FCLA Local-MXF Client Manual (2004)

MISSING IMAGE

Material Information

Title:
FCLA Local-MXF Client Manual (2004)
Physical Description:
Mixed Material
Creator:
Florida Center for Library Automation (FCLA)
Publisher:
Florida Center for Library Automation (FCLA)
Place of Publication:
Gainesville, FL
Publication Date:

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution.
System ID:
AA00001411:00001


This item is only available as the following downloads:


Full Text












MXF CLIENT MANUAL


Florida Center for Library Automation

rev. 11/04/04













MXF CLIENT MANUAL............................................................................................................................ 1

FLORIDA CENTER FOR LIBRARY AUTOMATION ....................................................................... 1

REV. 11/04/04 ............................................................................................................................................... 1

I. UNDERSTANDING THE MXF AND MXF CLIENT...................................................................... 5
I.a. What the MXF is and how it is used .......................................................................................... 5
I.b Sending data to F C L A ............... .................................................................................................... 6
I.e. The MXF Structure: Package, Entity, Division and File levels..................................................... 7
II. PREPARING DOCUMENTS ......................................................................................................... 9
II.a. Document structure ..................................................................... ...... ................. 9
II.b B o o k s ........ .. .................. ............ .................................................................................................... 9
II.c. Unitary visual materials (maps, posters, postcards, photographs).............................................. 10
map, poster, etc. [type of object, main....................................................................................... 10
II.d O their form ats ................................................................................................................... ........... 10
III. BASICS OF THE MXF CLIENT ................................................................................................ 11

III. BASICS OF THE MXF CLIENT ................................................................................................ 11
III.a. Logging on .............................................................................. ................. 11
III.b. Opening the Metadata Module ................................................................................................ 11
III.c. A areas of the screen ............................................................. ....................................................... 11
III.d T he T ree W window ...................... ............................................................................................... 12
III.d.1. Using the Tree: Nodes, Parents and Children..................................................................... 12
III.d.2. Using the Tree: Functions and Navigation.......................................................................... 13
III.e. T he W orkform W window ......................................................... ................................................... 14
III.e.]. M etadata W orkform s ............................................................................................................... 14
III.e.2. R ep eatable elem ents ................................................................................................................ 15
IV. SETTING DEFAULTS....................................................................................................................... 17
IV.a. When to use defaults................................................................. ...... ................. 17
IV .b. H ow to set defaults ............................. ........................................................ ................. 17
IV.c. Clearing, saving and changing defaults .................................................................................... 19
V. CREATING A NEW MXF PACKAGE .......................................................................................... 20
V .a. The Package control num ber...................................................... ................................................ 20
V.b. Setting the New Package directory..................................................................... ................. 20
V .c. Setting the Im age directory ...................................................... ................................................ 2 1
V.d. Package level Workform elements ......................................................................................... 23
VI. DESCRIBING PUBLICATIONS: ENTITIES AND DIVISIONS............................................... 23
VI.a. Adding Entity nodes ................................................................ ...... ................. 23
V I.b A adding D division nodes.......................................................... ................................................... 25
V I.c. U sing Insert R ange .............................................................................. ..................................... 27
VII. DESCRIBING PHYSICAL FILES ............................................................................................ 30
V II.a. A adding F ile nodes .............................................................................. ..................................... 30
V I.b. C creating D derived File nodes..................................................... ................................................ 33
VI.c. Copying Files.............................. ........................... 34
V I d V iew ing files ........................................................ ..................................................... ............ 35










V I.e. File w orkform elem ents.......................................................... ................................................... 36
VII. TREE MANIPULATION ................................................................................................................. 37
V II.a In sertin g n od es ................................................................................. ....................................... 37
V II.b R arranging nodes ............................................................................. ..................................... 37
VIII. VALIDATING PACKAGES .................................................... ............................................... 38
IX. OUTPUT TO XML ............................................................................................................................. 40

X: SPECIAL FORMATS: SERIALS .................................................................................................... 41
X .a. D directory setup for serials ......................................................... ................................................. 4 1
X.b. Package level information .......................................................................... ............ 42
X .c. E ntity level inform action .......................................................... ................................................... 42
X .d. D division level inform action ........................................................ .................................................. 43
X .d 1. Volum es..................................... ............................................................. .............. 43
X .d.2 Issu es........................................................................................................................ ........... 4 4
X.d.3. Contents of issues........................................................... 46
X.e. Serial title changes ..................................................................... ...... ................. 48
XI. SPECIAL FORMATS: MAPS AND PHOTOS ................................................................................. 49
XI.a. Directory structure..................................................................... ...... ................. 49
XI.b. Package level information.................................................................................................................... 49
X I.c. E ntity level inform action ......................................................... ................................................... 50
X I.d. D division level inform ation........................................................ ................................................. 50
XI.e. File level information ................................................................ ...... ................. 50
XI.f. Related Images........................................................... 52
M etadata for Im age F iles............................................................... ............................................... 52
M A R C record creation .............................................................................. .................................... 52
New Elements for Related Images ......................................................................... ................. 52
Image File Naming.............................. ........................... 55
APPENDIX A: WORKFORM ELEMENTS ........................................................................................ 56
PACKAGE LEVEL .............................. ........................... 56
E N T IT Y L E V E L ................. ............................................................... ........................................... 57
D IV IS IO N L E V E L ................................................................................................................................... 6 0
F IL E L E V E L .......................................................................................................................... ........... 6 1
D isp lay R ow ................................................................................................................ ............ .. 63
D isp lay C olum n ..................................................................................................................... ............ 63
APPENDIX B: HOW THE MXF IS ENCODED ................................................................................. 67

APPENDIX C: VALIDATION MESSAGES ........................................................................................ 68
A appendix C E rror m messages ......................................................................... ................................. 68
A appendix C .2. W warning m essages...................................................... ................................................ 68
APPENDIX D: OPTIMIZING WORKFLOW ..................................................................................... 70
a. Book, images exist.......................................................................... ...... ................. 70
APPENDIX E: TRAINING EXERCISES .......................................................................................... 71
E exercise 1: B ook w orksheet ............................................................................ ................................... 7 1
E exercise 2 : N ew package.............. ...................................................................................................... 74
Exercise 3: Adding Entity and Division level information................................................................ 77
Exercise 4: Adding File level information and finishing the Package................................................ 79
APPENDIX F: MXF QUICK GUIDES FOR TEXTUAL MATERIALS ......................... 80




3










M XF: Quick Guide for M onograph..................................................... ............................................... 80
M XF: Quick Guide for M ultipart...................................................... ................................................. 85
M XF: Quick Guide for Serial.......................................................... ................................................... 90












I. Understanding the MXF and MXF Client


l.a. What the MXF is and how it is used

The MXF (Metadata eXchange Format) is a locally-developed format for communicating
information about digital files representing library materials such as books, journals,
maps, and photographs. When we speak of "an MXF" we really mean a file of metadata
recorded in MXF format. Some of the information that can be recorded includes:

bibliographic metadata -- information describing the book, journal, etc., such as its title
and author

structural metadata -- information about how different images relate to each other in a
complex document, e.g., how pages are ordered to form chapters

technical metadata -- information about the digital file itself, such as its filename, and
when and how it was created.

The MXF can be used by a library to send metadata to FCLA, by FCLA to send metadata
to a library, or by two libraries to exchange metadata with each other. When used to send
metadata to FCLA, the MXF can include instructions to FCLA on how to process the
metadata and related files.










I.b. Sending data to FCLA


The MXF should be sent to FCLA along with the files it describes. For example, if a
library is sending PDF images of a book to FCLA for loading into Florida Heritage, an
MXF and the PDF files it describes should be sent to FCLA together. Generally this
means that the MXF itself and the files it describes should be placed in the same directory
and FTPed as a single folder.

However, note that files should only be sent to FCLA if there is a reason to do so. For
example, imagine that a library created TIFF masters, MrSid derivatives, and JPEG
thumbnails for maps. The MrSid and JPEG images were intended to be loaded in the
Visual Materials Collections server, and displayed to users. The TIFF masters were
intended to be burned onto CD by the library itself, and stored locally as archival copies.
In this case, even though for each map there are three image formats, only the MrSid and
JPEG formats would have to be sent to FCLA and described in an MXF. On the other
hand, if there were an agreement between the library and FCLA that FCLA would
archive the TIFF or burn the TIFF onto a CD and send that back to the library, then the
TIFF image would need to be sent to FCLA and described in the MXF.










I.c. The MXF Structure: Package, Entity, Division and File levels

The MXF is structured in four logical parts called "levels" : Package, Entity, Division
and File.

When an MXF and the digital files it describes are sent to FCLA (or from FCLA to
another library) the MXF itself and all the files transmitted together are called a package.
A package usually contains information about a single bibliographic entity but if
necessary multiple entities can be bundled up in a single package. For example,
generally a package will contain a single book, but there is no reason (apart from the
large size) that two books could not be sent together in a package. It might make sense to
bundle several short items together, such as the metadata and image files for several
maps.

The Package level contains information needed to process a single package. The type of
information contained at the Package level includes:

whether the package is new or a replacement for one sent earlier;
who is sending the package and should be contacted if there are problems;
specific processing instructions for the package, for example, whether to
create derivative PDF files from images sent in the package, or whether to
create MARC records from bibliographic data in the MXF;
what file formats are included in the package;
when the package was created.

The Entity level describes a single bibliographic entity, e.g. a single book, map, or
photograph. For a serial, the bibliographic entity is the serial title, even if only one issue
of the serial is being sent. The Entity level generally corresponds to what would be
cataloged in the library catalog, even if a catalog record does not exist for the particular
entity. The type of information indicated at the Entity level includes:

the bibliographic description of the entity;
what type of item it is (e.g. book, serial, map, photo...);
what projects) or collections) it belongs to (e.g. Florida Heritage, Literature
for Children, etc.);
a copyright statement;
a statement of who created the digitized version.

The Division level describes a logical structural division of the entity. For example, a
book may consist of a front cover, front matter, several chapters and a back cover. Each
of these would be described by a set of data elements at the Division level. Because the
structure of a publication is often hierarchical, Division level information can be nested.
For example, if the entity is a serial title, there may be Division level information for
articles nested within Division level information for issues nested within Division level
information for volumes. Information that can be given at the Division level includes:










what the structural subdivision is (e.g. a chapter, issue, article...)
what it is called (e.g. "Chapter IV")
a bibliographic description of the subdivision

The File level pertains to digital files. For every file included with a package, there must
be a description at the File level. The File level can include extensive information about
the physical file, including but not limited to:

the filename and file type
when it was created and by what institution and individual;
how it was created;
what device (e.g. digital camera or scanner) created the file, and what device
settings were used;
for images, what the image characteristics are (e.g. bit depth, resolution, color
space, descreening, etc.)

If a file was derived from a source file, then information about the source can also be
given.










II. Preparing documents


II.a. Document structure

Before you begin creating MXF data, you must understand the structure of the document
you are dealing with, and how it relates to the structure of the MXF. You will need to
know what the bibliographic Entity is and what Divisions exist within the Entity. For
each Division, you need to know what to call it, and what type of part it is.

If the document is a book, serial, or other textual work that will be displayed by the
FCLA document server, the name of the Division is what will appear in the online table
of contents that is displayed to the user. For example, if the Division is a chapter, the
name of the Division might be something like "Chapter I" or "Chapter 1: Tallahassee" or
"Tallahassee". The name should generally be what appears in the printed table of
contents of the work, although some Divisions like covers and other front matter have no
equivalent in the printed TOC.

The type of part must be taken from a list. It is used primarily for statistical purposes. It
includes such structural parts as: chapter, page, section, abstract, advertisement, volume,
issue, article, part, main, section, supply, cover, fonds, recordgrp, series, subgrp, subseries,
box, file, item.

When creating the MXF, you should either have a work in hand, or else have a worksheet
that lists all structural divisions of the work. The worksheet should list, for each
Division, the name, type of part, and range of pages included in that Division.


ll.b. Books


For a book, the Entity level will describe the book itself.
how to treat typical book divisions:


Division name


Here are some suggestions on


Type of Part


permission*
front cover

inside of front cover
other preliminary matter,
e.g. blank page
table of contents

foreword
preface


Copyright permission
Front cover or
Front matter
Front matter
Front matter

Table of Contents or
Contents
Foreword
Preface


division


permission
cover

cover
section

contents

section
section









chapters
parts
index
back matter
inside of back cover
back cover


Chapter X
Part X
Index
Back matter
Back matter
Back cover or
Back matter


* when a book is under copyright and digitized with permission, a statement describing
the copyright holder and the permissions granted should be included as the first digital
file for the entity.


ll.c. Unitary visual materials (maps, posters, postcards, photographs)

Most visual materials are unitary, that is, they have only one part, the map, poster,
photograph etc. itself. In this case the entity level describes the visual material, and there
is only one division.


Division name


Type of Part


map, poster, etc.


[type of object,
e.g. map, poster, etc.]


ll.d. Other formats

For how to treat divisions of other formats, see the chapters of this manual on "Special
Formats" (e.g., Special Formats: Serials).


chapter
chapter
index
section
cover
cover


division


main













III. Basics of the MXF Client




Ill.a. Logging on


You need a username and password to log onto the MXF Client. If you don't know your
name and password, contact your site administrator.



Ill.b. Opening the Metadata Module


When you invoke the MXF Client, you will initially see a blank screen with a menu bar
at the top. One of the options listed on the menu bar will be Module. Under Module
there are three sub-options: Administration, Metadata and Scanning. Select Metadata.



Ill.c. Areas of the screen


The MXF Client screen is divided into 3 areas. [Areas].

* rMXF Client--I[MX 11111Plxl


P I I "i ,- n_,h, I


SI ilhl i | I1?' ii I? I h i | II
h'i I r .i R hi R -: ii .1 i. R ,1



I I 1 _1 i '~ I i i : i
I4P-4

I Ii I l-- i I
I. I4"wi h --JI1" ,
-4, P-4, _. J.i ,


Type ol part
Seclion name
Namespace relerence(s I
Title
Alternative title
Creator


sarl| I A V F 0 'I41 ,1'-,,-,- I.' .r... I | i-, Mx E.-.. I


1 r i 'i-



'^- i'JI ''i ". _

.*:I |n .n I-

ll.: n -
nu ,'-,I .-,
a* -l',.-n,
nnn,* ," ,,'
nnn -n,.-
n, In_.- ,
,* "i-,1-h N


- ..1 a...











The Tree window is on the upper left. This displays all Package, Entity, Division and
File level information you have created as nodes in a hierarchical display called a "tree".
Whichever node is highlighted in the tree is the "current" node. In this example,
"Copyright permission" is the current node.


The Directory window is on the upper right. The top of this window displays the current
directory, and the lower portion lists any image files (.tif, .jpg, .sid) found in that
directory.


The Workform window is at the bottom of the screen. This window has a data entry area
for the metadata elements associated with the current node, for example, "Type of part".



III.d. The Tree Window


Ill.d.1. Using the Tree: Nodes, Parents and Children


Each of the four levels of the MXF are represented by nodes on the tree. [Tree]

Vi MF Clent [MX] pq-i F


0I,.. I I "1- I I 1.1 I


j,-, ~ ~~ R .i,4 ,-, .



- I ,.1 I .... R c I I
i iii i
I : .. 1. i ii "i' I? 1,1


Il-I



I.fl~.hfl III


BS2lI a -*E5 K V P, CA It


|| !1i~r.Iy* | F ~l .:iip i | .L-i l H MXF CI CEud.':rL | I Sn',.- 3 I|I '


There are four different kinds of nodes, each represented by different icons.


f .


* I II.-
* i 'i, -,1




jBBI
;I, I


Type of part
Page name
File name
File o, iatl
File order
Filesiz!









* Envelope icon = Package node. This is always the top-most, left-most node. In this
example, its name is "MXF".

* Book icon = Entity node. In this example, "FA00000033" is the Entity node.

* Folder icon = Division node. In this example, "Copyright permission", "Front
Matter", etc. are Division nodes.

* Document icon = File node. In this example, "Page 1" is a File node. (Derived files
have a slightly different icon. See "Creating Derived File nodes" below.)

The Workform window at the bottom always displays a metadata workform for the
current node. The current node is the node that is highlighted on the tree.

There are Parent and Child nodes on the tree. A Parent is hierarchically higher. On the
tree it appears to the left and above its Child nodes. A Child node appears below and to
the right of its parent. In this example, the Division node "Chapter 1" is the parent of one
Child node named "Page 1".


Ill.d.2. Using the Tree: Functions and Navigation


When a node has Child nodes, there is a plus or minus before the node on the tree. A
plus means that Child nodes are not shown. If you click the plus sign the node will be
expanded to show Child nodes. A minus means that all Child nodes are displayed. If
you click the minus, the Child nodes will be "rolled up" (not displayed) under the Parent
node.

The arrow keys move you up and down the tree:

* up arrow = Takes you to the node immediately above the current node
* down arrow = Takes you to the node immediately below the current node
* left arrow = Takes you up to the parent of the current node. (The current node in our
example is the node named Foreword; the left arrow would make the node named
FA00000033 current.)
* right arrow = Expands any plus nodes to show the Child nodes beneath. (Also, in
Windows 2000 systems, the right arrow will move you down the tree.)

The menu bar Tree option has many commands for manipulating nodes on the tree.
Among other things, you can add, remove, copy, and rename nodes. The most common
actions have shortcut key combinations consisting of holding the CTRL key while
pressing one other key. For example, CTRL-a has the same function as Tree/Add node.









III.e. The Workform Window


Ill.e.1. Metadata Workforms

Each type of node has a different Workform, listing the data elements that pertain to that
kind of node. The metadata Workform for the current node will appear in the workform
window. In the example above (Section 3.d, "Tree"), the File node "Page 1" is the
current node, so the workform for a File node displays.

Every workform has three areas. On the far left a label for the data element displays, e.g.
"Type of part", "Page name". The label color tells you if a value for the element is
required:

* black = optional
* red = required
* purple = conditionally required
* grey = supplied by the system

Purple labels always come in pairs; if one purple element is supplied, the other is
required. For example, for an Entity node, "Bib record key" and "Bib record system"
are purple. Both of these values can be left blank; however, if one is filled in, then the
other must be also.

A grey label means the value of the element will be supplied by the system if possible. If
the value is also grey, that means that you cannot change that value. For example, for a
File node, the element "Checksum" (label and value both grey) is calculated by the
system and cannot be changed. The element "File format" (label is grey, value is black)
is calculated by the system but the value can be overtyped by an operator.

In the center of the workform there is a box for data entry. Values can be supplied in four
ways:

1) by typing the value in the box
2) by selecting a value from a pulldown menu (down arrow on the right side of the box)
3) by setting a default (See "Setting Defaults" below)
4) by being automatically calculated by the Client program.

Values which are automatically calculated by the Client and cannot be changed are
shown in grey.

On the right there are buttons for performing certain functions.

* browse = when the value of an element is a filename or directory name, you can
browse through the directories known to your computer to find the name
* add new = if the element is repeatable, you can click this to add another instance of
the element











* remove = you can click this to remove the element


At the very far right is a scroll bar for moving through a long workform.



Ill.e.2. Repeatable elements


If an element is repeatable, an "Add new" button will display to the right of the data entry
box. Element values can be repeated in two ways:


1) use "Add new" to add another data entry box for the element [Add new]

1 .- F Cin r -.r =,lxl


.If JJdi


I ,,, : ,,11, -

"--l- Ii ,: ill1 ^ ^^ ^^ ^

--.--'--- ? ---- ---- --

y ,,, .... ff it, --,,i


1 I |..... I I '!'l IP I ... |I| ? ',l

?1 .I I .. | P .. 1. hil l
hi II| | I .. ^: .. .. I ^ I .. 1
iii IFA .... FA. l~iI ..


FAI .- iI l Ii


Entity type
Source institution
Project codes)
Project codes)
Version statement
Copyright


I,,, ,, , ,


I''
I '''I I,.-,i-'',.,.-,.,,,~,.. iii


.IBnlgI a SC K F- Rr C V


li r ,: ..:..... I| ..:.r.. ;l '..:.:...... || M yF CI I--:,, .. .Br,'-,. |,|,i' ll ^ 9 0 'J


2) type in multiple values, separated by commas [Commas]


* fr I 111
.r4,1


-II


t-, FIe FOL a Tr~c m ie i-/ridr:. Mud.1--E


Li












I XF Cient- [MF] 1111nF3


g ri.
.4 ilIIIIII


E ntily type
Source institution
Project codes I
Version statement
Copyright
Bib record key


I -i il I I I




.-. 4


isiarli I V R R V In .2T .i..Inn I .. ii ..,r.Fri I ,, 1i r.n I| ,MXF C 5i, -.. L I :.r.3.I, |I |(--n 09. '.




In the examples above, multiple values are entered for the data element "Project code(s)"
in two different ways. These are equivalent to each other; neither method is better than
the other. Use the method you prefer.


I 1...

















III-
_3 ,-,, : .
i ,, ,:-,-,n ,,.,
'- i.i. I: .: .: . ______._____' .._______, .,_



... 1 .... ..- .. .1 f .....
,. I."' .. ".I. f? ,,, I !f? ^










IV. Setting defaults


IV.a. When to use defaults

In general, before you create a new package, you will want to set defaults, or make sure
that the defaults that are set are correct. Each user has his own set of defaults that persist
until s/he changes them. Even if several people share a single workstation, each of them
has his/her own defaults.

The values you supply when you set defaults will appear automatically on the metadata
Workforms you use to enter data for a package. You can always type over a default
value with some other value.

Generally you want to default values which will not change from package to package,
such as the contributing institution. You may also want to default the values that most
frequently occur. For example, a book may have three types of divisions: front matter,
chapters and back matter. However, if most of the divisions will be chapters, you may
want to set the default type of division at the Division level to "chapter". Although this
will have to be overridden some of the time, it will be correct most of the time.


IV.b. How to set defaults

To set defaults, select File Defaults. A tree will appear in the Tree window, showing
Package, Entity, Division and File nodes. Select (highlight) a node from the tree to set
defaults for data elements at that level.











I MF Clint MXF]FlnF


E- Fl e .I T- ..-fnd.s-.- P0tduli


il F i-, .:,
a Ij-1 -.1,
5;


New Package?
Contributing institution
Make records
Make derivatives
Target server
Formats included


*asian|i f- K V UPCrA a


.-151 xl


I'-f ,,, I :. ::- -,- L. I I -1 -_j.1

J3.',-,
3.i....i i


1 II I
,,,,,., .... --,.., w ...

,., ,, ,,- ... -,


I --



I''"' "_n J,, ,- .



I| .ri-i-.. nn |I i1 'i? *..... ri-, I | MXF.CI i.. .I.r; r II i: I I|n r Q '


In the example above, "Package" was selected on the tree, and defaultable elements at
the Package level are shown in the Workform window. In this case the operator has set
"New Package?" to default to "yes" and "Contributing institution" to "FAU", which
would make sense for an FAU operator who mostly creates new packages. Defaults for
some processing instructions ("make derivatives", "target server" and "formats
included") are also set. Remember to use the scroll bar at the right side of the Workform
window to see all defaultable elements.


In the next example, "Entity" was selected on the tree, and defaultable elements at the
Entity level are shown in the Workform window.










I XF Cient- [MF] 1111nF3


n* : ,


E ntily type
Source institution
Project codes I
Version statement
Copyright
Bib record key


I 1-M I
I i1- n,n, n n .. n- I I. n ,r n


as2a| I S f-, K r rR 21 aV


I ....i..-* I .. I i'i... i, I ..n .. I -MXFCI L | 1 ; :.r.,.i", I | | Q .S


Most operators will probably want to default their "Source institution" and the "Version
statement" (which will only change once a year). If most work is being done for a
particular type of material and project, then defaulting "Entity type" and "Project
code(s)" also makes sense. It would not make sense to default elements such as "Bib
record key" (the key of the NOTIS record cataloging the entity) as these would change
from one entity to another. It is, however, a good practice to default "Bib record system."
Since this is conditionally required with "Bib record key," it will help the operator
remember to include the "Bib record key."



IV.c. Clearing, saving and changing defaults


To clear a default, simply blank out its value on the Workform.


To save defaults, select File Close defaults. They will be automatically saved.


To change a default, just go into default mode by selecting File Defaults, and change the
value on the workform.


I' ,. I :. : '".. L Ih -I














.1 4 II R
I .. .

_3 ,-,, : .
i ,, ,:-,-,n ,,.,
,,'-i.i.-I:..:.: .... ______._____'.._______,.,_


... 1 .... .,- ,,. .1 f .....
,, I."' ,. ".,I, f? ,,, I !f? ^










V. Creating a new MXF package


V.a. The Package control number

Every package must have a unique control number, called the "package control number"
or sometimes, the "package id". For packages to be contributed to FCLA, this number
must begin with a 2-character institution or project code assigned to you by FCLA. The
standard format is the 2-character code followed by a unique 8-character number, e.g.
"UF00001234". However, other formats may be accepted for certain projects. Before
you create a new package, you must decide what the package control number will be.

The package control number is meant to identify a transmission (that is, a bunch of stuff
FTPed together at some point in time). The package itself only exists for the length of
time it takes to be created, transmitted, and processed. After a package control number is
used once, it should not be used again, unless the entire package is re-transmitted for
some reason (for example, if there was an error in the original package that prevented
processing).


V.b. Setting the New Package directory

Every user should have his or her own directory for doing work with the MXF client. In
our example, the user is Priscilla and the folder (directory) for MXF work is called
"Digitized books". Many MXF packages can be stored within the folder "Digitized
books" at the same time.

Ordinarily, for a monograph or unitary work, each MXF package and the images it
describes should be stored in the same folder to make it easier to FTP the entire package
at one time. (To set up directories for a serial, see section X Serials Issues, below.) Best
practice is to give the folder the same name as the package control number. So, for
example, if the package id is going to be FA00000033, then both the MXF file and the
image files should be created in the folder named FA00000033.

In the example here, the directory structure is:

/Priscilla
/Digitized books
/FA00000033

To create a new MXF, select Package\'NeL,. A popup window will appear asking you
where to put the new package [Package new].











I* MF Clent Metaata]IV]F3


"I .. ,,,,- ,, ., i
-:IF
ill- .,'', I
"T I .I- I .
__ T,, ________
I. TI',I,|,F,lII t, I, I






.. .- T ., ..... I.... -_. .




Fole sI-, iF -il1 t |


.JIJAi
I' LI' I -I


*asian|i f- K V ri a1 R


I it I.r.. I| .. co i jE i i-, .nf.. -Ii (|' ZQ|| @9r 'J


Select the appropriate directory in the "Save in" box at the top of the popup window. Use
the down arrow at the end of the box to navigate through the directories on your
computer. To the right of the box, there are a set of buttons. The first button moves up
one level, the second button creates a new folder (directory). If the appropriate directory
(in this case FA00000133) does not already exist, use the "new folder" button to create it.


In this example, a new package named "FA00000133" is being created in the directory
FA00000133.



V.c. Setting the Image directory


When the new package has been created, a Package level node will appear at the top of
the Tree window [New Package].


E- F-IE Fe c 3 _- ,_ _,t e r'. n ..',, M.dul,:











16i -X rClie u [ IIO,,00 '2 F3


Q. File Fackage _iEee "i' '"nrdc.. CMdule


[J I:'. ..ge?:'
I- onlrnulir.ng i riluriron
ICoolllt.ulln ll'iln l ildual


r.la ke reCordC
P.JL ri'1 -Ti i-,.


_JjJA J


Si LI. -I

lI=,,: :,11.
-J i,.j,,ll:- ae,... .i


Ille pdlIi i ir


IWer I: ,11

1 :1
r 11
l .r l -II
.rlc: .Jr

_1,1. ," Ir
Ilr- I:


I-I
r 1 ... F ,,..,.II 1.*,, ,h :,--36 .,.,I I.Il- I.ii.lUi 1.1. ._
i .... :.


.iSlailli I fa 1 Pr [l lA VW IM eO:W: | jtcm ||3MXF 9EdA.|3 I S \napi



A metadata Workform window for Package level information appears at the bottom of
the screen. Note that defaults appear for some elements.


The Directory window at the upper right of the screen should show the directory that
contains the images to be included in the package (assuming these images have already
been created). When you create a new package, the directory containing the MXF
package is automatically highlighted in the Directory window, and the name of that
directory appears as the value of the Image directory element in the workform. If the
image files to be included in the package are in a different directory from the MXF
package, you will have to change the directory listed. Any image files in the directory
will appear just below the directory listing.


NOTE: If, after creating a package, you move the package and associated images to a
different location, you need to re-open the package and change the Image directory.


If image files have not been created yet, it does not matter which directory is highlighted
in the Directory window.


B I---:e


I .














V.d. Package level Workform elements


Package level Workform elements tell FCLA how to process what you have sent. See
Appendix A for explanation of element meanings and use. See also instructions for
specific formats of materials, such as serials.


VI. Describing publications: Entities and Divisions




VI.a. Adding Entity nodes


Every package must have at least one Entity level node. To add the node, make sure the
Package node is highlighted on the tree, and select Tree Add node [Tree/Add node].


I-l Fle Pacti B, '. ., V ,%:,,,. t,_o dul, ._j_ x)


_' II --


Ji.ii~
iL .~.' .dL.. I


l Ii' ii I I II




IF
> i l' i ..1 j H. h.i P :~ i.. II .1 J "ii



1 ,m 'I Ii i HI i i i P -11,1
,, : 111 I ., -.., 1,_ I. I
'1 I


New Package?
Contributing institution
Image directory
Make records
Make derivatives
Target server


I'll
I,"


A popup F Iwindow will prompt you Ito add the node name [Add entity]... .,




A popup window will prompt you to add the node name [Add entity].


I- .- 1 1-

i I i
I',n I F 5Iol_
E-.roand -II
5I11oP E-II


....1-ll .1 l.. h











I*MXFC et .... -IV] FI


, Fl IE Fbc6 3,- I T, '_,. _,n':.d ," _1mdul,


lrii I-ud r I.| tur. ,d ri0r,,-

p:7-1 FT--


-IJI Xl


'- 'I I I : : : -" L' -I |---.1L-I I

J3.,'
_Jl~i:iI ,:,n
I3'-,, ~ h",-,J- ,,:..:.:l ._____________________


ILi Add node. '


I :"51w l I


IF i i,.i
I i hi
Ic : : ,.,,
li..II:"


. ill,
- "II'.


V11,,.1


1.- 4
~I II I



ii l I
I I Iii I
Rio:.
IRA


New Package?
Contributing institution
Image directory
Make records
Make derivatives
Target server


*asian|i f- K V UP CA R


I --

i" 1 .. :.... ..... r-_] V.aF,-,
I:" -] ,. ne... -__


|I ..riii..nn | "BI ii"**. .....*r ic'' Irn | -I |i E L | ii ri :.n I,:'H:|lZ rS 'J


The Client assumes that any node added to the Package level will be an Entity node
inserted as a Child to the Package Node, which is almost always the case.


The name of the node given in the popup is also called the "Entity ID". For a work
completely contained within one package (for example, a monograph all chapters of
which are being sent at the same time), the Entity ID should be the same as the package
control number. For a journal or other serial which may be sent over time in a number of
packages (for example, volumes 1 and 2 in one package, volumes 3 and 4 in another), the
Entity ID should be the characters "SN" followed by the 8-character ISSN of the serial
(e.g. "SN00213392"). This allows all parts of the work to be brought together in the
system, even if the parts are sent separately. In this example, since we are doing a
complete book, the node name is the same as the package ID, "FA00000033".


If you made a mistake typing the node name, go to the Tree window and right click on
the node icon, then select Rename.


An Entity level workform appears in the Workform window, where you can now enter
entity level information [Entity].


jr1i"F


,r,,',


FA, ,, ,
w I,: 1,,1











I MF Clint MXF]FlnF


E- F'IE bc6 ,... I 51 'T- '. ..ncd.s" MtduliE


ErI r
r, l 0


Source institution
Project codes I
Version statement
Copyright
Bib record key
Bib record system


-A.A xl


I' ,,,l 1':, :: '", L....-I |j


J .. ..






1 4 ., ,.
I,, ,:, q
,-.. 4 .. .. 4 .1 P ,,,
14 ,,.,, I" : ,, _I:, ,I,, : ,,,f


-_
-_


I-I. I.:- in .I i I l I4 :


, I.- i
11- .1.- 1 --T _- -:, 1. l -1


as2a| I S f-, K r rR 21 aV


II 3'U i-i |-i..... n.. I l ."l 'I. II | MXF CI i- I-.1"I L ; :.r..-i', I' I | S '


Note that there are two ways to supply a bibliographic description of the title:


1) If a NOTIS record exists, you can supply the NOTIS record key in "Bib record key"
and "FCLANOTIS:QF" in "Bib record system". This associates this package with the
external bibliographic record.


2) If no NOTIS record exists, you can supply a complete Dublin Core description by
filling in the elements "Title", "Creator", "Contributor" etc.


In both cases, you must supply the element "Title" at this level. This is used by FCLA
staff to make sure the correct volume is being processed.


See Appendix A for explanation of Entity level element meanings and their use.



Vl.b. Adding Division nodes


Only Division nodes display in the Table of Contents for an Entity. The rule of thumb is
therefore to create a Division node for anything that should display in the TOC.
Sometimes people wonder, if a section is only one page, if they need to create a Division











node for it or if they can simply add a File node in the appropriate place. Either method
is acceptable. However, if you want it to display in the TOC, you must create a Division.

To create a Division, make sure that the parent node of the Division is highlighted in the
Tree window and select Tree Add node or use CTRL-a to add a node. Note that if a
Division node is current, the Client assumes that the next node added will be a Division
at the same level. If you are adding a File node or a child Division, you will need to
change the default node type. [Add node div]


Lj Llle tb.^,: ,- i-? -R,3 ,. um' r:- LJ-,,..1 _jgl _|
FI I i I I 'iiI I I "


... i


FW iFg iqW ,,,i'C
rig,
,,, Add,: nud ...


-j


E.... one.. re na ..-.. ..n o, .I"m p ...






Copyright --
Bib record key I--- : I


Bib record systern
Namespace relei ence(s
Title
Alternative title


I. 1- TI" n -.


-.I -I.. -


BsrlJ I A r" V V H I | .. Ih | I Ie".. o I | 5., r.. :ch rI |I ,- E 1.3,:., L I S...i. Q: I I 3 "'J



For each Division, there are four data elements that you need to know how to use:


Node name
Type of part
[type of part] name
Title

You are prompted to enter the node name in the box that appears when you request "Add
node". This node name labels the node in the tree, and is also what will ordinarily appear
in the online table of contents listing for the work. For example, if the Division is a
chapter, the node name for the division might be something like "Chapter I" or "Chapter
1: Tallahassee" or "Tallahassee". The node name should generally be what appears in


_ : I I, ^ : ,.,
t..,-, LM T ..


,_J









the printed table of contents of the work (unless it is a kind of division that doesn't
ordinarily appear in a printed table of contents, or if the name is very long, as will be
explained below). In this example the node name is "Copyright permission".

Once the Division node is created, the workform for Division level information will
appear at the bottom of the screen. In general, the only data element that needs to be
supplied at the Division level is "Type of part", which is used primarily for statistical
purposes. The value for this element should be taken from the pull-down menu. Values
include such structural parts as: chapter, page, section, abstract, advertisement, volume,
issue, article, part, main, section, supply, cover, fonds, recordgrp, series, subgrp, subseries,
box, file, item.

In the metadata workform, the label "[type of part] name" will vary depending on what
you enter for Type ofpart. For example, if you say that the type of part is "chapter", then
the next element on the metadata workform will be labeled Chapter name. If the Type
of part is "section", then the next element will be labeled Section name. In general, you
do not have to provide a value for this element. However, if the type of part is "chapter"
or "page" or some other unit that is numbered (e.g. Chapter 1, Chapter 2, Chapter 3...),
the number or name of the part exactly as it appears on the piece can optionally be given
in "Part name". A value is required, however, if the Type ofpart is "volume" or "issue",
in which case the Volume name should be the number of the volume, and the Issue name
should be the number of the issue. See "Describing Serials" below.

The element Title can usually be left blank. Anything entered in Title will be used
instead of the Node name in the table of contents displayed to the user. This is useful
when a title is very long. Typing in a long title as the node name can make the tree look
messy and hard to read. Instead, type an abbreviated form of the title as the node name,
and type the complete title in Title.

Additional bibliographic description can optionally be provided. For example, if the
Division is an article in a journal, then the author of the article might be supplied in
"Creator".

See Appendix A for full explanation of Division level element meanings and their use.


VI.c. Using Insert Range

The Tree/Insert range function can be used whenever you have multiple Entity, Division
or File nodes whose names increment sequentially. (Insert range for File nodes is
described in "Describing physical files" below.) For example, if you have a book with
chapters numbered "Chapter 1" through "Chapter 10", you can use Insert range to add ten
Divisions for the ten chapters.











Highlight the parent of the nodes you want to add. (That is, if you are adding a range of
Entity nodes, highlight the Package node on the tree. If you are adding a range of
Division nodes, highlight the Entity node on the tree.)


Select Tree Insert range [Insert range].

,1:1r I IxI


-Ill xl


'-- 3 I '- l- i',l l "I








lIF r .


S a urce insthlaon
F roleCI COdCel I
Var ion Ifate i1ant
S0op:. righi
8i record key


Ene ange.Ie n 'o II-Rom n mhaI r1. t. Cancel

- elecI lrnge rype -"
S ,:.c.n ( Eniry h le l[,i HI ,,i,,, .

Type ao pan i. 1.-. -
IChapier name


Tole .7-1
Crealor o ie'I
in ,il r rol" l.l. u I
F "'^----------,--,,


A popup box will prompt you for information about the nodes to be added. In this
example, we have a book with 10 chapters titled "Chapter 1", "Chapter 2", etc., so we
filled in "Enter range name" with "Chapter" and filled in "Enter range values ..." to
indicate we want the parts to be numbered from 1 to 10. We selected "chapter" as the
"Type of Part". If other bibliographic data was constant for all chapters, the information
could be entered on this workform.


When the popup workform is filled out, click OK. The range of nodes will be added as
the last set of children under the highlighted parent node. [Range added]


II- I II I
4-


E- Ede -a ag -D e '., '?'n C- Mo...


J


J-j













I MF Clint MXF]flnF


Sri.F
'r 'I- nI
/ .i I r I


I-I,--
i5 I,*
a Char-r 1




: I' -I|' I

I I-l~lI,-


Type of part

Chapter name

Namespace referencelsi


Alternative title

Creator


.dlJ Ilt


sia | IS K R V ` | 2TI 1 i .r..7.MX', I |I,1| )'r 9'




The normal workform for the first added node will be displayed. You can now add

information unique to this chapter if any.


I' ,, 1 ':, :.. L h -I j





















I-- q
L .. ....







,-,,. -- ,-,,_pr--,, -..












VII. Describing physical files



VII.a. Adding File nodes


There are three ways to add a File node:


1. Drag method. If images already exist in the directory, then the best way to add file
nodes is to drag the images from the Directory window to the Tree window. After
creating the Division in which the image belongs, right click on the image in the
Directory window and drag it directly onto the Division node in the Tree window. [Drag]


.t Fle FP c .a T e- \_ n ,,.'- Moduli -j j _j


sr i i'.



I^-III~ll.- __
: -, ,. ? t| I


iii 7, i n-
*I~Ii ,.-
I '-l'-lI~
I i, I-I,-


I I I' 'I,, I

Ji.ii~
iL .~.'


X ,1 ,,1 '! ... | t l ,l T.. | .,... : ,
7A ,,- ,,R !, :,q M -: 1-- 11 j



* .1 g II I PA. hi T ,Ilr

i; il-"q I :' .1 ,,,iB H
4j I4 j I




II ..:or h | I .. | ........... | MXF CI lg E ...:.. I :,,,i :,,


A subordinate node for the file will appear, and File level metadata will appear in the
Workform window. The file name, file format, and file creation date and time will
automatically be supplied to the workform. [Drag done]


To drag multiple files at the same time, click on the first file, press and hold CTRL, and
click on the other file(s). Continue holding the left mouse key down after selecting the
last file, and drag the group.













, F'IE c6 3,, IT-, '_,. -.n,1.:,," Mtdul,,


Sr.

0
r I 1 n-:1111:1111 n


a T i 1:.7 1-
I: .-,- ,i?,,), :, l
T ii
I ,1

m '.Ii'- ,

,. I, 1,, '
mm ,I,-I'n-
,. '-, I"'- '
,.iI-"- n
nlm ,I,-I n l


Type of part
Page name
File name
File lorinal
File order
FI-lesizp


-1A 1 xl


j ,:.















.14 I- M
14.,r,,,



I ,: ,, ,
,-..i .... i i ,-.1 P_ ,,,i
,, I_ ,,, ,? I. ,, I' ,,, ? ^


-II


I, ..~-..II;I.1.i.1I.. I.I,,,II,,,,,, I




II I


asian|i I f-, K r r17 A


I 2~-I -T "*'** I FIii'kr.r iri-.rn || *MXFCI L |1.r;[ :.r.a~i', I| QrSQ'


2. Node method. If you are creating the metadata before you have the images, then you
can create a File node by highlighting the Division and selecting Tree/Add node. By
default, the Client will assume you are adding another Division; change the default node
type from "div" to "file".


3. Range method. If you are creating the metadata before you have the images, there
will be an image for each page, and the images will be predictably named with the page
number, the easiest way to create the File nodes is to use the Insert range function. [Insert
page range]


I MXFClient [MX]FillnF












E Fl IE .c6 l._- I ,T _,'. d -,n .:.-" MIlduln


r II.F













NamespaceI-. reerencel
S,.,I- I-

















Type of part
Chapter name
Namespace relerencel
Title
Alternative title
Creator


11i, iN..Z. r.Cprl i i[|


or.if.F.- l. l im I ':

r l, ,:,I r ,,,r.


.16J Xl


:Ja


I aanetrneI


:anc-l i


r F,|I :,lpr. FIImI,- r" [I


Iiii I-




~~IRA
,,,1 : ....


.,-,,, : ,,,


: m ni I m

,III


fl 1-


-ddJ it.
-Jdd it.
-.dd it.


iasia|i I fI K r r2 a V


II 2 on-* .- I .i-i k ri .! ii-.r.. C" F1MXF CI nI-I ''L :.r.,-i,, I: | | ,S -


Highlight the node for the chapter or other Division to which the pages belong. In this
example we are inserting page images for Chapter 1. Select Tree/Insert range and a
popup window will appear. For "Enter range name" we enter "Page" so the pages will be
named "Page 1", "Page 2", etc. There are six pages in chapter 1 so we enter the range
values from 1 to 6. We select range type "file" so these will be file nodes, and the "Type
of part" is page.


The program supplies a default directory for the element File name, and a default file
type. In this example the directory is m:\user\Priscilla\DigitizedBooks\FA00000033\ and
the file type is "image/tiff". If the box "Supply Filename" is checked, when the Range
function is enacted, the actual file names will be supplied automatically using the
directory name, the range values, and the file type. So, for example, the first File node
created will be given the filename m:\user\Priscilla\DigitizedBooks\FA00000033\l.tif.


When all the information has been entered, click OK. The range will be automatically
created. The Workform for the first of the added nodes will display, so you can add any
information necessary. [Page range added]


Iiiiii e .. ... 1119r] II-f 3


Type of part
Page name
File name
File lormarr
Creation method
SoFt'jarfE 'ied


V. 1-- 1


V, .n
I












I MXFClient [MX]FillnF


r- FlIE T6 ,,. I r= '_,'. -n,.nd .-" Mtdul,,


jrI.I

I I n 1 : : 1 I- n n





_j T, i-- :
Jn,-I
iI i: I -


I' ':7'I

II I-I== I
J '-l 1l I

II:1,.i "
I JI'~ Ia I
I I~I~-I1

I I II l -lh


-1 1 xl1


Type of part
Page name
File name
while lorinal
File order
I-ilesizF


-II


sIa.l| ,S R r VF I | I 'r ... I i .r .'.r'. I" -.. Ir.. n I IAMXF Cl nlii.nL l1:.r.a._in, : 'll r 'J





VI.b. Creating Derived File nodes


If you have a file derived from another file in the same package, for example, a JPEG
image derived from a TIFF image, you can create a Derived File node for the derived file.
A Derived node automatically records information about the source file that can be useful
in the archival management of the derived file. If you do not plan to manage the derived
file archivally, there is no need to create a Derived node.


To create a Derived node, highlight the node for the source file. In this example it is the
"copyright" node for the TIFF image. Select Tree/Derive node. A new node with the
same name as the source node will appear. [Derive]


-_=,- I-.. I I:i ::"- "' L-.. ..- I iJ



j .. ..
,r,.:, ,. ,

,,, ,i: -,, ,:,,n ,, ,
^ L. i. i -I. .:. : ._,, ,,-___,____.. ..___,__,,_..__,



,i .... ,, 11 i ,, i ,,, ,?




,4I1,,-Y.f IM R,: .. 4 I. f : ,I

I I- P











I MXFClient [MX]FillnF


, F'IE c6 3,, IT-, '_,'. -,n1,:.-" M, dul,,


jrI.l
0
I I -1:1: 1:1: :' :

"I, i n
,U1' r'rr

II. I-I' iirr~
,I_-I = :-,,,,
.: m .-l- I
I|:' 4
*m '. i_,-

*m, I-, i_-,'
,. i-, i:, ,hlU


Type of part
Page name
File name
File lorinal
File order
FI-lesizp


-1A 1 xl


I' ,,, 1':, : ", L....-I IJ
A,.-
_ji:,, ",:,n
^ L.i. -J I:..:.l.____________________


"i ..... W 1i w 1P










I I1 I


I, *.-.~-~*.-.II;I..*1.i.*.-1I.-.- I.I,,,II,,,,,, I
Ii


-II


*asian|i f- K V ri CA R


I2I-*'* I Lir 4r.,ri-'4!1iri-.rn. || FIMXFCI TI1. .r.a~i', I| QrSQ'


The program assumes the derived file has the same filename, but a different file type, as
the source file. Your cursor will be in the metadata workform for the new node, at the end
of the default filename, so you can add the new file type.



VI.c. Copying Files


Sometimes a file needs to appear in more than one place in the Tree. For example, a
new chapter may begin on the same page that the previous chapter ended on, so the file
for that page needs to appear under two chapter Divisions. In this case you have two
options:


a you can create a new File node for the page by either using the "Add node" function
or dragging the file again from the Directory window;
a you can use the "Copy node" and "Paste Node" functions to copy the first file node to
the second location.


To use copy and paste, highlight the node to be copied and select Tree/Copy Node
(CTRL-N). Then highlight the node under which you want to copy to appear, and select
Tree/Paste Node (CTRL-P). The new node will appear with the same name as the copied
node, followed by (copy). [Copy node]





















-. MFe Cl age 1e [U e. x*0 ; o.r _.c.1le ].i[]jE
~-. Flie Package ire ''* ifdc.. .dvle j1
Ir


i:



I 1 .1. .1..-1 Hf I *I. I.:l.





11 1 iL.- 1-1 II.jII.- r.I .


II I 'I


U. ii-,~i .
i:,, .
I1-.. .11







^ -II' ;' I 1' II 'll ~ IJ JJJJ ..



.i ,'.. 1^ **. ,l:.
1 10I gI ,I-:


T:. pe of pan
FaIe rnam6
File name


File oidei
F.I ,-


I,


aSlailJ 0 fI U MR r v. IM rir J:.: |Vr w..:. ||E-.l. | `.9.ji l MXF




When you copy a File node all the data entered on the File workform for the first node
automatically carries over to the copy, so you won't have to re-enter any information.
This can be an advantage over using the Add Node method. It is an advantage over
dragging the file node twice only if you supplied non-default information after dragging.





VI.d. Viewing files


Sometimes it can be helpful to view a file to see what page or section it represents. To
view an image file, highlight the file name in the Directory window, and use
View/Image. The image will appear in a popup window, which can be resized using the
mouse. You can also highlight the file name and use CNTL-I to view the image. The









popup window will also allow you to move from one image to another, using back and
forward arrows.



VI.e. File workform elements

The file workform has several required elements but most of them can be defaulted or
supplied by the system.

"Type of part" at the File level can generally be defaulted using File/Defaults. For
example, if you are sending a book of page images as TIFF files and their derivative
JPEGs, it would make sense to default "Type of part" at the File level to "page".

"File name" is automatically supplied when you use the drag method to create the node.

"File format" will be supplied by the program if it can get this from the extension on the
filename.

"File size" will be supplied by program when the image is present in the Image Directory
and the information is present in the image header.

See Appendix A for full explanation of File level element meanings and their use.










VII. Tree manipulation


VII.a. Inserting nodes

The Add Node function will, by default, add the new node as a child to whatever node is
highlighted as the current node on the tree. It is possible to insert nodes in other places
by changing the default position from "child" on the AddNode popup box.

*Current node
*default position of add node

"Previous" will add a node directly above and at the same level as the current node.

*Position of added node
*Current node

"Next" will add a node below and at the same level as the current node. The new node
will be placed after any children of the current node.

*Current node
*child of current node
*Position of added node

"Last" will add a node as the last node in the tree at the same level of the current node.

*Current node
*Node
*Node
*Position of added node


VII.b. Rearranging nodes

Nodes can be moved from one place in the tree to another by dragging and dropping the
node to be moved onto another node. The node will be added as the previous peer of the
node on which it is dropped.

For example, say that by mistake you have inserted nodes out of order, such that the tree
is ordered as follows:

*Parent
*Child 2
*Child 3
*Child 1











You can drag and drop the node Child 1 up to Child 2. The node will be inserted before
Child 2 at the same level and the nodes will now be in order:


*Parent
*Child 1
*Child 2
*Child 3



VIII. VALIDATING PACKAGES


When you are finished creating a package, you should check it using the
Package/Validate function. This will tell you how many errors and possible errors
(warnings) there are. [Validate]


.- Ele Eaci age lee .',e "_'.dc- I dlpdM -JLjJ


5



31 1' 3~ .. ,,.:,..i,
h


5 i .L I ,) .3-! 30
-3-

I-- I-;,.l .,, I' ,,,l I, ,,,-... .1 E r T l C:,, ..1


1 1
I-I i L I I n l i ",3 ,,,l ---



il
3ii

3i
I--I I LIIi,, I. _, 2.1 ,


IFJxJ Pac age?
I- O ilrd .u ii .-l iri rilunrlc
I-onIrb.ulin. inCI Idual
Irr.cqe director,
r lape recoLlJs
r Iqb6 rolt, ,qT.,,oc


sia,.jl i K V0 PR' 2 v


e e ae 11 enior: anr.d I .iin.ng: Ic- Ih.: pa.:- age

w, I

1-'-9 F


1' Iil: U


Pi-
I-l II: 'm
.: II: ,*I
1 II: ,u
"1Iii = i


II-_







CI E ... r :: II n II MXF S' D.al | |;,


When you clear the popup box by pressing Enter or clicking "OK", a table of errors will
display. [Error table].


'---? ,.,. r 1 : -*' '"" "'-' ; LJ,, el ^
I-i iI
...- T i ,,)h Jh,'
I F, -- ,11IIII]


A,, ", ,, )


I I eI I









L F Ili Fent IlF *000052II1.,i I x F
Q, P~e Fackage lise ,e-,, '_,,rd&e.,, Ddul:,,e ..l19lxl


.iJ J\s "\ i i ",- LI -I
0 Package EFFOFI/Wans MI-I


Ij rJ I i I.L'LJ e IIr9P I~k


Although the message in the "message" column may be truncated, you can display the
full message by either double-clicking the error line (the full message will appear in a
pop-up box) or by holding the cursor over the error message (the full message will
display as a mouse-over).

If the error concerns bad data in the metadata Workform, you can see where the error
occurred by double clicking the error line and clearing the pop-up error message. The
affected node will be highlighted in the Tree window, and the cursor will be placed in the
data entry box of the Workform that contains the bad data. In order that you can see this,
the table of errors is minimized. You can bring it back to the foreground in the normal
way, by clicking its icon in the task bar

You can initiate validation at any time during package creation. FCLA recommends that
you validate periodically, after adding 10-15 page files. This step performs the function
of calculating the checksum values for each image, which can take some time, depending
on the size of the image file and the power of the computer. By validating frequently,
final validation goes more quickly. Validation is automatically performed by the system
when XML output is requested. Errors must be fixed before the MXF package can be
output as XML. Warnings will not prevent the output of XML.

For a list of error and warning messages and their meanings, see APPENDIX C.


I


Affsila~i~lgfa I a fi;K AV












IX. OUTPUT TO XML


When you finish creating your package, you need to output it as XML before sending it
to FCLA or another library.

Select Package Output to XML. First the MXF Client will automatically Validate the
data (see Validating Packages above). If there are errors, they must be fixed, and Output
to XML requested again. If there are warnings, you can either correct them or simply
click OK.

The MXF Client will then create an XML file and display popup message box telling you
the name of the XML file created.

If on the View menu, XML Tree on Output is checked, the Client will also open a new
window showing the XML in tree format. Like the Package tree, the XML tree can be
manipulated to expand and contract nodes by clicking the plus and minus boxes in front
of the node. You can get rid of the tree by closing the window. XML trees cannot be
saved, but can be recreated by redoing Output to XML.

The XML output file is saved by default in the Package directory. To view the XML file,
select View/XML File. You can also view the XML from outside the Client using
Notepad or a similar application. Never edit an XML output file directly.










X: SPECIAL FORMATS: SERIALS


The MXF allows flexible handling of serials, which may have many levels of hierarchy
(e.g. volumes, issues, articles). However, for serials to be processed correctly, it is
important to understand and follow special rules and conventions.

If many issues of a serial are to be digitized, it is required that each issue has to be sent
to FCLA in a separate package. There is no requirement that issues be received in
order. The important thing to remember is that the node name of the Entity (Entity ID) is
what ties all information for a single serial run together. If a serial is sent in several
packages, the Package ID must be unique for each package, but the Entity ID must be the
same.

X.a. Directory setup for serials

When you set up directories for your serial package, one and only one subdirectory
should be created for each serial issue:


\UF80001549
\SN00130098
\1 1


[package ID]
[entity ID]
[directory for images for volume 1 number 1]


Please use only numeric characters for the name of the subdirectory for each serial
issue. If the name of the subdirectory contains non-numeric character, the package will be
rejected.

Examples of valid directory setup:


\UF80001549
\SN00130098
\1

\UF80001549
\SN00130098
\1998_004

\UF80001549
\SN00130098
\123-124


[package ID]
[entity ID]
[directory for images for volume 1 with only one issue]

[package ID]
[entity ID]
[directory for images for volume 1998 number 4]

[package ID]
[entity ID]
[directory for a combined issue number 123 and 124]


Examples of invalid directory setup:


\UF80001549
\SN00130098
\voll nol


[package ID]
[entity ID]
[directory for images for volume 1 issue number 1]










\UF80001549 [package ID]
\SN00130098 [entity ID]
\vol3 [directory for images for volume 3]

X.b. Package level information

Create the package in the highest level directory you are using for the serial. In the
example above, you would create the package in the directory UF8001549.

Package ID (Node name). When you create the package node, the Package ID (node
name) should be in the standard package ID format, consisting of the 2-character
institution or project code followed by an 8-character numeric value, e.g. "UF80013549".

Image directory. All images being sent in the package should be in a single directory
(that is, the subdirectory for each serial issue). Set the Image directory to that directory.
In the example above this would be "...\SN00130098\1_1".

Make records. For some serials you may want author/title bibliographic records created
for each article published. If this is the case, select analyticss, article" as the value of
"Make records".

Make derivatives. For some serials, you may want a PDF file created for every article in
the serial. For other serials, you may want to create a PDF for each issue instead. Select
"pdf/article" or "pdf/issue" if you want either of these options.

Please note that you have to use sensible "Type of Part" for division nodes that you want
to create or wrap PDF files. If you select "pdf/article", the following "Type of Part"
nodes will be expected to appear: article, advertisement, contents, correspondence, cover,
editorial, index, section, permission. If you select "pdf/issue", nodes with the following
"Type of Part" will be expected to appear: issue, permission. In other words, even though
you select "pdf/article" or "pdf/issue" for "Make derivatives", but if you don't use any of
these "Types of Part" for any nodes, no PDF file will be created.


X.c. Entity level information

Entity ID (Node name). The node name of a serial entity must be the characters "SN"
followed by the 8-character ISSN of the serial. For example, if the ISSN is 0123-4567,
then the Entity ID (node name) for a UF serial must be "SN01234567". That way, even
if volumes or issues are sent in multiple packages, there is a unifying identifier that ties
all the pieces together as parts of the same serial.

Some older serial titles do not have ISSNs or some materials you want to package as
serial are not serials In this case there are two options: apply for a retrospective ISSN, or









assign a pseudo-ISSN. Instructions on how to apply for a retrospective ISSN, or assign a
pseudo-ISSN are available at http://palmm.fcla.edu/strucmeta/pseudoissn.html

Entity type. Must be "serial".


X.d. Division level information

Most serials will have at least three levels of hierarchy at the Division level: volume,
issue (number), and the contents of the issue (articles, correspondence, editorials, etc.).
This is represented on the Tree as parent and child nodes: the Division for the volume is
the parent of the Division for the issue, which is the parent for the Divisions for the
articles within the issue.


X.d.1. Volumes.

Node name. The node name for the volume should be what you want to display in the
online table of contents. In other words, if you would like the table of contents to display
as:

Volume 1
Volume 2
Volume 3

then you must record the node names for the volumes as "Volume 1", "Volume 2",
"Volume 3" etc.

If you want the chronology of the volume to display as part of the table of contents, then
you must include that in the node name as well:

Vol. 1 (1898/1899)
Vol. 2 (1899/1900)

It is important that you record all node names for volumes and issues consistently
throughout the entire serial run. Recommended practice is to use either "Volume x
(year)" or "Vol. x (year)" as this gives the most information to the reader.

Type of part. The Type ofpart for a volume must be "volume".

Volume name. This is required for a serial. It must be the numeric designation of the
volume. E.g. if this is volume 43, the Volume name must be "43".

Date. This is required for a serial. The date in this field will be used to make a
chronology tracing for the date of publication of the original materials. Use the date of
publication for the volume in hand. For volumes that don't actually have real publication











dates printed in, you need to infer the date or determine a probable date of publication
and enter this in brackets with a question mark, e.g. [1923?]. The date will be displayed
as "Publication date" for the item in Textual Collections as well as for sorting the search
results according to "date ascending", "date descending". The date in this field will also
be used for statistical purposes.

IA MXF C len -[ '800 1 x 11 l F9


a column 4 Ill1i)t


.JiJjl


I4 I i' 1.


Type of part
Volume name
Title
Date
Creator
Rights


A.sinai C K Fnri rE V


Il I,,


H I~ .I.~~rJ~j~r P*I~ r


X.d.2. Issues


Issue nodes are created as children of the parent volume node. Please note that for
annuals or serials that only have volumes, issues nodes are not required.

Node name. As with volumes, the node name for the issue should be what you want to
display in the online table of contents. In other words, if you would like the table of
contents to display as:

Number 1
Number 2
Number 3


t-, Fle FcL ... Trmee ,i,= i.,nj:.,. r,,Mud.ul-









then you must record the node names for the issues as "Number 1", "Number 2",
"Number 3" etc.

If you want the chronology of the issue to display as part of the table of contents, then
you must include that in the node name as well:

No. 1 (Spring)
No. 2 (Summer)

It is important that you record all node names for volumes and issues consistently
throughout the entire serial run. Recommended best practice is to use either "Number x"
or "No. x" and to put the chronology in parenthesis.

Type of part. The Type ofpart for an issue must be "issue".

Issue name. This is required for a serial when applicable. It must be the alphanumeric
designation of the issue. E.g. if this is issue 3, the Issue name must be "3".

Date. This is required for a serial when applicable. The date in this field will be used to
make a chronology tracing for the date of publication of the original materials. Use the
date of publication for the issue in hand. For issues that don't actually have real
publication dates printed in, you need to infer the date or determine a probable date of
publication and enter this in brackets with a question mark, e.g. [1923?]. The date will be
displayed as "Publication date" for the item in Textual Collections as well as for sorting
the search results according to "date ascending", "date descending". The date in this field
will also be used for statistical purposes. Note that the Date recorded for the issue should
be the same as the Date used for the volume. [Issue]










I*MX lin -[F80035 ~FlnF


F-lIE I6c 3,j.. IT-_ 'r,'. ,.n,:.," Mrtdul-


S

S


Type of part
Issue name
Title
Date
Creator
Rights


III''


*asian|i f-, K V 17CrA a


I I W' Ir0H~I Iii I I r9 I 'aLJA LJ-- IJL.jI M Z rS '


X.d.3. Contents of issues


Issues of serials generally contain material such as letters to the editor, articles,
advertisements, etc. Each article or other section should be a child node under the parent
node for the Issue.


Node name and Type of part. The name of the node (Division name) should be what you
want to appear in the online table of contents for the issue. If the node is an article, you
should generally use the title of the article; however, if the title is very long, then use a
short form as the node name, and put the complete title in Title. If the node is some
section of the issue that does not have a name, make up some descriptive heading, e.g.
"Letters". Here are suggestions on how to treat journal divisions:


division


permission*
volume
issue
front cover


Division name


Copyright permission
Volume X
Number X
Front cover or
Front matter


Type of Part


permission
volume
issue
cover


.-151 xl


1'-=-,,-- I I:i ::- "' L-.. .- -I

_Jl~i:iII ,:n
_ J ,j. ", ,Jl. ,J:..: .-1

J Ii- i iiii I*


9l.I I,-i lll'll'l'l~l, ,h l.,I


,i, re", I
Idd rI-,,
,dd i', |- .
^.dd rI-E- |









inside of front cover
table of contents

letters to the editor
article
abstract (within article)
advertisements
editorial
index
inside of back cover
back cover


Front matter
Table of Contents or
Contents
[name of letters section]
[name of article]
Abstract
Advertisements
Editorial
Index
Back matter
Back cover or
Back matter


cover
contents

correspondence
article
abstract
advertisements
editorial
index
cover
cover












Date. You do not need to supply a date here if one was provided at the issue level.

Creator. If the Division node is for an article with one or more named authors, you
should enter the name of the author in this element. Repeat the element for multiple
authors.

An example of a serial showing many sections as contents is given below. [Serial]




S-i I .I I I ., ,
.I MIFCiII, ,,,U,,1 ,-,,,,--I
.:, -jei X |
II




T e1J .-.1 "" I' ..: :


RN II ,"

TJy ,- I, ,



Type of par..t.'


Section name
Title
Date
Creatoi
Rights


I~i1,.=I


siJlaJ I Il r m, I[A 11 IfE c 7.. Li. r. j. 1-2 |'-'F 7 -T | MXF '*Srlr9 I '-



X.e. Serial title changes

For a serial with title changes, a unique Entity ID should be created when the title
changes. The Entity ID is a unifying identifier that ties all the pieces under each title
together.

Alternative title. This field is used to trace preceding titles) or/and succeeding titles) for
serials with title changes. You must record the preceding title in the form of "Preceding
title: preceding title", and the succeeding title as "Succeeding title: succeeding title", e.g.
"Preceding title: Report of the Board of Control of the State Institutions of Higher
Learning of Florida for the biennium ending ....", "Succeeding title: Report of Florida
Board of Regents".











XI. SPECIAL FORMATS: MAPS AND PHOTOS


Single sheet maps and photos can be added to the FCLA Document Loader as single page
documents, or can be added to the FCLA Visual Materials Collections server. If a map
is destined for a Visual Materials Collection, there must be two images of the map
supplied: a MrSid (.sid) vector image, and a JPEG thumbnail. Data entry for maps is
quick and easy, as most of the required Workform information can be defaulted.


Xl.a. Directory structure

Each single-sheet map or photo is treated as an Entity. A package can contain a single
map/photo entity, or multiple map/photo entities. If your package contains a single
map/photo entity, the directory structure can be very simple: a single directory to hold the
package and images.

\UF00005678 [packagelD -- should be the same as the Entity ID]

If the package contains multiple map entities, then a directory should be created with the
name of the Package ID, and subdirectories for each map should be named for the map
Entity IDs.

\UF800000019 [packagelD]
\UF00005678 [directory for first map]
\UF00005679 [directory for second map]


XI.b. Package level information

Package ID (Node name). When you create the package node, the Package ID (node
name) should be in the standard package ID format, consisting of the 2-character
institution or project code followed by an 8-character numeric value, e.g. "UF80013549".

Image directory. If all images being sent in the package are in a single directory (that is,
if you are sending a single map at a time) set the Image directory to that directory. If
multiple maps are being sent, set the Image directory to the highest level directory.

Make records. If maps are to be processed into a Visual Materials Collection, Make
records should specify "map".

Make derivatives. If maps are to be processed into a Visual Materials Collection, all map
derivatives should be sent in the package, and no Make derivatives instructions should be
specified.









Target server. If maps are to be processed into a Visual Materials Collection, the Target
server should be "IC".

Formats included. If maps are to be processed into a Visual Materials Collection, the two
Formats included should be "Image/jpeg" and "Image/sid".


XI.c. Entity level information

Entity ID (Node name). The node name of a map entity should be the two-character
institution or project code followed by a 8-digits to create a unique identifier. The Entity
ID should be the same as that entered in the bibliographic record for the map in the 035
field.

Entity type. Should be "map".

Project code(s), Bib record key, Bib record system, and Title should always be supplied.

Copyright. A copyright statement should generally not be supplied, as this is will be
taken from the associated bibliographic record.


XI.d. Division level information

Maps/photos intended for a Visual Materials Collection will not require any Division
nodes. The File nodes for the map images can be moved directly under the Entity node.


XI.e. File level information

File nodes for the map image can be moved directly under the Entity node.

Type of part. Type of part should be "main". [Map]












1 I1 .MXfllent 1D1 P 3


E- F'IE l c6 ,.j.. I rs 'I?. -. n rfls" ltdul.E


Type of part

Main name

File name

File lorinal

File order
Filesiz


-151 xl


-II


II'' I


IInill !#rr. I- Ir..:. I L. j i .- rr. Il nli Pi |' IM XF ,r:.na,. li, I J|" Q -l' .i ')


fluA in I'. LI' I 2.1









ii I
IzUUUUU1i~ ~d


0 l F ili,, i-I, Il
I ...h i-...qt,,.tl














XI.f. Related Images


For more information on Related Images see "File Level" in Editing Guide To Map &
Photo Metadata.

Metadata for Image Files

Images that are related in some way, such as the front and back of postcards, can be
displayed together in Visual Collections when "Other Views" is selected from the full
image display page. In order to associate the related images there are several elements
that need to be entered when creating the MXF.

Packages with non-related images contain only one image file, per image type, per
package, and all of the descriptive metadata are entered at the entity level in the MXF.
Packages for related images, however, contain two or more related images and the file
section for each image contains the descriptive metadata that is specific to that named
image file only, and is different from the other file(s) in the package. Any metadata that
the images share in common can be entered at the entity level. If all of the metadata are
the same, i.e., if all of the elements such as title, creator, date, etc. are the same, they can
be entered at the entity level. If any of the metadata is different it must be entered at the
file level.



MARC record creation

MARC records are created from the entity level metadata only. Metadata at the file level
will be displayed on the Web but will not become part of the MARC record.



New Elements for Related Images

There are four new required elements at the file level for 'Related Image' files. These
elements will indicate where the thumbnail images will be displayed in the "Other
Views" web page.










For example if you have nine images that you want to display together in "Other Views"
as follows:


For the image you want displayed in the upper left-most part of the web page, you would
code "Display Row"= 1 and "Display Column"=l, and so on. Keep in mind that four
normal size thumbnails will fit across a web page, while only three map thumbnails will
fit across a web page.

Each of the thumbnail image groupings in "Other Views" will show "Display Face" and
"Display Type" at the top of the group of related images.

For example: front|summary

Each new combination of "Display Face" and "Display Type" will give a new grouping
display in "Other Views". The number of thumbnails can vary for each grouping.

For example:


front|summary
1,1 1,2


1,1 1,2 1,3



2,1 2,2 2,3



3,1 3,2 3,3












back|detail
1,1 1,2



2,1 2,2


To show the front and back of a postcard you would get the following display in "Other
Views":

9 A


*-tS ual COLLECTIONS

view: C full image C description r( other views
other views of: (MM00012956)


Return to search results I << previous item I next item >>


.1Mtl B* 1* ^ -.,41-Q, JI,.&MesqeS.I;. 6:3sPM


10 Start|I I JO A| -4,












Image File Naming

In order to distinguish between multiple images in packages of 'Related Images' the file
names must differ. Each file name must include a suffix with an incremental number
after a dash (-).
For example, in the package named UF00005000, which contains three images, the files
would be named, UF00005000-1, UF00005000-2, UF00005000-3, and so on.










APPENDIX A: WORKFORM ELEMENTS


PACKAGE LEVEL

New Package?
This must be "yes" or "no". If this is a revision of a package sent to FCLA earlier with
the same control number, New Package should be "no". Otherwise, New Package should
be "yes".

Contributing institution
This element indicates the institution that did the scanning and/or metadata creation.
Ordinarily a short code for the institution (e.g. "FAU" "UF" "FGCU") is sufficient. This
is for "manual" use only, and helps to identify who should be contacted if problems exist.

Image directory
This must be set to the fully qualified directory name of the directory containing the
images for this package. If images are in multiple subdirectories, Image directory should
be set to the next higher level directory containing the subdirectories. The Client will
supply this directory name by default as the path to all image names. The Client will also
check that all images in this directory and its subdirectories are referenced in the package,
and give a warning message if any are not.

Make records
Indicates if bibliographic records are to be created by FCLA programs from data in this
package.
analyticss, article" = make a bibliographic records for every Division with
type="article".
analyticss, chapter" = make a bibliographic record for every Division with
type="chapter''.
"title, xx" = make a bibliographic record for the entity, and put it in the Institution Group
named by the xx.
For example, "title, QC" will make a MARC record for the title in QC.

If omitted, no bibliographic records are created.

Make derivatives
Indicates if derivative images should automatically be created by FCLA programs from
master images in this package. Only one pdf derivative level may be selected.
"image/jpeg" = create a JPEG derivative for each TIFF image sent with the MXF
"pdf/article" = create a PDF file at the article level; this includes a PDF for each
Division with a "type of part" article, advertisement, contents,
correspondence, cover, editorial, index or section.
"pdf/chapter" = create a PDF file at the chapter level; this includes a PDF for each
Division with a "type of part" chapter, advertisement, bookplate, contents,
correspondence, cover, editorial, index or section.









"pdf/issue" = create a PDF file at the issue level; this includes a PDF for each
Division with a "type of part" issue.
"pdf/main" = create a PDF file at the main level; this includes a PDF for each
Division with a "type of part" main.
"pdf/volume" = create a PDF file at the volume level; this includes a PDF for
each Division with a "type of part" volume or issue.

If omitted, no derivatives are created.

Target server
Indicates which systems) the files are intended for.
"DL" or "TC" = Text collection server, for page images and PDF files
"FT"= Text collection server, for SGML encoded full text and page images
"IC" = Image collection server, for collections of visual materials such as maps
and photographs

Formats included
The formats of digital files included in the package. Use the authority list in the
pulldown menu.

Timestamp
Date and time of package creation. Supplied by the system and cannot be changed.



ENTITY LEVEL

Entity type
Enter monograph, serial, etc. from pulldown.

Source institution
This is the institution that holds the original from which the contributed material was
digitized. If the original entity was "born digital", such as an ETD (electronic thesis or
dissertation) the source institution is the institution holding the digital original.

Project codes(s)
Enter all project codes applicable to this entity.

Version statement
Must use or default a value from the pulldown.

Copyright
If the source material is under copyright, a statement indicating the material is under
copyright and was digitized with permission of the rights holder is required.

Bib record key









If a NOTIS record exists for this entity, supply the NOTIS record number in the form
AAA9999 here.

Bib record system
This indicates the system and file in which the record with the specified "Bib record key"
will be found. For NOTIS records, use either "FCLANOTIS:QC" or "FCLANOTIS:QF"
depending on whether the bib data is in the institution group QC or QF.

Title
The title of the entity as it appears on the piece or as it would appear in a catalog record.
This is required at the Entity level, even if there is an external bibliographic record, to
serve as a double check that the correct bibliographic record has been referenced.

NOTE: The elements below, from Creator to xxx are part of an optional bibliographic
description which should not be used if a NOTIS record exists for the entity.

Alternative title
A second or alternative title for the entity. For a serial, this could be a former or
succeeding title.

Creator
The name of the primary author of the entity.

Contributor
The name of a secondary author, editor, illustrator or other person or organization
associated with the entity.

Publisher
The publisher of the entity. Use this data element to record the publisher of the source
material (e.g., Hawthorne Press) not of the electronic version (e.g. University of Florida).

Subject
LC Subject headings, or keywords, or other descriptors that describe the content of the
entity.

Description
A free text narrative description of the content of the entity. If the description is an
Abstract, use the Abstract element instead.

Abstract
An abstract of the content of the entity.

Table of Contents
The table of contents of the entity. This will not be used in generating a table of contents
display, but may be used to provide searchable keywords in the catalog system.









Description note
Do not use; use Description instead.

Description release
A version, release or edition statement pertaining to the entity. Use this to describe the
source material, not the digitized version.

Rights
Use Copyright for a short statement of copyright status and permission to digitize. Use
Rights for a longer explanation of complex terms and conditions.

Extent
Size of the entity. Use for filesize of the digital version. If the digital version is made up
of many small files, you can estimate the total filesize here, but indicate it is an estimate
and how it is derived, e.g. "Roughly 5MB in 15 PDF files".

Medium


Type


Coverage
Use for the geographical or temporal coverage of the content of the entity. For example,
"1860-1864" or "Alachua County".

Date
A date associated with the entity. This date is used for statistical purposes. If describing
a serial, see the chapter in this manual, Special Formats: Serials, for instructions how to
record this date.

Creation date
Creation date of the digital version.

Publication date
Publication date of the source or digital version, if formally published. Do not use for the
date of making a digitized entity available on the web.

Modification date
The date of last change of the entity.

Identifier
You can use this to record any standard numbers) assigned to the entity, such as ISSN or
ISBN, as well as digital identifiers such as DOI.

Source









Do not use.


Is part of
If the entity is part of a larger publication (for example, a map that is reproduced from an
atlas, or a book issued in series) give bibliographic information for the larger publication
here.

Has part
If the entity has component parts, you can give a bibliographic description of the parts
here.

Is version of
If the entity is a bibliographic version of another entity, you can describe the other entity
here. Do not use for physical versions (i.e. for TIFF and MrSid versions).

Is format of
If the entity is a derivative of another physical representations (e.g. a sid file made from a
TIFF) you can describe the source version here.

Has format
If the entity has other physical representations, you can describe them here. (Use "Is
format of" instead if this version was derived from another representation.)

Language
The language of the content of the entity, if not English. E.g. "French".

Thesis degree
If the entity is an ETD, then enter the degree granted here, e.g. "B.A", "M.S.", "PhD".

Thesis discipline
If the entity is an ETD, then enter the department in which the degree was granted here,
e.g. "Organic chemistry".

Thesis grantor

Thesis level


DIVISION LEVEL

Type of part
Select from pulldown. The type of structural part that this division represents, e.g.
"chapter", "article", "volume". Be as specific as possible, but if no part name in the
pulldown seems to apply, use "section". Use "main" for the main part of an ETD, if
represented by a single PDF.









[type of part] name
The label for this element will change depending on the type of part chosen, i.e. if Type
of part is "chapter", then this label will say "Chapter name". Use this element when the
parts have formal numbering that consist of the part designation and some alpha or
numeric enumeration, e.g. "Chapter 1", "Part II".

Namespace references)
Do not use.

Note: Remaining elements, from Title through Thesis level, are essentially the same as at
the Entity level. However, the meaning of the bibliographic data pertains to this
particular Division instead of the entire entity. E.g., Title refers to the title of the
Division.


FILE LEVEL

Note: If the file is for Visual Collections and there is more than one file in the package,
bibliographic data that is specific to an image file needs to be entered at the file level. The
elements, from Title to Thesis, are essentially the same as at the Entity Level. However,
the bibliographic data pertains to this particular file rather than to the whole entity. If an
element is the same for all the images in a package, it can be entered at the entity level.
(See XI.f Related Images for a full explanation of RelatedImages.)

Type of part
Select from pulldown. The type of structural part that this file represents. For example, a
PDF of a book chapter would have Type ofpart "chapter". A TIFF of a book page would
have Type ofpart "page".

[type of part] name
The label for this element will change depending on the type of part chosen, i.e. if Type
of part is "chapter", then this label will say "Chapter name". Use this element when the
parts have formal numbering that consist of the part designation and some alpha or
numeric enumeration, e.g. "Chapter 1", "Part II".

File name
The fully qualified filename (including extension) of the file, e.g.
m:\user\Priscilla\DigitizedBooks\FA00000033\1.tif

File format
The MIME Type (Internet media type and subtype) of the file, and the version of the file
if applicable; e.g. "image/tiff 6.0", "application/pdf x.x."

File order
Files will be ordered by default in the same sequence as the metadata describing them. If
some other order is required, you can use File order to give an absolute ordering. If any









one file in a package has a File order, then all files in the package must. This element is
generally not used.

Filesize
File size in bytes. This is supplied by the client from the file header whenever possible.
When you must supply a value, be sure to give the value in bytes, and do not include any
unit designation, e.g. "543210" not "5.4 MB".

Checksum
A calculated checksum for the file. This will be supplied by the Client and cannot be
altered.

Checksum type
The type of checksum calculated. This will be supplied by the Client and cannot be
altered.

Timestamp
The date and time of file creation. This is supplied by the Client from the file header
when possible.

File id
An internal identifier supplied by the Client.

Creating institution
The institution responsible for image creation. This should be given at the level of the
library, rather than of the university or library department, e.g. "University of Florida,
University Libraries".

Creating individual
The person who actually created the image via scanning, digital camera, etc. The
individual's department or other information can be supplied if there is a possibility of
confusion, e.g. "George Smith, Digital Library Center".

Creation method
How the image was created, for example, "Scanned from paper" or "Output from MS
Word in PDF format".

Software used
The name of the software package used to create the file, e.g. "Adobe Capture".

Software version
The version of the software used, e.g. "5.2".

Name of compression scheme
If the file is compressed, the name of the compression scheme used.









Compression
The compression ratio, or other textual information about compression apart from the
name of the scheme.

Note: The Following four elements (Display Row, Display Column, Display Type,
Display Face) are required only in packages that have one or more Related Images in a
package. (see Appendix F for a full explanation of Related Images.)

Display Row
The row number where the thumbnail image will display when 'Other Views" is selected.

Display Column
The column number where the thumbnail image will display when "Other Views" is
selected.

Display Type
The value that will display at the top of the group of thumbnail images when "Other
Views" is selected. Values are sum (summary), and det (detail). If not entered, will
default to "sum".

Display Face
The Second value that will display at the top of the group of thumbnail images when
"Other Views is selected. Values are:front, back right, left, top and bottom. If not
entered, will default to front.

Capture
A textual description of the capture method. It is better to use Capture device used and
Type of capture device.

Capture device used
The make and model of the digital camera or scanner used.

Type of capture device
Select from pulldown.

Scanner settings
If the capture device was a scanner, record the settings in Gamma, Brightness, Contrast,
etc. and set Scanner settings to "yes".

Gamma/Brightness/Contrast/Saturation/Percentage
Record the settings used to drive the scanner used as capture device.

Camera settings
Do not use.

Light source









Textual description of source of light used when scanning.


Tracking id
If you have an internal tracking id number for this batch of images, you can include it
here.

Image bit depth
Record as a number, e.g. 1, 16, 24.

Image resolution
For example, 600 DPI.

Storage segment
For TIFF images, whether recorded as strip or tile. Select from pulldown.

Storage planar configuration
For planar formats only, whether the image is chunky or planar. If not applicable, use
"unknown".

Unit of sampling frequency
Inch, centimeter, etc. Select from pulldown.

Sampling plane
The reference plane for sampling frequency. Select from pulldown.

Image dimensions (horizontal)
The number of pixels in the sampling frequency unit in the image width.

Image dimensions (vertical)
The number of pixels in the sampling frequency unit in the image length.

Image color space
Color space (color mode) of image, e.g. linear RGB.

Color management
Description of any process used to improve consistency of color across capture, display,
and output devices.

Color lookup table
Color lookup table used.

Targets included
Textual information about color bar or greyscale bar used as internal target.

Type of target included
Either "color" or "greyscale", select from pulldown.










Descreening software
Software used for descreening.

Descreening settings
Settings used with descreening software.

Orientation
Either "portrait" or "landscape", select from pulldown.

Derived from (filename)
If this file was derived from another file (e.g., a JPEG image made from a TIFF), you can
give the filename of the source file here. Alternatively, if the source file is described in
this MXF package, you can leave this element blank and give the File id of the source file
in the following element. NOTE if you use the Client's Derive Node function, this is
automatically supplied.

Derived from (file id)
If this file was derived from another file (e.g., a JPEG image made from a TIFF) included
in this MXF package, you can give the File id of the source file here. This is an
alternative to giving the filename in the previous element. NOTE if you use the Client's
Derive Node function, this is automatically supplied.

Derived from (description)
If this file was derived from a non-digital original, or if the relationship between the
source file(s) and this file more complex than a one-to-one derivation, use this element
for a textual description of the files involved and their relationship to each other.

Source dimensions (horizontal)
Horizontal (width) dimensions of a non-digital source file scanned to make this file, e.g.
"35". Do not include the unit of measurement here, but in the Unit of source dimensions
element.

Source dimensions (vertical)
Vertical (height) dimensions of a non-digital source file scanned to make this file, e.g.
"50". Do not include the unit of measurement here, but in the Unit of source dimensions
element.

Unit of source dimensions
The unit of measurement for horizontal and vertical dimensions of the source, e.g. "cm.".

Namespace references)
Do not use.

Note: Remaining elements, from Title through Thesis level, are essentially the same as at
the Division and Entity levels. However, the meaning of the bibliographic data pertains









to this particular File. As a general rule, bibliographic description should be given at the
Entity and or Division levels, and not at the File level.

An exception to this rule is made for "Related Images". If there is more than one JPEG
image in an MXF package, i.e., one or more related images, the bibliographic description
must be given at the file level if the element values differ for the images. For example, if
the images have different titles and descriptions, these must be indicated at the file level.
Bibliographic description that is shared between images is indicated at the Entity level.










Appendix B: How the MXF is encoded


The MXF is encoded in XML. Data elements are delimited by start tags and end tags.
Start tags look like this:


End tags look like this:


The value of the data element is usually given between the start and end tags, e.g.
FHP
means the value of the data element "projects" is "FHP".

Some data elements have attributes, or more specific sub-elements. In this case they are
shown within the start tag in attribute= "value format:



Elements can be nested within other elements, in which case the meaning is taken to be
hierarchical. For example, there may be two elements nested as follows:

...
...



In this case, it is understood that the page is contained within the chapter. (Note that
although this kind of nesting is often represented graphically using indentation, the actual
XML is not indented.)

The MXF Client software is designed to hide the actual XML syntax of the MXF from
the person doing the data entry. Rather than seeing start tags and end tags, you see screen
labels and boxes for entering data. When all the information has been entered, you can
tell the Client to export the data in the appropriate XML format.










APPENDIX C: VALIDATION MESSAGES


The MXF Client has two kinds of Validation Messages: Errors and Warnings. An Error
must be fixed before XML can be output. Warnings do not have to be fixed.


Appendix C. 1. Error messages

The following are Error messages:


"No value specified for [required element]"
The cursor will be at the required element in the Workform. Some value must be
supplied.

"The name "Volume name" or "Issue name" value (in the metadata workform) for a
volume or issue contained within a serial entity must be numeric."
If the Entity type is "serial" and the Division node's Type ofpart is "volume" or
"issue", then the Volume name or Issue name field must have some value. For
example, if you are describing Volume 43, then Volume name should be "43".

"[filename] is not an existing file."
In a File node, the file identified in File name was not found in the Image
directory for the package, or in any of the subdirectories of the Image directory.

"The 'Derived from' value [file id] is not a valid file id from this package"
If you use the "Tree/Derive node" function to create a derived file node, the
source node must exist in the package at the time of validation. If you get this
error message, somehow the source file must have been removed from the Image
directory.

"The specified image is not in the path given in the package level
image directory."
A file node references a file that was not found in the image directory specified in
the Package level field Image directory.


Appendix C.2. Warning messages

The following are Warning messages:

"No images have been referenced in this package."
The package does not refer to any files. That can only happen if the package does
not contain any File nodes, since a Filename is required on the File workform.









"Unable to determine image directory."

"Image directory cannot be located, XML file will be stored in the package directory."

"The file [file path] exists in the image directory, but is not referenced in
the package."
The directory designated on the Package workform as the Image directory
contains certain files that are not referenced in the package. This message will
only appear if the filetypes of the unreferenced files are .tif, [what else]










APPENDIX D: OPTIMIZING WORKFLOW


Organizing your workflow and using shortcuts, defaults and control keys can speed data
entry significantly. Here are some suggestions for a few common cases.


a. Book, images exist

In this example, there are TIFF and JPEG images for every page of a book.

a. 1. Set up your new package and Package node.
a.2. Set up your Entity node for the book.
a.3. Create nodes for all Divisions:
Have the Table of Contents or Division List in front of you.
For nodes with unique names (e.g. Copyright permission through Foreword):
Use CTRL-a (Add node) to create the first Division under the Entity node.
Hit the back arrow (up one level) to highlight the Entity node again.
Use CTRL-a (Add node) to create the next Division.
Hit the back arrow (up one level) to highlight the Entity node again.
Continue creating Division nodes with CTRL-a and back arrow until done.
For nodes with sequentially numbered names (e.g. Chapter 1 through 10):
Use CTRL-s (Insert Range) to create nodes for these Divisions.
a.4. When all Divisions have been created, create File nodes for the images:
For each division, highlight the TIFF images for the pages in that division (Hold
CTRL down and click the image names).
Drag the TIFFs from the Directory Window to the appropriate Division in the
Tree Window.
For each TIFF, highlight the TIFF node and use CTRL-d (Derive node) to create
a derived node for the JPEG.










APPENDIX E: TRAINING EXERCISES


Exercise 1: Book worksheet

Preparing a worksheet. To create an MXF you need to know the structure of the
publication. If you won't have the publication in front of you when you make the MXF,
you can record the important information on a worksheet. In this exercise, because we
don't have the source publication even to make the worksheet, you will have to pretend
you do by inferring the information from the online version in Florida Heritage.

1. Read Section I "Understanding the MXF" and Section II, "Preparing Documents".

2. In Florida Heritage, look up the title, Conjured into being: Zora Neale Hurstons Their
Eyes were watching God. Fill out Worksheet 1 (empty) on the following page.

The "Division Name" is what will appear in the table of contents display. "Type of Part"
must be taken from the following list: abstract, advertisement, article, box, chapter,
contents, correspondence, cover, editorial, file, item, main, page, part, permission,
recordgroup, section, series, subgroup, subseries, supply, volume. Fill in "Page or range"
when the division consists of numbered pages (e.g. "i-vii").

"Comments" can be used to record the filename, if other than the page number. (There is
a convention to name files the same as the pages, e.g. the TIFF file for page 1 would be
1.tif) Since you don't know the filenames for this exercise, you can omit this
information. "Comments" can also be used to record additional bibliographic
information for the section, such as an author or long title.

The first entry is filled in for you.

3. Compare your filled-out worksheet with Worksheet 1 (completed). Yours does not
have to look exactly like this but be sure you are comfortable with the differences.





Worksheet 1 (empty)
Division Name Type of Part Page or range (n-n) Comments
Front Matter section approval, title


_____________________________ _____________________________ _____________________________ I _____________________________


___ ___ ___ I ___


___ ___ ___ I ___







Worksheet 1 complexe ed)
Division Name Type of Part Page or range (n-n) Comments
Front Matter section approval, title
Table of Contents contents ii
Abstract abstract iii-v
Introduction chapter 1-11
Free indirect chapter 12-16
discourse
Signifying chapter 17-23
Conjuring chapter 24-28
Signifying/ chapter 29-33
Conjuring
Conclusion chapter 34-38
Literature cited section 39-41


_____________________________ _____________________________ _____________________________ I _____________________________










Exercise 2: New package


This exercise will use the images in the SF00000012 folder. Before beginning, create a
directory for your MXF work. For example, this might be called
"/path/yourName/MXFwork". Copy the SF00000012 folder into /MXFwork. Pretend
you are at USF, creating an MXF to transport these images to FCLA for loading.

1. Read sections III "Basics of the MXF Client", IV "Setting Defaults", and V "Creating
a New MXF package".

2. Log into the MXF Client.

3. Set the following Defaults:

Package:
New package? (yes)
Contributing institution (FSU)
Make derivatives (application/pdf)
Target server (DL)
Formats included (image/jpeg, image/tiff)

Entity:
Source institution (USF)
Version statement (Electronic version created 2001...)
Bib record system (FCLANOTIS:QF)

Division:
Type of part (chapter)

File:
Type of part (page)
Image format (image/tiff)

4. Create a new package named "SF00000012". Set Image Directory on the metadata
workform to the correct directory. Make sure the correct directory is in the Directory
window and that the images are listed. Fill in any missing information on the metadata
Workform at the Package level.

5. When you are done, compare your completed workform with the examples below.
[Exercise 2-1, Exercise 2-2]. They don't need to be identical but be sure you understand
any differences.













I1*I Cient- 11FOlDlIll ] -FlR l l


New Package?

Contributing institution

Contributing individual

Image directory

Make records

Make derivatives


I'A-- i l I"L:: I' L.... -I if



1:..







I ,I l" Il I
l iii l i: il in ii
i q i- ii i-ip,




II I-
m e iI i,- i,, a. i.i
i Piu,: iPu~i iili,:
.i.1i.I.: i _i_,,.i__.__ l.i
i i.i I i il .i ill.
i i l I. W iii i I Ii i



l ii i 11


I-.

hr

I *-*''


Eli. .t


sia fJ-lI K V .r.. r | .CA rT'ii.. | i iir.I ....rii | iI--. iiE L II'MXFCI -t :.r.i.I, Mi | 'J9












I1*111F Cliet-lFlIIfDl I 2llln ll3


E FlIE l6 ,q*. I


r'. rjinci-11I ilCtdui-


Make records
Make derivatives

Target server
Formats included
Formats included

Tilnei rtanip


*asian|i f- K V ri CA R


rr,: I i


ll .I- ,n
IllI F.I 77 7lI .I


-151 xl


I ji.. i :-ii L 1 I j
J3.,,
_3ii...iin


VIPI,1 :'. p. I
"i .w i i.i I il.a
SI I W ':Iil Wa I"
lii ira*)W ..P;
1 -T


-J "-1 r-e I


- P ...


I Y |. ". I -i- I.r. ..i.| I-rn I-.I .,.| L II MXFCI a l :.a..i_, I,-| .rSQ '


6. Close the package by selecting Package/Close. Close the Metadata Module by
clicking the close box at the right of the top Menu bar. Close the MXF Client by
clicking the close box in the upper right-hand corner.


- :i-i-i-ii-i-Iil "










Exercise 3: Adding Entity and Division level information


In this exercise you will use the worksheet you created in Exercise 1 to add Entity and
Division nodes to the tree. Assume there is already a NOTIS catalog record for this title
in the QF (PALMM) institution, and that its NOTIS record number is AAA1070.

1. Open the MXF client, using your username and password.

2. Open the Metadata module ( l,,/il,/e _1\ /A',//).

3. Open the Package for SF00000012: Select Pac kIge Open. A popup box will suggest
the last package file that you had open. If this is correct (it should be) highlight the name
of the package and click Open. Note that the correct Image directory will display in the
Directory Window, and the Package workform displays.

4. Create an Entity node: Select Tree Add node. The Client assumes you want to add an
Entity node as a child to the Package node; this is correct. Type the name of the Entity
Node in the box. For non-serial publications, this should be the same as the name of the
Package Node, so enter "SF00000012" and click OK or hit Enter. The Entity node will
appear on the tree, and an Entity workform will display.

5. Add Entity level information: Fill out the Entity workform. Make sure you have
filled out Bib record key (AAA1070) and Title (Conjured into being: Zora Neale
Hurstons Their Eyes were watching God).

6. Create a Division node: Use CTRL-a (same as Tree Add node). The Client assumes
you want to add a Division node as a child to the Entity node; this is correct. Type the
name of the Division node in the box. This should be whatever you recorded on the
worksheet -- it will display in the table of contents for this entity. If you followed the
sample worksheet, it should say "Front Matter".

7. Add Division level information: Probably the only thing you need to look at is "Type
of part". You may have to change the default to match what you have on your
worksheet. If you followed the sample worksheet, it should say "section".

8. Add remaining Divisions: Make the Entity node the current node. If the Tree window
is active, just hit the Left arrow to take you to the parent of the current node, which will
be the Entity node. If the Workform window is active, press F2 to make the Tree
window active, then Left arrow. Alternatively, you can just click the Entity node to make
it current, but using the mouse is usually slower than the other methods.

Use CTRL-a to create a new node. The default node type (division, child) should be
correct. Enter the Division node name here, and the correct "Type of part" on the
workform. Repeat for all remaining divisions on your worksheet. (Tip: if the default
"Type of part" was correct and you did not need to enter anything in the Workform
window, the Tree window will be the current window. Hitting the Left arrow will take













you to the parent of the current node, which will be the Entity node. This is faster than
clicking on the Entity node.)


9. When you are done, compare your work to the example below. [Exercise 3]. Be sure
you understand any differences.

r-MX Client ,- [SFOO00001 2] P lnxl


r-, F,Ie F0c1 ,.} Tr-t i1.. ,,irid,:.. ML.ul-


mm -i .,. : )lil : ,~





mm i:1,-i i i1 l n, ,.
4 m ,I-- ,r ,, ::,, .,
* 1':,, i,.:-,1,< -
U _nn ~n :1 ,


Type of part
Chapter name
Namespace references i

Title
Alternative title
Crealor



.isinai gC K Fr n 2 V


.JIJJ


,'.,, .







'I, r rFP 4 ,,.
l:,,, ,_,,,1 ..






II',II i 'II .
'iI *i I I I. ii 'I
I s i1i Iw hi I I. 1 *
Siii 1
PWi ':|')P ''1k-i- Wi T 'i'=i


I : ..... .. I -ih .E....:... .. ..3. -.. I" I' '-" *, t ir.-; i || iSnaI"gl/3 Z | S 0


10. Close the Package: Close and save the package with Package/Close. You can leave
the client open in the Metadata module.


-'J'j
.,Jl
-'J'j










Exercise 4: Adding File level information and finishing the Package

In this exercise you will finish the package you started building in Exercise 2 and 3.

1. Open the Package for SF00000012. If you left the tree in expanded form (with all
nodes showing) it should still be expanded. If any node is compressed (preceded by a
plus), click on the plus to expand it, or select Tree/Expand all.

2. Make nodes for TIFF files: Using your worksheet to relate pages to divisions, copy
over the TIFF images for the first division. In this example, the first division is Front
Matter and there are two pages in it, represented by 1 .tif and 2.tif. To move more than
one image at a time, hold down the CTRL key while clicking on the name of each image
in the Directory window. Then clicking on one of the images, drag it over to the correct
node.

3. Make nodes for derivatives: Click on the first TIFF (1 .tif) and check that the
Workform data is correct. Then use Tree/Derive node or CTRL-d to create a node for
the derivative JPEG. The cursor will be at the end of the default File name in the
Workform window; type the file type (.jpg) and press ENTER. Default information for
the JPEG image will be supplied.

4. Make remaining file nodes: Continue making nodes for the TIFF and JPEG files by
dragging the TIFF files from the Directory window and making derivative nodes for the
JPEGs.

5. Validate and output the package. Make sure that "View/XML Tree on Output" is not
checked. Select Package/Output to XML. This will cause the MXF client to validate the
data you have entered. There will be a popup box giving the number of errors and
warnings. Click "OK" to clear the box, then use PF5 to review each error in turn.
Correct any errors found and repeat "Package/Output to XML" until no errors occur.
You will get a message saying the XML file has been created.

6. Now select "View/XML Tree on Output" so that it is checked. Select
"Package/Output to XML" again. Note that this time you see a tree version of the output
XML file. If you want to continue viewing the tree for other packages, leave "XML Tree
on Output" checked, otherwise turn it off

7. Select "View/XML file". The output XML will display in a Notepad window. Close
the window. Close the package ("Package/Close"). Close the MXF Client.










APPENDIX F: MXF Quick Guides for Textual Materials

Before creating MXF for your textual materials, it is very important to understand and
decide on the correct entity type for your materials. Currently, Textual Collections
supports three entity types for all the textual materials, monograph, multipart and serial.

A monograph (a single part monograph or single volume monograph) is an item
complete, or intended to be complete, in one part.

A serial is a publication issued in successive parts, bearing numerical or chronological
designations, and intended to be continued indefinitely. Examples include journals,
periodicals, newspapers, and annuals (reports, yearbooks, etc.)

A multipart (a multipart monograph or multi-volume monograph) is an item complete, or
intended to be complete, in a finite number of separate parts.


MXF: Quick Guide for Monograph

Submission A complete work with all chapters and other sections should be sent in one
package to FCLA.


Directory setup for monograph


\SF00000012


[package ID]


JPEG, TIFF, PDF, XML, SGML, DTD, etc


Package Level information


Package ID (Node)



Image directory


Make records


10-digit ID in the standard package ID format, consisting of the 2-
character institution or project code followed by an 8-character
numeric value, e.g. "FI12345678", "FA12345678".

The directory where image files are found. In the example above
this would be "...\SF0000012\".

Use "title, QF" or "title, QC" if you want a MARC record made
from the metadata in the MXF. In general, records should only go
in QF if they have valid authority-controlled name and subject
headings.









Make derivatives


Target server


Select "pdf/chapter" if you want PDF files created according to the
division nodes of the following "Type of Part": chapter,
advertisement, bookplate, contents, correspondence, cover,
editorial, index, main, section, permission.

The PDF files created based on JPEGs you send us will be
displayed parallel to the JPEG version of the item in Textual
Collections.

Use "DL" or "TC" to send JPEG page images and have us create
the SGML file to be loaded into Textual Collections.


Use "FT" to send JPEG and/or PDF images and your own fulltext
SGML file for us to load directly into Textual Collections.

Entity Level information


Entity ID (Node)


Entity type

Source institution


Project code(s)


Version statement


Monograph entity ID should be the same as the package ID. In the
example above this would be SF00000012.

"monograph"

The institution who holds the original from which the material was
digitized. The institution code supplied here will govern the "Print
Source" statement in the full citation of the item in Textual
Collections.

For example, if you supply "FIU", the "Print Source" statement
will be "Digitized from original source held at Florida
International University Libraries"; if you supply "FAU", the
"Print Source" will say "Digitized from original source held at
Florida Atlantic University Libraries".

Supply only one primary project code and any applicable sub-
project code(s). Make sure this is consistent with what is recorded
in 852$b of the bib record, if such a bib record exists.

For proj ect code(s), see http://allegro.fcla.edu/Proj ectCodes.cfm

Either select a version statement from the pull-down or enter a
statement that describes the origin of the digital version, e.g.
"Electronic version created yyyy, State University System of
Florida." Version statement will be displayed in full citation of the
item in Textual Collections.









Bib Record Key



Bib record system


Title


Rights


If a NOTIS record exists, supply the NOTIS record key in the form
of AAA1234 here. This associates this package with the external
bibliographic record.

Use either "FCLANOTIS:QF" or "FCLANOTIS:QC" to indicate
the system and file in which the record with the specified "Bib
record key" will be found.

"Bib record key" and "Bib record system" must come in pairs. The
bibliographic record will be used to extract descriptive metadata
and displayed as "citation" for the item in Textual Collections.


The title of the monograph. This is required.


Use for a statement of rights related to the original or digitized
version. If the material is under copyright and digitized with
permission, note this here. In Textual Collections, the Rights is
displayed as Rights Information.


Do not use Copyright to record this information.

Division Level information

Most monographs will have at least one Division level. A monograph generally contains
logical structural parts such as chapters, table of contents, index, cover, etc. Each of these
parts or other section should be a division node. Each of the division nodes usually
parents a number of file nodes which contain the page images. However, there is no
reason that division nodes cannot be nested in division nodes.


Node name








Type of part


Title


The name of the node (Division name) should be what you want to
appear in the online table of contents for the item. If the node is a
chapter, you should generally use the title of the chapter; however,
if the title is very long, then use a short form as the node name, and
put the complete title in Title. If the node is some section of the
item that does not have a name, make up some descriptive heading,
e.g. "Letters".

Commonly used value: permission, cover, contents, section,
chapter, index, cover.

Title can usually be left blank. It's used only when a title is very
long. Anything entered in Title will be used instead of the Node
name in the table of contents displayed to the user. Typing in a
long title as the node name can make the tree look messy and hard
to read. Instead, type an abbreviated form of the title as the node
name, and type the complete title in Title.










File Level information


File nodes should be children nodes under the parent nodes for the contents of book
divisions. File nodes generally contain page images. There are several required elements
at the file level, but most of them can be supplied by the system automatically.

Type of Part Usually use "Page".

File name Automatically supplied when you browse and select the file or use
the drag method to create the file node.

It is recommended that if you have different file formats for the
same page image, they should use the same filename, but with
different file extension.

File format Automatically supplied by the system if it can get this from the
extension on the filename.

It is required that if you have different file formats for the same
page image, you have to create sibling file nodes for them
continuously. For example, if you have both "cover.tif" and
"cover.jpg" for the Cover page image, the file nodes for "cover.tif"
and "cover.jpg" should come in pair consecutively.

File size Automatically supplied by the system when the information is
present in the image header.















An example of a monograph showing many sections as contents.


10 MXFm Ci.= .t -=, [MX -llIn


..3,,,










RlI ii ILK i'.l
l :,,i ,:, i






-- .... i-l






.1 W-4 1 4 -
Ii 1 4 4 1.
I I- 1;,. PQ
14 i q i, ii : ii






FA ,,,, 11 FA i.


II--


USlri| Mr -K E | r .h |MRI ler[i |Iet. 5 .c. c.h || MXF Cl iEu3'L |I li |


/ J :,:,- ,,;l-.,
I"- rn1 lnn



fl-i ,J
*: i I.l l'. l I
Ul .,II
U ',b2?
U ,I.
*ll ,I, I, ,
,.IU i
Ul .I %
SHI.
Ul ., II ,h


Page name

File name

File Fornial

File ordei

Filesi r










MXF: Quick Guide for Multipart


Submission All parts have to be wrapped in one package and sent to FCLA together.

Directory setup for multipart

When set up directories for your multipart package, you must create a separate
subdirectory for each part/volume:

\UF80001549 [package ID]
\UF80001549 [entity ID]
\1 [images for volume 1]
\2 [images for volume 2]

etc.

Use numeric characters only for the name of the subdirectory for each part/volume.


Package Level information


Package ID (Node)




Image directory



Make records




Make derivatives


10-digit ID in the standard package ID format, consisting of
the 2-character institution or project code followed by an 8-
character numeric value, e.g. "FI12345678",
"FA12345678".

Set the Image directory to the directory corresponding to
the Entity ID. In the example above this would be
"...\UF80001549\ UF80001549".

Use "title, QF" or "title, QC" if you want a MARC record
made from the metadata in the MXF. In general, records
should only go in QF if they have valid authority-controlled
name and subject headings.

Select "pdf/chapter" if you want PDF files created
according to the division nodes of the following "Type of
Part": chapter, advertisement, bookplate, contents,
correspondence, cover, editorial, index, main, section,
permission.

The PDF files created based on JPEGs you send us will be
displayed parallel to the JPEG version of the item in
Textual Collections.









Target server


Use "DL" or "TC" to send JPEG page images and have us
create the SGML file to be loaded into Textual Collections.

Use "FT" to send JPEG and/or PDF images and your own
fulltext SGML file for us to load directly into Textual
Collections.


Entity Level information


Entity ID (Node)


Entity type


Multipart entity ID should be the same as package ID.


"multipart"


Source institution


Project code(s)


The institution who holds the original from which the
material was digitized. The institution code supplied here
will govern the "Print Source" statement in the full citation
of the item in Textual Collections.

For example, if you supply "FIU", the "Print Source"
statement will be "Digitized from original source held at
Florida International University Libraries"; if you supply
"FAU", the "Print Source" will say "Digitized from
original source held at Florida Atlantic University
Libraries".

Please be aware of the difference between "Contributing
institution" at the package level and "Source institution" at
the entity level.

Supply only one primary project code and any applicable
sub-project code(s). Make sure this is consistent with what
is recorded in 852$b of the bib record, if such a bib record
exists.


For project code(s), see
http://allegro.fcla.edu/ProiectCodes.cfm


Version statement





Bib Record Key


Either select a version statement from the pull-down or
enter a statement that describes the origin of the digital
version, e.g. "Electronic version created yyyy, State
University System of Florida." Version statement will be
displayed in full citation of the item in Textual Collections.

If a NOTIS record exists, supply the NOTIS record key in
the form of AAA1234 here. This associates this package
with the external bibliographic record.










Bib record system


Title


Rights


Use either "FCLANOTIS:QF" or "FCLANOTIS:QC" to
indicate the system and file in which the record with the
specified "Bib record key" will be found.

"Bib record key" and "Bib record system" must come in
pairs. The bibliographic record will be used to extract
descriptive metadata and displayed as "citation" for the
item in Textual Collections.


The title of the multipart. This is required.


Use for a statement of rights related to the original or
digitized version. If the material is under copyright and
digitized with permission, note this here. In Textual
Collections, the Rights is displayed as Rights Information.


Do not use Copyright to record this information.

Division level information

Most multipart will have at least two levels of hierarchy at the Division level: volume and
the contents of the volume (chapters, index, table of contents, etc.). This is represented
on the Tree as parent and child nodes: the Division for the volume is the parent of the
Divisions for the contents of the volume.

Volume Node


Node name


The node name for the volume should be what you want to
display as part of the title for the item in Textual
Collections, e.g. "Part X: part title"

For example the volume node name "Part 1 General
Statistics" will be concatenated with the title at entity level
"Industrial and economic survey of Pensacola: Industrial
and economic survey of Pensacola", thus forming a title of
"Industrial and economic survey of Pensacola: Industrial
and economic survey of Pensacola Part 1 General
Statistics", which is displayed to users as the title of the
item in Textual Collections.


Type of part

Volume name


Use "volume".


This is required for a multipart. It must be the numeric
designation of the volume, e.g. if this is volume 1, the









Volume name must be "1". If there isn't a numeric
designation, you have to establish one.

Date This is required for a multipart. The date in this field will
be used to make a chronology tracing for the date of
publication of the original materials. Use the date of
publication for the volume in hand. For volumes that don't
actually have real publication dates printed in, you need to
infer the date or determine a probable date of publication
and enter this in brackets with a question mark, e.g.
[1923?]. The date will be displayed as "Publication date"
for the item in Textual Collections as well as for sorting the
search results according to "date ascending", "date
descending". The date in this field will also be used for
statistical purposes.

Contents of volume Nodes

Multipart is multi-volume monograph. Each volume generally contains logical structural
parts such as chapters, table of contents, index, cover, etc. just like monograph does. Each
of these parts or other section should be a division node nested under the volume node.


Node name


Type of part


The name of the node (Division name) should be what you
want to appear in the online table of contents for the item.
If the node is a chapter, you should generally use the title of
the chapter; however, if the title is very long, then use a
short form as the node name, and put the complete title in
Title. If the node is some section of the item that does not
have a name, make up some descriptive heading, e.g.
"Letters".

Commonly used value: permission, cover, contents,
section, chapter, index, cover.

Title can usually be left blank. It's used only when a title is
very long. Anything entered in Title will be used instead
of the Node name in the table of contents displayed to the
user. Typing in a long title as the node name can make the
tree look messy and hard to read. Instead, type an
abbreviated form of the title as the node name, and type the
complete title in Title.


Title


File Level information









File nodes should be children nodes under the parent nodes for the contents of volume
divisions. File nodes generally contain page images. There are several required elements
at the file level, but most of them can be supplied by the system automatically.

Type of Part Usually use "Page".

File name Automatically supplied when you browse and select the file
or use the drag method to create the file node.

It is recommended that if you have different file formats for
the same page image, they should use the same filename,
but with different file extension.

File format Automatically supplied by the system if it can get this from
the extension on the filename.

It is required that if you have different file formats for the
same page image, you have to create sibling file nodes for
them continuously. For example, if you have both
"cover.tif" and "cover.jpg" for the Cover page image, the
file nodes for "cover.tif" and "cover.jpg" should come in
pair consecutively.

File size Automatically supplied by the system when the information
is present in the image header.










MXF: Quick Guide for Serial


Submission

Each issue of the serial has to be sent to FCLA in a separate package, the Package ID
must be unique for each package (issue), but the Entity ID must be the same for all
issues.

For a serial with title changes, a unique Entity ID should be created when the title
changes. The Entity ID is a unifying identifier that ties all the pieces under each title
together.

Directory setup for serials


\UF80001549
\SN00130098
\1 1


[package ID]
[entity ID]
[directory for volume 1 issue number 1]


Use numeric characters only for the name of the subdirectory for each serial issue.

Package Level information


Package ID (Node)




Image directory


Make records




Make derivatives


10-digit ID in the standard package ID format, consisting of
the 2-character institution or project code followed by an 8-
character numeric value, e.g. "FI12345678",
"FA12345678".

The subdirectory for each serial issue. In the example
above this would be "...\SN00130098\1 1".

Use "title, QF" or "title, QC" if you want a MARC record
made from the metadata in the MXF. In general, records
should only go in QF if they have valid authority-controlled
name and subject headings.

Select "pdf/article" if you want a PDF file created for every
article in the issue, the division nodes of following "Type
of Part" will be wrapped into PDFs: article, advertisement,
contents, correspondence, cover, editorial, index, section,
permission.
Select "pdf/issue" if you want a PDF file created for the
issue, the division nodes of the following "Type of Part"
will be wrapped into PDFs: issue, permission.













Target server


The PDF files created based on JPEGs you send us will be
displayed parallel to the JPEG version of the item in
Textual Collections.

Use "DL" or "TC" to send JPEG page images and have us
create the SGML file to be loaded into Textual Collections.

Use "FT" to send JPEG and/or PDF images and your own
fulltext SGML file for us to load directly into Textual
Collections.


Entity Level information


Entity ID (Node)


Entity type


Source institution


Project code(s)


Serial entity ID must consist of the 2-character "SN" and
the 8-character ISSN of the serial. For example, if the
ISSN is 0123-4567, then the Entity ID must be
"SN01234567". The entity ID stays the same for all the
issues regardless of the package IDs that are used.

Some older serial titles do not have ISSNs or some
materials you want to package as serial are not serials In
this case there are two options: apply for a retrospective
ISSN, or assign a pseudo-ISSN. Instructions on how to
apply for a retrospective ISSN, or assign a pseudo-ISSN are
available at
httD://Dalmm.fcla.edu/strucmeta/Dseudoissn.html


"serial"


The institution who holds the original from which the
material was digitized. The institution code supplied here
will govern the "Print Source" statement in the full citation
of the item in Textual Collections.

For example, if you supply "FIU", the "Print Source"
statement will be "Digitized from original source held at
Florida International University Libraries"; if you supply
"FAU", the "Print Source" will say "Digitized from
original source held at Florida Atlantic University
Libraries".

Supply only one primary project code and any applicable
sub-project code(s). Make sure this is consistent with what
is recorded in 852$b of the bib record, if such a bib record
exists.









For project code(s), see
http://allegro.fcla.edu/ProiectCodes.cfm


Version statement





Bib Record Key



Bib record system


Title


Either select a version statement from the pull-down or
enter a statement that describes the origin of the digital
version, e.g. "Electronic version created yyyy, State
University System of Florida." Version statement will be
displayed in full citation of the item in Textual Collections.

If a NOTIS record exists, supply the NOTIS record key in
the form of AAA1234 here. This associates this package
with the external bibliographic record.

Use either "FCLANOTIS:QF" or "FCLANOTIS:QC" to
indicate the system and file in which the record with the
specified "Bib record key" will be found.

"Bib record key" and "Bib record system" must come in
pairs. The bibliographic record will be used to extract
descriptive metadata and displayed as "citation" for the
item in Textual Collections.


The title of the serial. This is required.


Rights


Alternative title.


Use for a statement of rights related to the original or
digitized version. If the material is under copyright and
digitized with permission, note this here. In Textual
Collections, the Rights is displayed as Rights Information.

Do not use Copyright to record this information.

This field is used to trace preceding titles) or/and
succeeding titles) for serials with title changes. You must
record the preceding title in the form of "Preceding title:
preceding title", and the succeeding title as "Succeeding
title: succeeding title", e.g. "Preceding title: Report of the
Board of Control of the State Institutions of Higher
Learning of Florida for the biennium ending ....",
"Succeeding title: Report of Florida Board of Regents".


Division level information

Most serials will have at least three levels of hierarchy at the Division level: volume,
issue (number), and the contents of the issue (articles, correspondence, editorials, etc.).
This is represented on the Tree as parent and child nodes: the Division for the volume is









the parent of the Division for the issue, which is the parent for the Divisions for the
articles within the issue.

Volume Node


Node name


The node name for the volume should be what you want to
display as part of the title for the item in Textual
Collections. It is important that you record all node names
for volumes and issues consistently throughout the entire
serial run. Recommended practice is to use either "Volume
x (year)"or "Vol. x (year)" as this gives the most
information to the reader. For example, "Volume 1", "Vol.
1 (1898/1899)".

"The Florida Historical Quarterly volume 1 issue 1"


Type of part

Volume name


Date


Use "volume".


This is required for a serial. It must be the numeric
designation of the volume, e.g. if this is volume 43, the
Volume name must be "43". If there isn't a numeric
designation, you have to establish one.

This is required for a serial. The date in this field will be
used to make a chronology tracing for the date of
publication of the original materials. Use the date of
publication for the volume in hand. For volumes that don't
actually have real publication dates printed in, you need to
infer the date or determine a probable date of publication
and enter this in brackets with a question mark, e.g.
[1923?]. The date will be displayed as "Publication date"
for the item in Textual Collections as well as for sorting the
search results according to "date ascending", "date
descending". The date in this field will also be used for
statistical purposes.


Issue Node

Issue node is created as child of the parent volume node. Please note that for annuals or
serials that only have volumes, issues node is not required.


Node name


As with volumes, the node name for the issue should be
want to display as part of the title for the item in Textual
Collections. It is important that you record all node names
for volumes and issues consistently throughout the entire
serial run. Recommended best practice is to use either









"Number x" or "No. x" and to put the chronology in
parenthesis. For instance, "Number 1", "No. 1 (Spring)".

"The Florida Historical Quarterly volume 1 issue 1"


Type of part

Issue name


Use"issue".


This is required for a serial when applicable. It must be the
numeric designation of the issue. E.g. if this is issue 3, the
Issue name must be "3". If there isn't a numeric
designation, you have to establish one.

This is required for a serial when applicable. The date in
this field will be used to make a chronology tracing for the
date of publication of the original materials. Use the date of
publication for the issue in hand. For issues that don't
actually have real publication dates printed in, you need to
infer the date or determine a probable date of publication
and enter this in brackets with a question mark, e.g.
[1923?]. The date will be displayed as "Publication date"
for the item in Textual Collections as well as for sorting the
search results according to "date ascending", "date
descending". The date in this field will also be used for
statistical purposes. Note that the Date recorded for the
issue should be the same as the Date used for the volume.


Contents of Issues Nodes

Issues of serials generally contain material such as letters to the editor, articles,
advertisements, etc. Each article or other section should be a child node under the parent
node for the Issue.


Node name


The name of the node (Division name) should be what you
want to appear in the online table of contents for the issue
in Textual Collections. If the node is an article, you should
generally use the title of the article; however, if the title is
very long, then use a short form as the node name, and put
the complete title in Title. If the node is some section of
the issue that does not have a name, make up some
descriptive heading, e.g. "Letters".

Commonly used values: permission, cover, contents,
correspondence, article, abstract, advertisements, editorial,
index.


Type of part


Date









Title Title can usually be left blank. It's used only when a title is
very long. Anything entered in Title will be used instead
of the Node name in the table of contents displayed to the
user. Typing in a long title as the node name can make the
tree look messy and hard to read. Instead, type an
abbreviated form of the title as the node name, and type the
complete title in Title.

Creator If the Division node is for an article with one or more
named authors, you should enter the name of the author in
this element.


File level information

File nodes should be children nodes under the parent nodes for contents of issue
divisions. File nodes generally contain page images. There are several required elements
at the file level, but most of them can be supplied by the system automatically.

Type of Part Usually use "Page".

File name Automatically supplied when you browse and select the file
or use the drag method to create the file node.

It is recommended that if you have different file formats for
the same page image, they should use the same filename,
but with different file extension.

File format Automatically supplied by the system if it can get this from
the extension on the filename.

It is required that if you have different file formats for the
same page image, you have to create sibling file nodes for
them continuously. For example, if you have both
"cover.tif" and "cover.jpg" for the Cover page image, the
file nodes for "cover.tif" and "cover.jpg" should come in
pair consecutively.

File size Automatically supplied by the system when the information
is present in the image header.













,a F.I. ,i. ae e Tree ._.. .rT -,., 15 led___
- I'r ... .. I "-'- 'F .. .. .... J "_.
,. IF ,,,,I.- u, *-------_-------------,,------















I I .
-Type of part
I.. I .. .

,1 ., I ,, v








hiType o part IJ ....


Section name
Title
Dale
Creator
Rights


S


-.. ..j


- .' ,,,.. I


Silarn l a t". K r MI E.,j1 I j I r 1,rI I' T M F E -" ..r 'i l Il | o


An example of a serial showing many sections as contents.




Full Text

PAGE 1

MXF CLIENT MANUAL Florida Center for Library Automation rev. 11/04/04 1

PAGE 2

MXF CLIENT MANUAL ............................................................................................................................1 FLORIDA CENTER FOR LIBRARY AUTOMATION ..........................................................................1 REV. 11/04/04 ...............................................................................................................................................1 I. UNDERSTANDING THE MXF AND MXF CLIENT .........................................................................5 I.a. What the MXF is and how it is used ...................................................................................................5 I.b. Sending data to FCLA .........................................................................................................................6 I.c. The MXF Structure: Package, Entity, Division and File levels ...........................................................7 II. PREPARING DOCUMENTS ...............................................................................................................9 II.a. Document structure ............................................................................................................................9 II.b. Books .................................................................................................................................................9 II.c. Unitary visual materials (maps, posters, postcards, photographs) ....................................................10 map, poster, etc. [type of object, main ............................................................................................10 II.d. Other formats ....................................................................................................................................10 III. BASICS OF THE MXF CLIENT ......................................................................................................11 III. BASICS OF THE MXF CLIENT ......................................................................................................11 III.a. Logging on ......................................................................................................................................11 III.b. Opening the Metadata Module .......................................................................................................11 III.c. Areas of the screen ..........................................................................................................................11 III.d. The Tree Window ...........................................................................................................................12 III.d.1. Using the Tree: Nodes, Parents and Children ..........................................................................12 III.d.2. Using the Tree: Functions and Navigation ...............................................................................13 III.e. The Workform Window ..................................................................................................................14 III.e.1. Metadata Workforms ...............................................................................................................14 III.e.2. Repeatable elements ................................................................................................................15 IV. SETTING DEFAULTS .......................................................................................................................17 IV.a. When to use defaults .......................................................................................................................17 IV.b. How to set defaults .........................................................................................................................17 IV.c. Clearing, saving and changing defaults ..........................................................................................19 V. CREATING A NEW MXF PACKAGE .............................................................................................20 V.a. The Package control number ............................................................................................................20 V.b. Setting the New Package directory ..................................................................................................20 V.c. Setting the Image directory ............................................................................................................21 V.d. Package level Workform elements ..................................................................................................23 VI. DESCRIBING PUBLICATIONS: ENTITIES AND DIVISIONS ..................................................23 VI.a. Adding Entity nodes .......................................................................................................................23 VI.b. Adding Division nodes ...................................................................................................................25 VI.c. Using Insert Range .........................................................................................................................27 VII. DESCRIBING PHYSICAL FILES ..................................................................................................30 VII.a. Adding File nodes .........................................................................................................................30 VI.b. Creating Derived File nodes ...........................................................................................................33 VI.c. Copying Files ..................................................................................................................................34 VI.d. Viewing files ...................................................................................................................................35 2

PAGE 3

VI.e. File workform elements ...................................................................................................................36 VII. TREE MANIPULATION .................................................................................................................37 VII.a. Inserting nodes ..............................................................................................................................37 VII.b. Rearranging nodes ........................................................................................................................37 VIII. VALIDATING PACKAGES .........................................................................................................38 IX. OUTPUT TO XML .............................................................................................................................40 X: SPECIAL FORMATS: SERIALS .......................................................................................................41 X.a. Directory setup for serials ................................................................................................................41 X.b. Package level information ................................................................................................................42 X.c. Entity level information ...................................................................................................................42 X.d. Division level information ................................................................................................................43 X.d.1. Volumes. ....................................................................................................................................43 X.d.2. Issues .........................................................................................................................................44 X.d.3. Contents of issues ......................................................................................................................46 X.e. Serial title changes ............................................................................................................................48 XI. SPECIAL FORMATS: MAPS AND PHOTOS .................................................................................49 XI.a. Directory structure ...........................................................................................................................49 XI.b. Package level information ...............................................................................................................49 XI.c. Entity level information ..................................................................................................................50 XI.d. Division level information ...............................................................................................................50 XI.e. File level information ......................................................................................................................50 XI.f. Related Images .................................................................................................................................52 Metadata for Image Files ....................................................................................................................52 MARC record creation ........................................................................................................................52 New Elements for Related Images ......................................................................................................52 Image File Naming ..............................................................................................................................55 APPENDIX A: WORKFORM ELEMENTS ...........................................................................................56 PACKAGE LEVEL ..................................................................................................................................56 ENTITY LEVEL ......................................................................................................................................57 DIVISION LEVEL ...................................................................................................................................60 FILE LEVEL ............................................................................................................................................61 Display Row .........................................................................................................................................63 Display Column ....................................................................................................................................63 APPENDIX B: HOW THE MXF IS ENCODED ....................................................................................67 APPENDIX C: VALIDATION MESSAGES ...........................................................................................68 Appendix C.1. Error messages ................................................................................................................68 Appendix C.2. Warning messages ............................................................................................................68 APPENDIX D: OPTIMIZING WORKFLOW ........................................................................................70 a. Book, images exist ...............................................................................................................................70 APPENDIX E: TRAINING EXERCISES ................................................................................................71 Exercise 1: Book worksheet .....................................................................................................................71 Exercise 2: New package ..........................................................................................................................74 Exercise 3: Adding Entity and Division level information ......................................................................77 Exercise 4: Adding File level information and finishing the Package ......................................................79 APPENDIX F: MXF QUICK GUIDES FOR TEXTUAL MATERIALS .............................................80 3

PAGE 4

MXF: Quick Guide for Monograph ..........................................................................................................80 MXF: Quick Guide for Multipart .............................................................................................................85 MXF: Quick Guide for Serial ...................................................................................................................90 4

PAGE 5

I. Understanding the MXF and MXF Client I.a. What the MXF is and how it is used The MXF (Metadata eXchange Format) is a locally-developed format for communicating information about digital files representing library materials such as books, journals, maps, and photographs. When we speak of an MXF we really mean a file of metadata recorded in MXF format. Some of the information that can be recorded includes: bibliographic metadata -information describing the book, journal, etc., such as its title and author structural metadata -information about how different images relate to each other in a complex document, e.g., how pages are ordered to form chapters technical metadata -information about the digital file itself, such as its filename, and when and how it was created. The MXF can be used by a library to send metadata to FCLA, by FCLA to send metadata to a library, or by two libraries to exchange metadata with each other. When used to send metadata to FCLA, the MXF can include instructions to FCLA on how to process the metadata and related files. 5

PAGE 6

I.b. Sending data to FCLA The MXF should be sent to FCLA along with the files it describes. For example, if a library is sending PDF images of a book to FCLA for loading into Florida Heritage, an MXF and the PDF files it describes should be sent to FCLA together. Generally this means that the MXF itself and the files it describes should be placed in the same directory and FTPed as a single folder. However, note that files should only be sent to FCLA if there is a reason to do so. For example, imagine that a library created TIFF masters, MrSid derivatives, and JPEG thumbnails for maps. The MrSid and JPEG images were intended to be loaded in the Visual Materials Collections server, and displayed to users. The TIFF masters were intended to be burned onto CD by the library itself, and stored locally as archival copies. In this case, even though for each map there are three image formats, only the MrSid and JPEG formats would have to be sent to FCLA and described in an MXF. On the other hand, if there were an agreement between the library and FCLA that FCLA would archive the TIFF or burn the TIFF onto a CD and send that back to the library, then the TIFF image would need to be sent to FCLA and described in the MXF. 6

PAGE 7

I.c. The MXF Structure: Package, Entity, Division and File levels The MXF is structured in four logical parts called levels : Package, Entity, Division and File. When an MXF and the digital files it describes are sent to FCLA (or from FCLA to another library) the MXF itself and all the files transmitted together are called a package. A package usually contains information about a single bibliographic entity but if necessary multiple entities can be bundled up in a single package. For example, generally a package will contain a single book, but there is no reason (apart from the large size) that two books could not be sent together in a package. It might make sense to bundle several short items together, such as the metadata and image files for several maps. The Package level contains information needed to process a single package. The type of information contained at the Package level includes: whether the package is new or a replacement for one sent earlier; who is sending the package and should be contacted if there are problems; specific processing instructions for the package, for example, whether to create derivative PDF files from images sent in the package, or whether to create MARC records from bibliographic data in the MXF; what file formats are included in the package; when the package was created. The Entity level describes a single bibliographic entity, e.g. a single book, map, or photograph. For a serial, the bibliographic entity is the serial title, even if only one issue of the serial is being sent. The Entity level generally corresponds to what would be cataloged in the library catalog, even if a catalog record does not exist for the particular entity. The type of information indicated at the Entity level includes: the bibliographic description of the entity; what type of item it is (e.g. book, serial, map, photo...); what project(s) or collection(s) it belongs to (e.g. Florida Heritage, Literature for Children, etc.); a copyright statement; a statement of who created the digitized version. The Division level describes a logical structural division of the entity. For example, a book may consist of a front cover, front matter, several chapters and a back cover. Each of these would be described by a set of data elements at the Division level. Because the structure of a publication is often hierarchical, Division level information can be nested. For example, if the entity is a serial title, there may be Division level information for articles nested within Division level information for issues nested within Division level information for volumes. Information that can be given at the Division level includes: 7

PAGE 8

what the structural subdivision is (e.g. a chapter, issue, article...) what it is called (e.g. Chapter IV) a bibliographic description of the subdivision The File level pertains to digital files. For every file included with a package, there must be a description at the File level. The File level can include extensive information about the physical file, including but not limited to: the filename and file type when it was created and by what institution and individual; how it was created; what device (e.g. digital camera or scanner) created the file, and what device settings were used; for images, what the image characteristics are (e.g. bit depth, resolution, color space, descreening, etc.) If a file was derived from a source file, then information about the source can also be given. 8

PAGE 9

II. Preparing documents II.a. Document structure Before you begin creating MXF data, you must understand the structure of the document you are dealing with, and how it relates to the structure of the MXF. You will need to know what the bibliographic Entity is and what Divisions exist within the Entity. For each Division, you need to know what to call it, and what type of part it is. If the document is a book, serial, or other textual work that will be displayed by the FCLA document server, the name of the Division is what will appear in the online table of contents that is displayed to the user. For example, if the Division is a chapter, the name of the Division might be something like Chapter I or Chapter 1: Tallahassee or Tallahassee. The name should generally be what appears in the printed table of contents of the work, although some Divisions like covers and other front matter have no equivalent in the printed TOC. The type of part must be taken from a list. It is used primarily for statistical purposes. It includes such structural parts as: chapter, page, section, abstract, advertisement, volume, issue, article, part, main, section, suppl, cover, fonds, recordgrp, series, subgrp, subseries, box, file, item. When creating the MXF, you should either have a work in hand, or else have a worksheet that lists all structural divisions of the work. The worksheet should list, for each Division, the name, type of part, and range of pages included in that Division. II.b. Books For a book, the Entity level will describe the book itself. Here are some suggestions on how to treat typical book divisions: division Division name Type of Part permission* Copyright permission permission front cover Front cover or cover Front matter inside of front cover Front matter cover other preliminary matter, Front matter section e.g. blank page table of contents Table of Contents or contents Contents foreword Foreword section preface Preface section 9

PAGE 10

chapters Chapter X chapter parts Part X chapter index Index index back matter Back matter section inside of back cover Back matter cover back cover Back cover or cover Back matter when a book is under copyright and digitized with permission, a statement describing the copyright holder and the permissions granted should be included as the first digital file for the entity. II.c. Unitary visual materials (maps, posters, postcards, photographs) Most visual materials are unitary, that is, they have only one part, the map, poster, photograph etc. itself. In this case the entity level describes the visual material, and there is only one division. division Division name Type of Part map, poster, etc. [type of object, main e.g. map, poster, etc.] II.d. Other formats For how to treat divisions of other formats, see the chapters of this manual on Special Formats (e.g., Special Formats: Serials). 10

PAGE 11

III. Basics of the MXF Client III.a. Logging on You need a username and password to log onto the MXF Client. If you dont know your name and password, contact your site administrator. III.b. Opening the Metadata Module When you invoke the MXF Client, you will initially see a blank screen with a menu bar at the top. One of the options listed on the menu bar will be Module. Under Module there are three sub-options: Administration, Metadata and Scanning. Select Metadata. III.c. Areas of the screen The MXF Client screen is divided into 3 areas. [Areas]. 11

PAGE 12

The Tree window is on the upper left. This displays all Package, Entity, Division and File level information you have created as nodes in a hierarchical display called a tree. Whichever node is highlighted in the tree is the current node. In this example, Copyright permission is the current node. The Directory window is on the upper right. The top of this window displays the current directory, and the lower portion lists any image files (.tif, .jpg, .sid) found in that directory. The Workform window is at the bottom of the screen. This window has a data entry area for the metadata elements associated with the current node, for example, Type of part. III.d. The Tree Window III.d.1. Using the Tree: Nodes, Parents and Children Each of the four levels of the MXF are represented by nodes on the tree. [Tree] There are four different kinds of nodes, each represented by different icons. 12

PAGE 13

Envelope icon = Package node. This is always the top-most, left-most node. In this example, its name is MXF. Book icon = Entity node. In this example, FA00000033 is the Entity node. Folder icon = Division node. In this example, Copyright permission, Front Matter, etc. are Division nodes. Document icon = File node. In this example, Page 1 is a File node. (Derived files have a slightly different icon. See Creating Derived File nodes below.) The Workform window at the bottom always displays a metadata workform for the current node. The current node is the node that is highlighted on the tree. There are Parent and Child nodes on the tree. A Parent is hierarchically higher. On the tree it appears to the left and above its Child nodes. A Child node appears below and to the right of its parent. In this example, the Division node Chapter 1 is the parent of one Child node named Page 1. III.d.2. Using the Tree: Functions and Navigation When a node has Child nodes, there is a plus or minus before the node on the tree. A plus means that Child nodes are not shown. If you click the plus sign the node will be expanded to show Child nodes. A minus means that all Child nodes are displayed. If you click the minus, the Child nodes will be rolled up (not displayed) under the Parent node. The arrow keys move you up and down the tree: up arrow = Takes you to the node immediately above the current node down arrow = Takes you to the node immediately below the current node left arrow = Takes you up to the parent of the current node. (The current node in our example is the node named Foreword; the left arrow would make the node named FA00000033 current.) right arrow = Expands any plus nodes to show the Child nodes beneath. (Also, in Windows 2000 systems, the right arrow will move you down the tree.) The menu bar Tree option has many commands for manipulating nodes on the tree. Among other things, you can add, remove, copy, and rename nodes. The most common actions have shortcut key combinations consisting of holding the CTRL key while pressing one other key. For example, CTRL-a has the same function as Tree/Add node. 13

PAGE 14

III.e. The Workform Window III.e.1. Metadata Workforms Each type of node has a different Workform, listing the data elements that pertain to that kind of node. The metadata Workform for the current node will appear in the workform window. In the example above (Section 3.d, Tree), the File node Page 1 is the current node, so the workform for a File node displays. Every workform has three areas. On the far left a label for the data element displays, e.g. Type of part, Page name. The label color tells you if a value for the element is required: black = optional red = required purple = conditionally required grey = supplied by the system Purple labels always come in pairs; if one purple element is supplied, the other is required. For example, for an Entity node, Bib record key and Bib record system are purple. Both of these values can be left blank; however, if one is filled in, then the other must be also. A grey label means the value of the element will be supplied by the system if possible. If the value is also grey, that means that you cannot change that value. For example, for a File node, the element Checksum (label and value both grey) is calculated by the system and cannot be changed. The element File format (label is grey, value is black) is calculated by the system but the value can be overtyped by an operator. In the center of the workform there is a box for data entry. Values can be supplied in four ways: 1) by typing the value in the box 2) by selecting a value from a pulldown menu (down arrow on the right side of the box) 3) by setting a default (See Setting Defaults below) 4) by being automatically calculated by the Client program. Values which are automatically calculated by the Client and cannot be changed are shown in grey. On the right there are buttons for performing certain functions. browse = when the value of an element is a filename or directory name, you can browse through the directories known to your computer to find the name add new = if the element is repeatable, you can click this to add another instance of the element 14

PAGE 15

remove = you can click this to remove the element At the very far right is a scroll bar for moving through a long workform. III.e.2. Repeatable elements If an element is repeatable, an Add new button will display to the right of the data entry box. Element values can be repeated in two ways: 1) use Add new to add another data entry box for the element [Add new] 2) type in multiple values, separated by commas [Commas] 15

PAGE 16

In the examples above, multiple values are entered for the data element Project code(s) in two different ways. These are equivalent to each other; neither method is better than the other. Use the method you prefer. 16

PAGE 17

IV. Setting defaults IV.a. When to use defaults In general, before you create a new package, you will want to set defaults, or make sure that the defaults that are set are correct. Each user has his own set of defaults that persist until s/he changes them. Even if several people share a single workstation, each of them has his/her own defaults. The values you supply when you set defaults will appear automatically on the metadata Workforms you use to enter data for a package. You can always type over a default value with some other value. Generally you want to default values which will not change from package to package, such as the contributing institution. You may also want to default the values that most frequently occur. For example, a book may have three types of divisions: front matter, chapters and back matter. However, if most of the divisions will be chapters, you may want to set the default type of division at the Division level to chapter. Although this will have to be overridden some of the time, it will be correct most of the time. IV.b. How to set defaults To set defaults, select File/Defaults. A tree will appear in the Tree window, showing Package, Entity, Division and File nodes. Select (highlight) a node from the tree to set defaults for data elements at that level. 17

PAGE 18

In the example above, Package was selected on the tree, and defaultable elements at the Package level are shown in the Workform window. In this case the operator has set New Package? to default to yes and Contributing institution to FAU, which would make sense for an FAU operator who mostly creates new packages. Defaults for some processing instructions (make derivatives, target server and formats included) are also set. Remember to use the scroll bar at the right side of the Workform window to see all defaultable elements. In the next example, Entity was selected on the tree, and defaultable elements at the Entity level are shown in the Workform window. 18

PAGE 19

Most operators will probably want to default their Source institution and the Version statement (which will only change once a year). If most work is being done for a particular type of material and project, then defaulting Entity type and Project code(s) also makes sense. It would not make sense to default elements such as Bib record key (the key of the NOTIS record cataloging the entity) as these would change from one entity to another. It is, however, a good practice to default "Bib record system." Since this is conditionally required with "Bib record key," it will help the operator remember to include the "Bib record key." IV.c. Clearing, saving and changing defaults To clear a default, simply blank out its value on the Workform. To save defaults, select File/Close defaults. They will be automatically saved. To change a default, just go into default mode by selecting File/Defaults, and change the value on the workform. 19

PAGE 20

V. Creating a new MXF package V.a. The Package control number Every package must have a unique control number, called the package control number or sometimes, the package id. For packages to be contributed to FCLA, this number must begin with a 2-character institution or project code assigned to you by FCLA. The standard format is the 2-character code followed by a unique 8-character number, e.g. UF00001234. However, other formats may be accepted for certain projects. Before you create a new package, you must decide what the package control number will be. The package control number is meant to identify a transmission (that is, a bunch of stuff FTPed together at some point in time). The package itself only exists for the length of time it takes to be created, transmitted, and processed. After a package control number is used once, it should not be used again, unless the entire package is re-transmitted for some reason (for example, if there was an error in the original package that prevented processing). V.b. Setting the New Package directory Every user should have his or her own directory for doing work with the MXF client. In our example, the user is Priscilla and the folder (directory) for MXF work is called Digitized books. Many MXF packages can be stored within the folder Digitized books at the same time. Ordinarily, for a monograph or unitary work, each MXF package and the images it describes should be stored in the same folder to make it easier to FTP the entire package at one time. (To set up directories for a serial, see section X. Serials Issues, below.) Best practice is to give the folder the same name as the package control number. So, for example, if the package id is going to be FA00000033, then both the MXF file and the image files should be created in the folder named FA00000033. In the example here, the directory structure is: /Priscilla /Digitized books /FA00000033 To create a new MXF, select Package/New. A popup window will appear asking you where to put the new package [Package new]. 20

PAGE 21

Select the appropriate directory in the Save in box at the top of the popup window. Use the down arrow at the end of the box to navigate through the directories on your computer. To the right of the box, there are a set of buttons. The first button moves up one level, the second button creates a new folder (directory). If the appropriate directory (in this case FA00000133) does not already exist, use the new folder button to create it. In this example, a new package named FA00000133 is being created in the directory FA00000133. V.c. Setting the Image directory When the new package has been created, a Package level node will appear at the top of the Tree window [New Package]. 21

PAGE 22

A metadata Workform window for Package level information appears at the bottom of the screen. Note that defaults appear for some elements. The Directory window at the upper right of the screen should show the directory that contains the images to be included in the package (assuming these images have already been created). When you create a new package, the directory containing the MXF package is automatically highlighted in the Directory window, and the name of that directory appears as the value of the Image directory element in the workform. If the image files to be included in the package are in a different directory from the MXF package, you will have to change the directory listed. Any image files in the directory will appear just below the directory listing. NOTE: If, after creating a package, you move the package and associated images to a different location, you need to re-open the package and change the Image directory. If image files have not been created yet, it does not matter which directory is highlighted in the Directory window. 22

PAGE 23

V.d. Package level Workform elements Package level Workform elements tell FCLA how to process what you have sent. See Appendix A for explanation of element meanings and use. See also instructions for specific formats of materials, such as serials. VI. Describing publications: Entities and Divisions VI.a. Adding Entity nodes Every package must have at least one Entity level node. To add the node, make sure the Package node is highlighted on the tree, and select Tree/Add node [Tree/Add node]. A popup window will prompt you to add the node name [Add entity]. 23

PAGE 24

The Client assumes that any node added to the Package level will be an Entity node inserted as a Child to the Package Node, which is almost always the case. The name of the node given in the popup is also called the Entity ID. For a work completely contained within one package (for example, a monograph all chapters of which are being sent at the same time), the Entity ID should be the same as the package control number. For a journal or other serial which may be sent over time in a number of packages (for example, volumes 1 and 2 in one package, volumes 3 and 4 in another), the Entity ID should be the characters SN followed by the 8-character ISSN of the serial (e.g. SN00213392). This allows all parts of the work to be brought together in the system, even if the parts are sent separately. In this example, since we are doing a complete book, the node name is the same as the package ID, FA00000033. If you made a mistake typing the node name, go to the Tree window and right click on the node icon, then select Rename. An Entity level workform appears in the Workform window, where you can now enter entity level information [Entity]. 24

PAGE 25

Note that there are two ways to supply a bibliographic description of the title: 1) If a NOTIS record exists, you can supply the NOTIS record key in Bib record key and FCLANOTIS:QF in Bib record system. This associates this package with the external bibliographic record. 2) If no NOTIS record exists, you can supply a complete Dublin Core description by filling in the elements Title, Creator, Contributor etc. In both cases, you must supply the element Title at this level. This is used by FCLA staff to make sure the correct volume is being processed. See Appendix A for explanation of Entity level element meanings and their use. VI.b. Adding Division nodes Only Division nodes display in the Table of Contents for an Entity. The rule of thumb is therefore to create a Division node for anything that should display in the TOC. Sometimes people wonder, if a section is only one page, if they need to create a Division 25

PAGE 26

node for it or if they can simply add a File node in the appropriate place. Either method is acceptable. However, if you want it to display in the TOC, you must create a Division. To create a Division, make sure that the parent node of the Division is highlighted in the Tree window and select Tree/Add node or use CTRL-a to add a node. Note that if a Division node is current, the Client assumes that the next node added will be a Division at the same level. If you are adding a File node or a child Division, you will need to change the default node type. [Add node div] For each Division, there are four data elements that you need to know how to use: Node name Type of part [type of part] name Title You are prompted to enter the node name in the box that appears when you request Add node. This node name labels the node in the tree, and is also what will ordinarily appear in the online table of contents listing for the work. For example, if the Division is a chapter, the node name for the division might be something like Chapter I or Chapter 1: Tallahassee or Tallahassee. The node name should generally be what appears in 26

PAGE 27

the printed table of contents of the work (unless it is a kind of division that doesnt ordinarily appear in a printed table of contents, or if the name is very long, as will be explained below). In this example the node name is Copyright permission. Once the Division node is created, the workform for Division level information will appear at the bottom of the screen. In general, the only data element that needs to be supplied at the Division level is Type of part, which is used primarily for statistical purposes. The value for this element should be taken from the pull-down menu. Values include such structural parts as: chapter, page, section, abstract, advertisement, volume, issue, article, part, main, section, suppl, cover, fonds, recordgrp, series, subgrp, subseries, box, file, item. In the metadata workform, the label [type of part] name will vary depending on what you enter for Type of part. For example, if you say that the type of part is chapter, then the next element on the metadata workform will be labeled Chapter name. If the Type of part is section, then the next element will be labeled Section name. In general, you do not have to provide a value for this element. However, if the type of part is chapter or page or some other unit that is numbered (e.g. Chapter 1, Chapter 2, Chapter 3...), the number or name of the part exactly as it appears on the piece can optionally be given in Part name. A value is required, however, if the Type of part is volume or issue, in which case the Volume name should be the number of the volume, and the Issue name should be the number of the issue. See Describing Serials below. The element Title can usually be left blank. Anything entered in Title will be used instead of the Node name in the table of contents displayed to the user. This is useful when a title is very long. Typing in a long title as the node name can make the tree look messy and hard to read. Instead, type an abbreviated form of the title as the node name, and type the complete title in Title. Additional bibliographic description can optionally be provided. For example, if the Division is an article in a journal, then the author of the article might be supplied in Creator. See Appendix A for full explanation of Division level element meanings and their use. VI.c. Using Insert Range The Tree/Insert range function can be used whenever you have multiple Entity, Division or File nodes whose names increment sequentially. (Insert range for File nodes is described in Describing physical files below.) For example, if you have a book with chapters numbered Chapter 1 through Chapter 10, you can use Insert range to add ten Divisions for the ten chapters. 27

PAGE 28

Highlight the parent of the nodes you want to add. (That is, if you are adding a range of Entity nodes, highlight the Package node on the tree. If you are adding a range of Division nodes, highlight the Entity node on the tree.) Select Tree/Insert range [Insert range]. A popup box will prompt you for information about the nodes to be added. In this example, we have a book with 10 chapters titled Chapter 1, Chapter 2, etc., so we filled in Enter range name with Chapter and filled in Enter range values ... to indicate we want the parts to be numbered from 1 to 10. We selected chapter as the Type of Part. If other bibliographic data was constant for all chapters, the information could be entered on this workform. When the popup workform is filled out, click OK. The range of nodes will be added as the last set of children under the highlighted parent node. [Range added] 28

PAGE 29

The normal workform for the first added node will be displayed. You can now add information unique to this chapter if any. 29

PAGE 30

VII. Describing physical files VII.a. Adding File nodes There are three ways to add a File node: 1. Drag method. If images already exist in the directory, then the best way to add file nodes is to drag the images from the Directory window to the Tree window. After creating the Division in which the image belongs, right click on the image in the Directory window and drag it directly onto the Division node in the Tree window. [Drag] A subordinate node for the file will appear, and File level metadata will appear in the Workform window. The file name, file format, and file creation date and time will automatically be supplied to the workform. [Drag done] To drag multiple files at the same time, click on the first file, press and hold CTRL, and click on the other file(s). Continue holding the left mouse key down after selecting the last file, and drag the group. 30

PAGE 31

2. Node method. If you are creating the metadata before you have the images, then you can create a File node by highlighting the Division and selecting Tree/Add node. By default, the Client will assume you are adding another Division; change the default node type from div to file. 3. Range method. If you are creating the metadata before you have the images, there will be an image for each page, and the images will be predictably named with the page number, the easiest way to create the File nodes is to use the Insert range function. [Insert page range] 31

PAGE 32

Highlight the node for the chapter or other Division to which the pages belong. In this example we are inserting page images for Chapter 1. Select Tree/Insert range and a popup window will appear. For Enter range name we enter Page so the pages will be named Page 1, Page 2, etc. There are six pages in chapter 1 so we enter the range values from 1 to 6. We select range type file so these will be file nodes, and the Type of part is page. The program supplies a default directory for the element File name, and a default file type. In this example the directory is m:\user\Priscilla\DigitizedBooks\FA00000033\ and the file type is image/tiff. If the box Supply Filename is checked, when the Range function is enacted, the actual file names will be supplied automatically using the directory name, the range values, and the file type. So, for example, the first File node created will be given the filename m:\user\Priscilla\DigitizedBooks\FA00000033\1.tif. When all the information has been entered, click OK. The range will be automatically created. The Workform for the first of the added nodes will display, so you can add any information necessary. [Page range added] 32

PAGE 33

VI.b. Creating Derived File nodes If you have a file derived from another file in the same package, for example, a JPEG image derived from a TIFF image, you can create a Derived File node for the derived file. A Derived node automatically records information about the source file that can be useful in the archival management of the derived file. If you do not plan to manage the derived file archivally, there is no need to create a Derived node. To create a Derived node, highlight the node for the source file. In this example it is the copyright node for the TIFF image. Select Tree/Derive node. A new node with the same name as the source node will appear. [Derive] 33

PAGE 34

The program assumes the derived file has the same filename, but a different file type, as the source file. Your cursor will be in the metadata workform for the new node, at the end of the default filename, so you can add the new file type. VI.c. Copying Files Sometimes a file needs to appear in more than one place in the Tree. For example, a new chapter may begin on the same page that the previous chapter ended on, so the file for that page needs to appear under two chapter Divisions. In this case you have two options: you can create a new File node for the page by either using the Add node function or dragging the file again from the Directory window; you can use the Copy node and Paste Node functions to copy the first file node to the second location. To use copy and paste, highlight the node to be copied and select Tree/Copy Node (CTRL-N). Then highlight the node under which you want to copy to appear, and select Tree/Paste Node (CTRL-P). The new node will appear with the same name as the copied node, followed by (copy). [Copy node] 34

PAGE 35

When you copy a File node all the data entered on the File workform for the first node automatically carries over to the copy, so you wont have to re-enter any information. This can be an advantage over using the Add Node method. It is an advantage over dragging the file node twice only if you supplied non-default information after dragging. VI.d. Viewing files Sometimes it can be helpful to view a file to see what page or section it represents. To view an image file, highlight the file name in the Directory window, and use View/Image. The image will appear in a popup window, which can be resized using the mouse. You can also highlight the file name and use CNTL-I to view the image. The 35

PAGE 36

popup window will also allow you to move from one image to another, using back and forward arrows. VI.e. File workform elements The file workform has several required elements but most of them can be defaulted or supplied by the system. Type of part at the File level can generally be defaulted using File/Defaults. For example, if you are sending a book of page images as TIFF files and their derivative JPEGs, it would make sense to default Type of part at the File level to page. File name is automatically supplied when you use the drag method to create the node. File format will be supplied by the program if it can get this from the extension on the filename. File size will be supplied by program when the image is present in the Image Directory and the information is present in the image header. See Appendix A for full explanation of File level element meanings and their use. 36

PAGE 37

VII. Tree manipulation VII.a. Inserting nodes The Add Node function will, by default, add the new node as a child to whatever node is highlighted as the current node on the tree. It is possible to insert nodes in other places by changing the default position from child on the Add Node popup box. *Current node *default position of add node Previous will add a node directly above and at the same level as the current node. *Position of added node *Current node Next will add a node below and at the same level as the current node. The new node will be placed after any children of the current node. *Current node *child of current node *Position of added node Last will add a node as the last node in the tree at the same level of the current node. *Current node *Node *Node *Position of added node VII.b. Rearranging nodes Nodes can be moved from one place in the tree to another by dragging and dropping the node to be moved onto another node. The node will be added as the previous peer of the node on which it is dropped. For example, say that by mistake you have inserted nodes out of order, such that the tree is ordered as follows: *Parent *Child 2 *Child 3 *Child 1 37

PAGE 38

You can drag and drop the node Child 1 up to Child 2. The node will be inserted before Child 2 at the same level and the nodes will now be in order: *Parent *Child 1 *Child 2 *Child 3 VIII. VALIDATING PACKAGES When you are finished creating a package, you should check it using the Package/Validate function. This will tell you how many errors and possible errors (warnings) there are. [Validate] When you clear the popup box by pressing Enter or clicking OK, a table of errors will display. [Error table]. 38

PAGE 39

Although the message in the message column may be truncated, you can display the full message by either double-clicking the error line (the full message will appear in a pop-up box) or by holding the cursor over the error message (the full message will display as a mouse-over). If the error concerns bad data in the metadata Workform, you can see where the error occurred by double clicking the error line and clearing the pop-up error message. The affected node will be highlighted in the Tree window, and the cursor will be placed in the data entry box of the Workform that contains the bad data. In order that you can see this, the table of errors is minimized. You can bring it back to the foreground in the normal way, by clicking its icon in the task bar You can initiate validation at any time during package creation. FCLA recommends that you validate periodically, after adding 10-15 page files. This step performs the function of calculating the checksum values for each image, which can take some time, depending on the size of the image file and the power of the computer. By validating frequently, final validation goes more quickly. Validation is automatically performed by the system when XML output is requested. Errors must be fixed before the MXF package can be output as XML. Warnings will not prevent the output of XML. For a list of error and warning messages and their meanings, see APPENDIX C. 39

PAGE 40

IX. OUTPUT TO XML When you finish creating your package, you need to output it as XML before sending it to FCLA or another library. Select Package/Output to XML. First the MXF Client will automatically Validate the data (see Validating Packages above). If there are errors, they must be fixed, and Output to XML requested again. If there are warnings, you can either correct them or simply click OK. The MXF Client will then create an XML file and display popup message box telling you the name of the XML file created. If on the View menu, XML Tree on Output is checked, the Client will also open a new window showing the XML in tree format. Like the Package tree, the XML tree can be manipulated to expand and contract nodes by clicking the plus and minus boxes in front of the node. You can get rid of the tree by closing the window. XML trees cannot be saved, but can be recreated by redoing Output to XML. The XML output file is saved by default in the Package directory. To view the XML file, select View/XML File. You can also view the XML from outside the Client using Notepad or a similar application. Never edit an XML output file directly. 40

PAGE 41

X: SPECIAL FORMATS: SERIALS The MXF allows flexible handling of serials, which may have many levels of hierarchy (e.g. volumes, issues, articles). However, for serials to be processed correctly, it is important to understand and follow special rules and conventions. If many issues of a serial are to be digitized, it is required that each issue has to be sent to FCLA in a separate package. There is no requirement that issues be received in order. The important thing to remember is that the node name of the Entity (Entity ID) is what ties all information for a single serial run together. If a serial is sent in several packages, the Package ID must be unique for each package, but the Entity ID must be the same. X.a. Directory setup for serials When you set up directories for your serial package, one and only one subdirectory should be created for each serial issue: \UF80001549 [package ID] \SN00130098 [entity ID] \1_1 [directory for images for volume 1 number 1] Please use only numeric characters for the name of the subdirectory for each serial issue. If the name of the subdirectory contains non-numeric character, the package will be rejected. Examples of valid directory setup: \UF80001549 [package ID] \SN00130098 [entity ID] \1 [directory for images for volume 1 with only one issue] \UF80001549 [package ID] \SN00130098 [entity ID] \1998_004 [directory for images for volume 1998 number 4] \UF80001549 [package ID] \SN00130098 [entity ID] \123-124 [directory for a combined issue number 123 and 124] Examples of invalid directory setup: \UF80001549 [package ID] \SN00130098 [entity ID] \vol1_no1 [directory for images for volume 1 issue number 1] 41

PAGE 42

\UF80001549 [package ID] \SN00130098 [entity ID] \vol3 [directory for images for volume 3] X.b. Package level information Create the package in the highest level directory you are using for the serial. In the example above, you would create the package in the directory UF8001549. Package ID (Node name). When you create the package node, the Package ID (node name) should be in the standard package ID format, consisting of the 2-character institution or project code followed by an 8-character numeric value, e.g. UF80013549. Image directory. All images being sent in the package should be in a single directory (that is, the subdirectory for each serial issue). Set the Image directory to that directory. In the example above this would be ...\SN00130098\1_1. Make records. For some serials you may want author/title bibliographic records created for each article published. If this is the case, select analytics, article as the value of Make records. Make derivatives. For some serials, you may want a PDF file created for every article in the serial. For other serials, you may want to create a PDF for each issue instead. Select pdf/article or pdf/issue if you want either of these options. Please note that you have to use sensible Type of Part for division nodes that you want to create or wrap PDF files. If you select pdf/article, the following Type of Part nodes will be expected to appear: article, advertisement, contents, correspondence, cover, editorial, index, section, permission. If you select pdf/issue, nodes with the following Type of Part will be expected to appear: issue, permission. In other words, even though you select pdf/article or pdf/issue for Make derivatives, but if you dont use any of these Types of Part for any nodes, no PDF file will be created. X.c. Entity level information Entity ID (Node name). The node name of a serial entity must be the characters SN followed by the 8-character ISSN of the serial. For example, if the ISSN is 0123-4567, then the Entity ID (node name) for a UF serial must be SN01234567. That way, even if volumes or issues are sent in multiple packages, there is a unifying identifier that ties all the pieces together as parts of the same serial. Some older serial titles do not have ISSNs or some materials you want to package as serial are not serials In this case there are two options: apply for a retrospective ISSN, or 42

PAGE 43

assign a pseudo-ISSN. Instructions on how to apply for a retrospective ISSN, or assign a pseudo-ISSN are available at http://palmm.fcla.edu/strucmeta/pseudoissn.html Entity type. Must be serial. X.d. Division level information Most serials will have at least three levels of hierarchy at the Division level: volume, issue (number), and the contents of the issue (articles, correspondence, editorials, etc.). This is represented on the Tree as parent and child nodes: the Division for the volume is the parent of the Division for the issue, which is the parent for the Divisions for the articles within the issue. X.d.1. Volumes. Node name. The node name for the volume should be what you want to display in the online table of contents. In other words, if you would like the table of contents to display as: Volume 1 Volume 2 Volume 3 then you must record the node names for the volumes as Volume 1, Volume 2, Volume 3 etc. If you want the chronology of the volume to display as part of the table of contents, then you must include that in the node name as well: Vol. 1 (1898/1899) Vol. 2 (1899/1900) It is important that you record all node names for volumes and issues consistently throughout the entire serial run. Recommended practice is to use either Volume x (year) or Vol. x (year) as this gives the most information to the reader. Type of part. The Type of part for a volume must be volume. Volume name. This is required for a serial. It must be the numeric designation of the volume. E.g. if this is volume 43, the Volume name must be Date. This is required for a serial. The date in this field will be used to make a chronology tracing for the date of publication of the original materials. Use the date of publication for the volume in hand. For volumes that don't actually have real publication 43

PAGE 44

dates printed in, you need to infer the date or determine a probable date of publication and enter this in brackets with a question mark, e.g. [1923?]. The date will be displayed as Publication date for the item in Textual Collections as well as for sorting the search results according to date ascending, date descending. The date in this field will also be used for statistical purposes. X.d.2. Issues Issue nodes are created as children of the parent volume node. Please note that for annuals or serials that only have volumes, issues nodes are not required. Node name. As with volumes, the node name for the issue should be what you want to display in the online table of contents. In other words, if you would like the table of contents to display as: Number 1 Number 2 Number 3 44

PAGE 45

then you must record the node names for the issues as Number 1, Number 2, Number 3 etc. If you want the chronology of the issue to display as part of the table of contents, then you must include that in the node name as well: No. 1 (Spring) No. 2 (Summer) It is important that you record all node names for volumes and issues consistently throughout the entire serial run. Recommended best practice is to use either Number x or No. x and to put the chronology in parenthesis. Type of part. The Type of part for an issue must be issue. Issue name. This is required for a serial when applicable. It must be the alphanumeric designation of the issue. E.g. if this is issue 3, the Issue name must be Date. This is required for a serial when applicable. The date in this field will be used to make a chronology tracing for the date of publication of the original materials. Use the date of publication for the issue in hand. For issues that don't actually have real publication dates printed in, you need to infer the date or determine a probable date of publication and enter this in brackets with a question mark, e.g. [1923?]. The date will be displayed as Publication date for the item in Textual Collections as well as for sorting the search results according to date ascending, date descending. The date in this field will also be used for statistical purposes. Note that the Date recorded for the issue should be the same as the Date used for the volume. [Issue] 45

PAGE 46

X.d.3. Contents of issues Issues of serials generally contain material such as letters to the editor, articles, advertisements, etc. Each article or other section should be a child node under the parent node for the Issue. Node name and Type of part. The name of the node (Division name) should be what you want to appear in the online table of contents for the issue. If the node is an article, you should generally use the title of the article; however, if the title is very long, then use a short form as the node name, and put the complete title in Title. If the node is some section of the issue that does not have a name, make up some descriptive heading, e.g. Letters. Here are suggestions on how to treat journal divisions: division Division name Type of Part permission* Copyright permission permission volume Volume X volume issue Number X issue front cover Front cover or cover Front matter 46

PAGE 47

inside of front cover Front matter cover table of contents Table of Contents or contents Contents letters to the editor [name of letters section] correspondence article [name of article] article abstract (within article) Abstract abstract advertisements Advertisements advertisements editorial Editorial editorial index Index index inside of back cover Back matter cover back cover Back cover or cover Back matter 47

PAGE 48

Date. You do not need to supply a date here if one was provided at the issue level. Creator. If the Division node is for an article with one or more named authors, you should enter the name of the author in this element. Repeat the element for multiple authors. An example of a serial showing many sections as contents is given below. [Serial] X.e. Serial title changes For a serial with title changes, a unique Entity ID should be created when the title changes. The Entity ID is a unifying identifier that ties all the pieces under each title together. Alternative title. This field is used to trace preceding title(s) or/and succeeding title(s) for serials with title changes. You must record the preceding title in the form of Preceding title: preceding title, and the succeeding title as Succeeding title: succeeding title, e.g. Preceding title: Report of the Board of Control of the State Institutions of Higher Learning of Florida for the biennium ending ...., Succeeding title: Report of Florida Board of Regents. 48

PAGE 49

XI. SPECIAL FORMATS: MAPS AND PHOTOS Single sheet maps and photos can be added to the FCLA Document Loader as single page documents, or can be added to the FCLA Visual Materials Collections server. If a map is destined for a Visual Materials Collection, there must be two images of the map supplied: a MrSid (.sid) vector image, and a JPEG thumbnail. Data entry for maps is quick and easy, as most of the required Workform information can be defaulted. XI.a. Directory structure Each single-sheet map or photo is treated as an Entity. A package can contain a single map/photo entity, or multiple map/photo entities. If your package contains a single map/photo entity, the directory structure can be very simple: a single directory to hold the package and images. \UF00005678 [packageID -should be the same as the Entity ID] If the package contains multiple map entities, then a directory should be created with the name of the Package ID, and subdirectories for each map should be named for the map Entity IDs. \UF800000019 [packageID] \UF00005678 [directory for first map] \UF00005679 [directory for second map] XI.b. Package level information Package ID (Node name). When you create the package node, the Package ID (node name) should be in the standard package ID format, consisting of the 2-character institution or project code followed by an 8-character numeric value, e.g. UF80013549. Image directory. If all images being sent in the package are in a single directory (that is, if you are sending a single map at a time) set the Image directory to that directory. If multiple maps are being sent, set the Image directory to the highest level directory. Make records. If maps are to be processed into a Visual Materials Collection, Make records should specify map. Make derivatives. If maps are to be processed into a Visual Materials Collection, all map derivatives should be sent in the package, and no Make derivatives instructions should be specified. 49

PAGE 50

Target server. If maps are to be processed into a Visual Materials Collection, the Target server should be IC. Formats included. If maps are to be processed into a Visual Materials Collection, the two Formats included should be Image/jpeg and Image/sid. XI.c. Entity level information Entity ID (Node name). The node name of a map entity should be the two-character institution or project code followed by a 8-digits to create a unique identifier. The Entity ID should be the same as that entered in the bibliographic record for the map in the 035 field. Entity type. Should be map. Project code(s), Bib record key, Bib record system, and Title should always be supplied. Copyright. A copyright statement should generally not be supplied, as this is will be taken from the associated bibliographic record. XI.d. Division level information Maps/photos intended for a Visual Materials Collection will not require any Division nodes. The File nodes for the map images can be moved directly under the Entity node. XI.e. File level information File nodes for the map image can be moved directly under the Entity node. Type of part. Type of part should be main. [Map] 50

PAGE 51

51

PAGE 52

XI.f. Related Images For more information on Related Images see File Level in Editing Guide To Map & Photo Metadata. Metadata for Image Files Images that are related in some way, such as the front and back of postcards, can be displayed together in Visual Collections when Other Views is selected from the full image display page. In order to associate the related images there are several elements that need to be entered when creating the MXF. Packages with non-related images contain only one image file, per image type, per package, and all of the descriptive metadata are entered at the entity level in the MXF. Packages for related images, however, contain two or more related images and the file section for each image contains the descriptive metadata that is specific to that named image file only, and is different from the other file(s) in the package. Any metadata that the images share in common can be entered at the entity level. If all of the metadata are the same, i.e., if all of the elements such as title, creator, date, etc. are the same, they can be entered at the entity level. If any of the metadata is different it must be entered at the file level. MARC record creation MARC records are created from the entity level metadata only. Metadata at the file level will be displayed on the Web but will not become part of the MARC record. New Elements for Related Images There are four new required elements at the file level for Related Image files. These elements will indicate where the thumbnail images will be displayed in the Other Views web page. 52

PAGE 53

For example if you have nine images that you want to display together in Other Views as follows: 1,1 1,2 1,3 2,1 2,2 2,3 3,1 3,2 3,3 For the image you want displayed in the upper left-most part of the web page, you would code Display Row= 1 and Display Column=1, and so on. Keep in mind that four normal size thumbnails will fit across a web page, while only three map thumbnails will fit across a web page. Each of the thumbnail image groupings in Other Views will show Display Face and Display Type at the top of the group of related images. For example: front|summary Each new combination of Display Face and Display Type will give a new grouping display in Other Views. The number of thumbnails can vary for each grouping. For example: front|summary 1,1 1,2 53

PAGE 54

or: back|detail 1,1 1,2 2,1 2,2 To show the front and back of a postcard you would get the following display in Other Views: 54

PAGE 55

Image File Naming In order to distinguish between multiple images in packages of Related Images the file names must differ. Each file name must include a suffix with an incremental number after a dash (-). For example, in the package named UF00005000, which contains three images, the files would be named, UF00005000-1, UF00005000-2, UF00005000-3, and so on. 55

PAGE 56

APPENDIX A: WORKFORM ELEMENTS PACKAGE LEVEL New Package? This must be yes or no. If this is a revision of a package sent to FCLA earlier with the same control number, New Package should be no. Otherwise, New Package should be yes. Contributing institution This element indicates the institution that did the scanning and/or metadata creation. Ordinarily a short code for the institution (e.g. FAU UF FGCU) is sufficient. This is for manual use only, and helps to identify who should be contacted if problems exist. Image directory This must be set to the fully qualified directory name of the directory containing the images for this package. If images are in multiple subdirectories, Image directory should be set to the next higher level directory containing the subdirectories. The Client will supply this directory name by default as the path to all image names. The Client will also check that all images in this directory and its subdirectories are referenced in the package, and give a warning message if any are not. Make records Indicates if bibliographic records are to be created by FCLA programs from data in this package. analytics, article = make a bibliographic records for every Division with type=article. analytics, chapter = make a bibliographic record for every Division with type=chapter. title, xx = make a bibliographic record for the entity, and put it in the Institution Group named by the xx. For example, title, QC will make a MARC record for the title in QC. If omitted, no bibliographic records are created. Make derivatives Indicates if derivative images should automatically be created by FCLA programs from master images in this package. Only one pdf derivative level may be selected. image/jpeg = create a JPEG derivative for each TIFF image sent with the MXF pdf/article = create a PDF file at the article level; this includes a PDF for each Division with a type of part article,advertisment, contents, correspondence, cover, editorial, index or section. pdf/chapter = create a PDF file at the chapter level; this includes a PDF for each Division with a type of part chapter, advertisement, bookplate, contents, correspondence, cover, editorial, index or section. 56

PAGE 57

pdf/issue = create a PDF file at the issue level; this includes a PDF for each Division with a type of part issue. pdf/main = create a PDF file at the main level; this includes a PDF for each Division with a type of part main. pdf/volume = create a PDF file at the volume level; this includes a PDF for each Division with a type of part volume or issue. If omitted, no derivatives are created. Target server Indicates which system(s) the files are intended for. DL or TC = Text collection server, for page images and PDF files FT = Text collection server, for SGML encoded full text and page images IC = Image collection server, for collections of visual materials such as maps and photographs Formats included The formats of digital files included in the package. Use the authority list in the pulldown menu. Timestamp Date and time of package creation. Supplied by the system and cannot be changed. ENTITY LEVEL Entity type Enter monograph, serial, etc. from pulldown. Source institution This is the institution that holds the original from which the contributed material was digitized. If the original entity was born digital, such as an ETD (electronic thesis or dissertation) the source institution is the institution holding the digital original. Project codes(s) Enter all project codes applicable to this entity. Version statement Must use or default a value from the pulldown. Copyright If the source material is under copyright, a statement indicating the material is under copyright and was digitized with permission of the rights holder is required. Bib record key 57

PAGE 58

If a NOTIS record exists for this entity, supply the NOTIS record number in the form AAA9999 here. Bib record system This indicates the system and file in which the record with the specified Bib record key will be found. For NOTIS records, use either FCLANOTIS:QC or FCLANOTIS:QF depending on whether the bib data is in the institution group QC or QF. Title The title of the entity as it appears on the piece or as it would appear in a catalog record. This is required at the Entity level, even if there is an external bibliographic record, to serve as a double check that the correct bibliographic record has been referenced. NOTE: The elements below, from Creator to xxx are part of an optional bibliographic description which should not be used if a NOTIS record exists for the entity. Alternative title A second or alternative title for the entity. For a serial, this could be a former or succeeding title. Creator The name of the primary author of the entity. Contributor The name of a secondary author, editor, illustrator or other person or organization associated with the entity. Publisher The publisher of the entity. Use this data element to record the publisher of the source material (e.g., Hawthorne Press) not of the electronic version (e.g. University of Florida). Subject LC Subject headings, or keywords, or other descriptors that describe the content of the entity. Description A free text narrative description of the content of the entity. If the description is an Abstract, use the Abstract element instead. Abstract An abstract of the content of the entity. Table of Contents The table of contents of the entity. This will not be used in generating a table of contents display, but may be used to provide searchable keywords in the catalog system. 58

PAGE 59

Description note Do not use; use Description instead. Description release A version, release or edition statement pertaining to the entity. Use this to describe the source material, not the digitized version. Rights Use Copyright for a short statement of copyright status and permission to digitize. Use Rights for a longer explanation of complex terms and conditions. Extent Size of the entity. Use for filesize of the digital version. If the digital version is made up of many small files, you can estimate the total filesize here, but indicate it is an estimate and how it is derived, e.g. Roughly 5MB in 15 PDF files. Medium ? Type ? Coverage Use for the geographical or temporal coverage of the content of the entity. For example, -1864 or Alachua County. Date A date associated with the entity. This date is used for statistical purposes. If describing a serial, see the chapter in this manual, Special Formats: Serials, for instructions how to record this date. Creation date Creation date of the digital version. Publication date Publication date of the source or digital version, if formally published. Do not use for the date of making a digitized entity available on the web. Modification date The date of last change of the entity. Identifier You can use this to record any standard number(s) assigned to the entity, such as ISSN or ISBN, as well as digital identifiers such as DOI. Source 59

PAGE 60

Do not use. Is part of If the entity is part of a larger publication (for example, a map that is reproduced from an atlas, or a book issued in series) give bibliographic information for the larger publication here. Has part If the entity has component parts, you can give a bibliographic description of the parts here. Is version of If the entity is a bibliographic version of another entity, you can describe the other entity here. Do not use for physical versions (i.e. for TIFF and MrSid versions). Is format of If the entity is a derivative of another physical representations (e.g. a sid file made from a TIFF) you can describe the source version here. Has format If the entity has other physical representations, you can describe them here. (Use Is format of instead if this version was derived from another representation.) Language The language of the content of the entity, if not English. E.g. French. Thesis degree If the entity is an ETD, then enter the degree granted here, e.g. B.A, M.S., PhD. Thesis discipline If the entity is an ETD, then enter the department in which the degree was granted here, e.g. Organic chemistry. Thesis grantor Thesis level DIVISION LEVEL Type of part Select from pulldown. The type of structural part that this division represents, e.g. chapter, article, volume. Be as specific as possible, but if no part name in the pulldown seems to apply, use section. Use main for the main part of an ETD, if represented by a single PDF. 60

PAGE 61

[type of part] name The label for this element will change depending on the type of part chosen, i.e. if Type of part is chapter, then this label will say Chapter name. Use this element when the parts have formal numbering that consist of the part designation and some alpha or numeric enumeration, e.g. Chapter 1, Part II. Namespace reference(s) Do not use. Note: Remaining elements, from Title through Thesis level, are essentially the same as at the Entity level. However, the meaning of the bibliographic data pertains to this particular Division instead of the entire entity. E.g., Title refers to the title of the Division. FILE LEVEL Note: If the file is for Visual Collections and there is more than one file in the package, bibliographic data that is specific to an image file needs to be entered at the file level. The elements, from Title to Thesis, are essentially the same as at the Entity Level. However, the bibliographic data pertains to this particular file rather than to the whole entity. If an element is the same for all the images in a package, it can be entered at the entity level. (See XI.f. Related Images for a full explanation of Related Images.) Type of part Select from pulldown. The type of structural part that this file represents. For example, a PDF of a book chapter would have Type of part chapter. A TIFF of a book page would have Type of part page. [type of part] name The label for this element will change depending on the type of part chosen, i.e. if Type of part is chapter, then this label will say Chapter name. Use this element when the parts have formal numbering that consist of the part designation and some alpha or numeric enumeration, e.g. Chapter 1, Part II. File name The fully qualified filename (including extension) of the file, e.g. m:\user\Priscilla\DigitizedBooks\FA00000033\1.tif. File format The MIME Type (Internet media type and subtype) of the file, and the version of the file if applicable; e.g. image/tiff 6.0, application/pdf x.x. File order Files will be ordered by default in the same sequence as the metadata describing them. If some other order is required, you can use File order to give an absolute ordering. If any 61

PAGE 62

one file in a package has a File order, then all files in the package must. This element is generally not used. Filesize File size in bytes. This is supplied by the client from the file header whenever possible. When you must supply a value, be sure to give the value in bytes, and do not include any unit designation, e.g. not .4 MB. Checksum A calculated checksum for the file. This will be supplied by the Client and cannot be altered. Checksum type The type of checksum calculated. This will be supplied by the Client and cannot be altered. Timestamp The date and time of file creation. This is supplied by the Client from the file header when possible. File id An internal identifier supplied by the Client. Creating institution The institution responsible for image creation. This should be given at the level of the library, rather than of the university or library department, e.g. University of Florida, University Libraries. Creating individual The person who actually created the image via scanning, digital camera, etc. The individuals department or other information can be supplied if there is a possibility of confusion, e.g. George Smith, Digital Library Center. Creation method How the image was created, for example, Scanned from paper or Output from MS Word in PDF format. Software used The name of the software package used to create the file, e.g. Adobe Capture. Software version The version of the software used, e.g. .2. Name of compression scheme If the file is compressed, the name of the compression scheme used. 62

PAGE 63

Compression The compression ratio, or other textual information about compression apart from the name of the scheme. Note: The Following four elements (Display Row, Display Column, Display Type, Display Face) are required only in packages that have one or more Related Images in a package. (see Appendix F for a full explanation of Related Images.) Display Row The row number where the thumbnail image will display when Other Views is selected. Display Column The column number where the thumbnail image will display when Other Views is selected. Display Type The value that will display at the top of the group of thumbnail images when Other Views is selected. Values are sum (summary), and det (detail). If not entered, will default to sum. Display Face The Second value that will display at the top of the group of thumbnail images when Other Views is selected. Values are: front, back right, left, top and bottom. If not entered, will default to front. Capture A textual description of the capture method. It is better to use Capture device used and Type of capture device. Capture device used The make and model of the digital camera or scanner used. Type of capture device Select from pulldown. Scanner settings If the capture device was a scanner, record the settings in Gamma, Brightness, Contrast, etc. and set Scanner settings to yes. Gamma/Brightness/Contrast/Saturation/Percentage Record the settings used to drive the scanner used as capture device. Camera settings Do not use. Light source 63

PAGE 64

Textual description of source of light used when scanning. Tracking id If you have an internal tracking id number for this batch of images, you can include it here. Image bit depth Record as a number, e.g. 1, 16, 24. Image resolution For example, 600 DPI. Storage segment For TIFF images, whether recorded as strip or tile. Select from pulldown. Storage planar configuration For planar formats only, whether the image is chunky or planar. If not applicable, use unknown. Unit of sampling frequency Inch, centimeter, etc. Select from pulldown. Sampling plane The reference plane for sampling frequency. Select from pulldown. Image dimensions (horizontal) The number of pixels in the sampling frequency unit in the image width. Image dimensions (vertical) The number of pixels in the sampling frequency unit in the image length. Image color space Color space (color mode) of image, e.g. linear RGB. Color management Description of any process used to improve consistency of color across capture, display, and output devices. Color lookup table Color lookup table used. Targets included Textual information about color bar or greyscale bar used as internal target. Type of target included Either color or greyscale, select from pulldown. 64

PAGE 65

Descreening software Software used for descreening. Descreening settings Settings used with descreening software. Orientation Either portrait or landscape, select from pulldown. Derived from (filename) If this file was derived from another file (e.g., a JPEG image made from a TIFF), you can give the filename of the source file here. Alternatively, if the source file is described in this MXF package, you can leave this element blank and give the File id of the source file in the following element. NOTE if you use the Clients Derive Node function, this is automatically supplied. Derived from (file id) If this file was derived from another file (e.g., a JPEG image made from a TIFF) included in this MXF package, you can give the File id of the source file here. This is an alternative to giving the filename in the previous element. NOTE if you use the Clients Derive Node function, this is automatically supplied. Derived from (description) If this file was derived from a non-digital original, or if the relationship between the source file(s) and this file more complex than a one-to-one derivation, use this element for a textual description of the files involved and their relationship to each other. Source dimensions (horizontal) Horizontal (width) dimensions of a non-digital source file scanned to make this file, e.g. Do not include the unit of measurement here, but in the Unit of source dimensions element. Source dimensions (vertical) Vertical (height) dimensions of a non-digital source file scanned to make this file, e.g. Do not include the unit of measurement here, but in the Unit of source dimensions element. Unit of source dimensions The unit of measurement for horizontal and vertical dimensions of the source, e.g. cm.. Namespace reference(s) Do not use. Note: Remaining elements, from Title through Thesis level, are essentially the same as at the Division and Entity levels. However, the meaning of the bibliographic data pertains 65

PAGE 66

to this particular File. As a general rule, bibliographic description should be given at the Entity and or Division levels, and not at the File level. An exception to this rule is made for Related Images. If there is more than one JPEG image in an MXF package, i.e., one or more related images, the bibliographic description must be given at the file level if the element values differ for the images. For example, if the images have different titles and descriptions, these must be indicated at the file level. Bibliographic description that is shared between images is indicated at the Entity level. 66

PAGE 67

Appendix B: How the MXF is encoded The MXF is encoded in XML. Data elements are delimited by start tags and end tags. Start tags look like this: End tags look like this: The value of the data element is usually given between the start and end tags, e.g. FHP means the value of the data element projects is FHP. Some data elements have attributes, or more specific sub-elements. In this case they are shown within the start tag in attribute=value format: Elements can be nested within other elements, in which case the meaning is taken to be hierarchical. For example, there may be two elements nested as follows:
...
...
In this case, it is understood that the page is contained within the chapter. (Note that although this kind of nesting is often represented graphically using indentation, the actual XML is not indented.) The MXF Client software is designed to hide the actual XML syntax of the MXF from the person doing the data entry. Rather than seeing start tags and end tags, you see screen labels and boxes for entering data. When all the information has been entered, you can tell the Client to export the data in the appropriate XML format. 67

PAGE 68

APPENDIX C: VALIDATION MESSAGES The MXF Client has two kinds of Validation Messages: Errors and Warnings. An Error must be fixed before XML can be output. Warnings do not have to be fixed. Appendix C.1. Error messages The following are Error messages: "No value specified for [required element]" The cursor will be at the required element in the Workform. Some value must be supplied. "The name Volume name or Issue name value (in the metadata workform) for a volume or issue contained within a serial entity must be numeric." If the Entity type is serial and the Division nodes Type of part is volume or issue, then the Volume name or Issue name field must have some value. For example, if you are describing Volume 43, then Volume name should be "[filename] is not an existing file." In a File node, the file identified in File name was not found in the Image directory for the package, or in any of the subdirectories of the Image directory. "The 'Derived from' value [file id] is not a valid file id from this package" If you use the Tree/Derive node function to create a derived file node, the source node must exist in the package at the time of validation. If you get this error message, somehow the source file must have been removed from the Image directory. "The specified image is not in the path given in the package level image directory." A file node references a file that was not found in the image directory specified in the Package level field Image directory. Appendix C.2. Warning messages The following are Warning messages: "No images have been referenced in this package." The package does not refer to any files. That can only happen if the package does not contain any File nodes, since a Filename is required on the File workform. 68

PAGE 69

Unable to determine image directory. Image directory cannot be located, XML file will be stored in the package directory. "The file [file path] exists in the image directory, but is not referenced in the package." The directory designated on the Package workform as the Image directory contains certain files that are not referenced in the package. This message will only appear if the filetypes of the unreferenced files are .tif, [what else] 69

PAGE 70

APPENDIX D: OPTIMIZING WORKFLOW Organizing your workflow and using shortcuts, defaults and control keys can speed data entry significantly. Here are some suggestions for a few common cases. a. Book, images exist In this example, there are TIFF and JPEG images for every page of a book. a.1. Set up your new package and Package node. a.2. Set up your Entity node for the book. a.3. Create nodes for all Divisions: Have the Table of Contents or Division List in front of you. For nodes with unique names (e.g. Copyright permission through Foreword): Use CTRL-a (Add node) to create the first Division under the Entity node. Hit the back arrow (up one level) to highlight the Entity node again. Use CTRL-a (Add node) to create the next Division. Hit the back arrow (up one level) to highlight the Entity node again. Continue creating Division nodes with CTRL-a and back arrow until done. For nodes with sequentially numbered names (e.g. Chapter 1 through 10): Use CTRL-s (Insert Range) to create nodes for these Divisions. a.4. When all Divisions have been created, create File nodes for the images: For each division, highlight the TIFF images for the pages in that division (Hold CTRL down and click the image names) Drag the TIFFs from the Directory Window to the appropriate Division in the Tree Window. For each TIFF, highlight the TIFF node and use CTRL-d (Derive node) to create a derived node for the JPEG. 70

PAGE 71

APPENDIX E: TRAINING EXERCISES Exercise 1: Book worksheet Preparing a worksheet. To create an MXF you need to know the structure of the publication. If you wont have the publication in front of you when you make the MXF, you can record the important information on a worksheet. In this exercise, because we dont have the source publication even to make the worksheet, you will have to pretend you do by inferring the information from the online version in Florida Heritage. 1. Read Section I Understanding the MXF and Section II, Preparing Documents. 2. In Florida Heritage, look up the title, Conjured into being: Zora Neale Hurstons Their Eyes were watching God. Fill out Worksheet 1 (empty) on the following page. The Division Name is what will appear in the table of contents display. Type of Part must be taken from the following list: abstract, advertisement, article, box, chapter, contents, correspondence, cover, editorial, file, item, main, page, part, permission, recordgroup, section, series, subgroup, subseries, suppl, volume. Fill in Page or range when the division consists of numbered pages (e.g. i-vii). Comments can be used to record the filename, if other than the page number. (There is a convention to name files the same as the pages, e.g. the TIFF file for page 1 would be 1.tif.) Since you dont know the filenames for this exercise, you can omit this information. Comments can also be used to record additional bibliographic information for the section, such as an author or long title. The first entry is filled in for you. 3. Compare your filled-out worksheet with Worksheet 1 (completed). Yours does not have to look exactly like this but be sure you are comfortable with the differences. 71

PAGE 72

Worksheet 1 (empty) Division Name Type of Part Page or range (n-n) Comments Front Matter section approval, title 72

PAGE 73

Worksheet 1 (completed) Division Name Type of Part Page or range (n-n) Comments Front Matter section approval, title Table of Contents contents ii Abstract abstract iii-v Introduction chapter 1-11 Free indirect discourse chapter 12-16 Signifying chapter 17-23 Conjuring chapter 24-28 Signifying/ Conjuring chapter 29-33 Conclusion chapter 34-38 Literature cited section 39-41 73

PAGE 74

Exercise 2: New package This exercise will use the images in the SF00000012 folder. Before beginning, create a directory for your MXF work. For example, this might be called /path/yourName/MXFwork. Copy the SF00000012 folder into /MXFwork. Pretend you are at USF, creating an MXF to transport these images to FCLA for loading. 1. Read sections III Basics of the MXF Client, IV Setting Defaults, and V Creating a New MXF package. 2. Log into the MXF Client. 3. Set the following Defaults: Package: New package? (yes) Contributing institution (FSU) Make derivatives (application/pdf) Target server (DL) Formats included (image/jpeg, image/tiff) Entity: Source institution (USF) Version statement (Electronic version created 2001...) Bib record system (FCLANOTIS:QF) Division: Type of part (chapter) File: Type of part (page) Image format (image/tiff) 4. Create a new package named SF00000012. Set Image Directory on the metadata workform to the correct directory. Make sure the correct directory is in the Directory window and that the images are listed. Fill in any missing information on the metadata Workform at the Package level. 5. When you are done, compare your completed workform with the examples below. [Exercise 2-1, Exercise 2-2]. They dont need to be identical but be sure you understand any differences. 74

PAGE 75

75

PAGE 76

6. Close the package by selecting Package/Close. Close the Metadata Module by clicking the close box at the right of the top Menu bar. Close the MXF Client by clicking the close box in the upper right-hand corner. 76

PAGE 77

Exercise 3: Adding Entity and Division level information In this exercise you will use the worksheet you created in Exercise 1 to add Entity and Division nodes to the tree. Assume there is already a NOTIS catalog record for this title in the QF (PALMM) institution, and that its NOTIS record number is AAA1070. 1. Open the MXF client, using your username and password. 2. Open the Metadata module (Module/Metadata). 3. Open the Package for SF00000012: Select Package/Open. A popup box will suggest the last package file that you had open. If this is correct (it should be) highlight the name of the package and click Open. Note that the correct Image directory will display in the Directory Window, and the Package workform displays. 4. Create an Entity node: Select Tree/Add node. The Client assumes you want to add an Entity node as a child to the Package node; this is correct. Type the name of the Entity Node in the box. For non-serial publications, this should be the same as the name of the Package Node, so enter SF00000012 and click OK or hit Enter. The Entity node will appear on the tree, and an Entity workform will display. 5. Add Entity level information: Fill out the Entity workform. Make sure you have filled out Bib record key (AAA1070) and Title (Conjured into being: Zora Neale Hurstons Their Eyes were watching God). 6. Create a Division node: Use CTRL-a (same as Tree/Add node). The Client assumes you want to add a Division node as a child to the Entity node; this is correct. Type the name of the Division node in the box. This should be whatever you recorded on the worksheet -it will display in the table of contents for this entity. If you followed the sample worksheet, it should say Front Matter. 7. Add Division level information: Probably the only thing you need to look at is Type of part. You may have to change the default to match what you have on your worksheet. If you followed the sample worksheet, it should say section. 8. Add remaining Divisions: Make the Entity node the current node. If the Tree window is active, just hit the Left arrow to take you to the parent of the current node, which will be the Entity node. If the Workform window is active, press F2 to make the Tree window active, then Left arrow. Alternatively, you can just click the Entity node to make it current, but using the mouse is usually slower than the other methods. Use CTRL-a to create a new node. The default node type (division, child) should be correct. Enter the Division node name here, and the correct Type of part on the workform. Repeat for all remaining divisions on your worksheet. (Tip: if the default Type of part was correct and you did not need to enter anything in the Workform window, the Tree window will be the current window. Hitting the Left arrow will take 77

PAGE 78

you to the parent of the current node, which will be the Entity node. This is faster than clicking on the Entity node.) 9. When you are done, compare your work to the example below. [Exercise 3]. Be sure you understand any differences. 10. Close the Package: Close and save the package with Package/Close. You can leave the client open in the Metadata module. 78

PAGE 79

Exercise 4: Adding File level information and finishing the Package In this exercise you will finish the package you started building in Exercise 2 and 3. 1. Open the Package for SF00000012. If you left the tree in expanded form (with all nodes showing) it should still be expanded. If any node is compressed (preceded by a plus), click on the plus to expand it, or select Tree/Expand all. 2. Make nodes for TIFF files: Using your worksheet to relate pages to divisions, copy over the TIFF images for the first division. In this example, the first division is Front Matter and there are two pages in it, represented by 1.tif and 2.tif. To move more than one image at a time, hold down the CTRL key while clicking on the name of each image in the Directory window. Then clicking on one of the images, drag it over to the correct node. 3. Make nodes for derivatives: Click on the first TIFF (1.tif) and check that the Workform data is correct. Then use Tree/Derive node or CTRL-d to create a node for the derivative JPEG. The cursor will be at the end of the default File name in the Workform window; type the file type (.jpg) and press ENTER. Default information for the JPEG image will be supplied. 4. Make remaining file nodes: Continue making nodes for the TIFF and JPEG files by dragging the TIFF files from the Directory window and making derivative nodes for the JPEGs. 5. Validate and output the package. Make sure that View/XML Tree on Output is not checked. Select Package/Output to XML. This will cause the MXF client to validate the data you have entered. There will be a popup box giving the number of errors and warnings. Click OK to clear the box, then use PF5 to review each error in turn. Correct any errors found and repeat Package/Output to XML until no errors occur. You will get a message saying the XML file has been created. 6. Now select View/XML Tree on Output so that it is checked. Select Package/Output to XML again. Note that this time you see a tree version of the output XML file. If you want to continue viewing the tree for other packages, leave XML Tree on Output checked, otherwise turn it off. 7. Select View/XML file. The output XML will display in a Notepad window. Close the window. Close the package (Package/Close). Close the MXF Client. 79

PAGE 80

APPENDIX F: MXF Quick Guides for Textual Materials Before creating MXF for your textual materials, it is very important to understand and decide on the correct entity type for your materials. Currently, Textual Collections supports three entity types for all the textual materials, monograph, multipart and serial. A monograph (a single part monograph or single volume monograph) is an item complete, or intended to be complete, in one part. A serial is a publication issued in successive parts, bearing numerical or chronological designations, and intended to be continued indefinitely. Examples include journals, periodicals, newspapers, and annuals (reports, yearbooks, etc.) A multipart (a multipart monograph or multi-volume monograph) is an item complete, or intended to be complete, in a finite number of separate parts. MXF: Quick Guide for Monograph Submission A complete work with all chapters and other sections should be sent in one package to FCLA. Directory setup for monograph \SF00000012 [package ID] JPEG, TIFF, PDF, XML, SGML, DTD, etc Package Level information Package ID (Node) 10-digit ID in the standard package ID format, consisting of the 2-character institution or project code followed by an 8-character numeric value, e.g. FI12345678, FA12345678. Image directory The directory where image files are found. In the example above this would be ...\SF00000012\. Make records Use title, QF or title, QC if you want a MARC record made from the metadata in the MXF. In general, records should only go in QF if they have valid authority-controlled name and subject headings. 80

PAGE 81

Make derivatives Select pdf/chapter if you want PDF files created according to the division nodes of the following Type of Part: chapter, advertisement, bookplate, contents, correspondence, cover, editorial, index, main, section, permission. The PDF files created based on JPEGs you send us will be displayed parallel to the JPEG version of the item in Textual Collections. Target server Use DL or TC to send JPEG page images and have us create the SGML file to be loaded into Textual Collections. Use FT to send JPEG and/or PDF images and your own fulltext SGML file for us to load directly into Textual Collections. Entity Level information Entity ID (Node) Monograph entity ID should be the same as the package ID. In the example above this would be SF00000012. Entity type monograph Source institution The institution who holds the original from which the material was digitized. The institution code supplied here will govern the Print Source statement in the full citation of the item in Textual Collections. For example, if you supply FIU, the Print Source statement will be Digitized from original source held at Florida International University Libraries; if you supply FAU, the Print Source will say Digitized from original source held at Florida Atlantic University Libraries. Project code(s) Supply only one primary project code and any applicable sub-project code(s). Make sure this is consistent with what is recorded in 852$b of the bib record, if such a bib record exists. For project code(s), see http://allegro.fcla.edu/ProjectCodes.cfm Version statement Either select a version statement from the pull-down or enter a statement that describes the origin of the digital version, e.g. Electronic version created yyyy, State University System of Florida. Version statement will be displayed in full citation of the item in Textual Collections. 81

PAGE 82

Bib Record Key If a NOTIS record exists, supply the NOTIS record key in the form of AAA1234 here. This associates this package with the external bibliographic record. Bib record system Use either FCLANOTIS:QF or FCLANOTIS:QC to indicate the system and file in which the record with the specified Bib record key will be found. Bib record key and Bib record system must come in pairs. The bibliographic record will be used to extract descriptive metadata and displayed as citation for the item in Textual Collections. Title The title of the monograph. This is required. Rights Use for a statement of rights related to the original or digitized version. If the material is under copyright and digitized with permission, note this here. In Textual Collections, the Rights is displayed as Rights Information. Do not use Copyright to record this information. Division Level information Most monographs will have at least one Division level. A monograph generally contains logical structural parts such as chapters, table of contents, index, cover, etc. Each of these parts or other section should be a division node. Each of the division nodes usually parents a number of file nodes which contain the page images. However, there is no reason that division nodes cannot be nested in division nodes. Node name The name of the node (Division name) should be what you want to appear in the online table of contents for the item. If the node is a chapter, you should generally use the title of the chapter; however, if the title is very long, then use a short form as the node name, and put the complete title in Title. If the node is some section of the item that does not have a name, make up some descriptive heading, e.g. Letters. Type of part Commonly used value: permission, cover, contents, section, chapter, index, cover. Title Title can usually be left blank. Its used only when a title is very long. Anything entered in Title will be used instead of the Node name in the table of contents displayed to the user. Typing in a long title as the node name can make the tree look messy and hard to read. Instead, type an abbreviated form of the title as the node name, and type the complete title in Title. 82

PAGE 83

File Level information File nodes should be children nodes under the parent nodes for the contents of book divisions. File nodes generally contain page images. There are several required elements at the file level, but most of them can be supplied by the system automatically. Type of Part Usually use Page. File name Automatically supplied when you browse and select the file or use the drag method to create the file node. It is recommended that if you have different file formats for the same page image, they should use the same filename, but with different file extension. File format Automatically supplied by the system if it can get this from the extension on the filename. It is required that if you have different file formats for the same page image, you have to create sibling file nodes for them continuously. For example, if you have both cover.tif and cover.jpg for the Cover page image, the file nodes for cover.tif and cover.jpg should come in pair consecutively. File size Automatically supplied by the system when the information is present in the image header. 83

PAGE 84

An example of a monograph showing many sections as contents. 84

PAGE 85

MXF: Quick Guide for Multipart Submission All parts have to be wrapped in one package and sent to FCLA together. Directory setup for multipart When set up directories for your multipart package, you must create a separate subdirectory for each part/volume: \UF80001549 [package ID] \ UF80001549 [entity ID] \1 [images for volume 1] \2 [images for volume 2] etc. Use numeric characters only for the name of the subdirectory for each part/volume. Package Level information Package ID (Node) 10-digit ID in the standard package ID format, consisting of the 2-character institution or project code followed by an 8-character numeric value, e.g. FI12345678, FA12345678. Image directory Set the Image directory to the directory corresponding to the Entity ID. In the example above this would be ...\UF80001549\ UF80001549. Make records Use title, QF or title, QC if you want a MARC record made from the metadata in the MXF. In general, records should only go in QF if they have valid authority-controlled name and subject headings. Make derivatives Select pdf/chapter if you want PDF files created according to the division nodes of the following Type of Part: chapter, advertisement, bookplate, contents, correspondence, cover, editorial, index, main, section, permission. The PDF files created based on JPEGs you send us will be displayed parallel to the JPEG version of the item in Textual Collections. 85

PAGE 86

Target server Use DL or TC to send JPEG page images and have us create the SGML file to be loaded into Textual Collections. Use FT to send JPEG and/or PDF images and your own fulltext SGML file for us to load directly into Textual Collections. Entity Level information Entity ID (Node) Multipart entity ID should be the same as package ID. Entity type multipart Source institution The institution who holds the original from which the material was digitized. The institution code supplied here will govern the Print Source statement in the full citation of the item in Textual Collections. For example, if you supply FIU, the Print Source statement will be Digitized from original source held at Florida International University Libraries; if you supply FAU, the Print Source will say Digitized from original source held at Florida Atlantic University Libraries. Please be aware of the difference between Contributing institution at the package level and Source institution at the entity level. Project code(s) Supply only one primary project code and any applicable sub-project code(s). Make sure this is consistent with what is recorded in 852$b of the bib record, if such a bib record exists. For project code(s), see http://allegro.fcla.edu/ProjectCodes.cfm Version statement Either select a version statement from the pull-down or enter a statement that describes the origin of the digital version, e.g. Electronic version created yyyy, State University System of Florida. Version statement will be displayed in full citation of the item in Textual Collections. Bib Record Key If a NOTIS record exists, supply the NOTIS record key in the form of AAA1234 here. This associates this package with the external bibliographic record. 86

PAGE 87

Bib record system Use either FCLANOTIS:QF or FCLANOTIS:QC to indicate the system and file in which the record with the specified Bib record key will be found. Bib record key and Bib record system must come in pairs. The bibliographic record will be used to extract descriptive metadata and displayed as citation for the item in Textual Collections. Title The title of the multipart. This is required. Rights Use for a statement of rights related to the original or digitized version. If the material is under copyright and digitized with permission, note this here. In Textual Collections, the Rights is displayed as Rights Information. Do not use Copyright to record this information. Division level information Most multipart will have at least two levels of hierarchy at the Division level: volume and the contents of the volume (chapters, index, table of contents, etc.). This is represented on the Tree as parent and child nodes: the Division for the volume is the parent of the Divisions for the contents of the volume. Volume Node Node name The node name for the volume should be what you want to display as part of the title for the item in Textual Collections, e.g. Part X: part title For example the volume node name Part 1 General Statistics will be concatenated with the title at entity level Industrial and economic survey of Pensacola: Industrial and economic survey of Pensacola, thus forming a title of Industrial and economic survey of Pensacola: Industrial and economic survey of Pensacola Part 1 General Statistics, which is displayed to users as the title of the item in Textual Collections. Type of part Use volume. Volume name This is required for a multipart. It must be the numeric designation of the volume, e.g. if this is volume 1, the 87

PAGE 88

Volume name must be If there isnt a numeric designation, you have to establish one. Date This is required for a multipart. The date in this field will be used to make a chronology tracing for the date of publication of the original materials. Use the date of publication for the volume in hand. For volumes that don't actually have real publication dates printed in, you need to infer the date or determine a probable date of publication and enter this in brackets with a question mark, e.g. [1923?]. The date will be displayed as Publication date for the item in Textual Collections as well as for sorting the search results according to date ascending, date descending. The date in this field will also be used for statistical purposes. Contents of volume Nodes Multipart is multi-volume monograph. Each volume generally contains logical structural parts such as chapters, table of contents, index, cover, etc. just like monograph does. Each of these parts or other section should be a division node nested under the volume node. Node name The name of the node (Division name) should be what you want to appear in the online table of contents for the item. If the node is a chapter, you should generally use the title of the chapter; however, if the title is very long, then use a short form as the node name, and put the complete title in Title. If the node is some section of the item that does not have a name, make up some descriptive heading, e.g. Letters. Type of part Commonly used value: permission, cover, contents, section, chapter, index, cover. Title Title can usually be left blank. Its used only when a title is very long. Anything entered in Title will be used instead of the Node name in the table of contents displayed to the user. Typing in a long title as the node name can make the tree look messy and hard to read. Instead, type an abbreviated form of the title as the node name, and type the complete title in Title. File Level information 88

PAGE 89

File nodes should be children nodes under the parent nodes for the contents of volume divisions. File nodes generally contain page images. There are several required elements at the file level, but most of them can be supplied by the system automatically. Type of Part Usually use Page. File name Automatically supplied when you browse and select the file or use the drag method to create the file node. It is recommended that if you have different file formats for the same page image, they should use the same filename, but with different file extension. File format Automatically supplied by the system if it can get this from the extension on the filename. It is required that if you have different file formats for the same page image, you have to create sibling file nodes for them continuously. For example, if you have both cover.tif and cover.jpg for the Cover page image, the file nodes for cover.tif and cover.jpg should come in pair consecutively. File size Automatically supplied by the system when the information is present in the image header. 89

PAGE 90

MXF: Quick Guide for Serial Submission Each issue of the serial has to be sent to FCLA in a separate package, the Package ID must be unique for each package (issue), but the Entity ID must be the same for all issues. For a serial with title changes, a unique Entity ID should be created when the title changes. The Entity ID is a unifying identifier that ties all the pieces under each title together. Directory setup for serials \UF80001549 [package ID] \SN00130098 [entity ID] \1_1 [directory for volume 1 issue number 1] Use numeric characters only for the name of the subdirectory for each serial issue. Package Level information Package ID (Node) 10-digit ID in the standard package ID format, consisting of the 2-character institution or project code followed by an 8-character numeric value, e.g. FI12345678, FA12345678. Image directory The subdirectory for each serial issue. In the example above this would be ...\SN00130098\1_1. Make records Use title, QF or title, QC if you want a MARC record made from the metadata in the MXF. In general, records should only go in QF if they have valid authority-controlled name and subject headings. Make derivatives Select pdf/article if you want a PDF file created for every article in the issue, the division nodes of following Type of Part will be wrapped into PDFs: article, advertisement, contents, correspondence, cover, editorial, index, section, permission. Select pdf/issue if you want a PDF file created for the issue, the division nodes of the following Type of Part will be wrapped into PDFs: issue, permission. 90

PAGE 91

The PDF files created based on JPEGs you send us will be displayed parallel to the JPEG version of the item in Textual Collections. Target server Use DL or TC to send JPEG page images and have us create the SGML file to be loaded into Textual Collections. Use FT to send JPEG and/or PDF images and your own fulltext SGML file for us to load directly into Textual Collections. Entity Level information Entity ID (Node) Serial entity ID must consist of the 2-character SN and the 8-character ISSN of the serial. For example, if the ISSN is 0123-4567, then the Entity ID must be SN01234567. The entity ID stays the same for all the issues regardless of the package IDs that are used. Some older serial titles do not have ISSNs or some materials you want to package as serial are not serials In this case there are two options: apply for a retrospective ISSN, or assign a pseudo-ISSN. Instructions on how to apply for a retrospective ISSN, or assign a pseudo-ISSN are available at http://palmm.fcla.edu/strucmeta/pseudoissn.html Entity type serial Source institution The institution who holds the original from which the material was digitized. The institution code supplied here will govern the Print Source statement in the full citation of the item in Textual Collections. For example, if you supply FIU, the Print Source statement will be Digitized from original source held at Florida International University Libraries; if you supply FAU, the Print Source will say Digitized from original source held at Florida Atlantic University Libraries. Project code(s) Supply only one primary project code and any applicable sub-project code(s). Make sure this is consistent with what is recorded in 852$b of the bib record, if such a bib record exists. 91

PAGE 92

For project code(s), see http://allegro.fcla.edu/ProjectCodes.cfm Version statement Either select a version statement from the pull-down or enter a statement that describes the origin of the digital version, e.g. Electronic version created yyyy, State University System of Florida. Version statement will be displayed in full citation of the item in Textual Collections. Bib Record Key If a NOTIS record exists, supply the NOTIS record key in the form of AAA1234 here. This associates this package with the external bibliographic record. Bib record system Use either FCLANOTIS:QF or FCLANOTIS:QC to indicate the system and file in which the record with the specified Bib record key will be found. Bib record key and Bib record system must come in pairs. The bibliographic record will be used to extract descriptive metadata and displayed as citation for the item in Textual Collections. Title The title of the serial. This is required. Rights Use for a statement of rights related to the original or digitized version. If the material is under copyright and digitized with permission, note this here. In Textual Collections, the Rights is displayed as Rights Information. Do not use Copyright to record this information. Alternative title. This field is used to trace preceding title(s) or/and succeeding title(s) for serials with title changes. You must record the preceding title in the form of Preceding title: preceding title, and the succeeding title as Succeeding title: succeeding title, e.g. Preceding title: Report of the Board of Control of the State Institutions of Higher Learning of Florida for the biennium ending ...., Succeeding title: Report of Florida Board of Regents. Division level information Most serials will have at least three levels of hierarchy at the Division level: volume, issue (number), and the contents of the issue (articles, correspondence, editorials, etc.). This is represented on the Tree as parent and child nodes: the Division for the volume is 92

PAGE 93

the parent of the Division for the issue, which is the parent for the Divisions for the articles within the issue. Volume Node Node name The node name for the volume should be what you want to display as part of the title for the item in Textual Collections. It is important that you record all node names for volumes and issues consistently throughout the entire serial run. Recommended practice is to use either Volume x (year)or Vol. x (year) as this gives the most information to the reader. For example, Volume 1, Vol. 1 (1898/1899). The Florida Historical Quarterly volume 1 issue 1 Type of part Use volume. Volume name This is required for a serial. It must be the numeric designation of the volume, e.g. if this is volume 43, the Volume name must be If there isnt a numeric designation, you have to establish one. Date This is required for a serial. The date in this field will be used to make a chronology tracing for the date of publication of the original materials. Use the date of publication for the volume in hand. For volumes that don't actually have real publication dates printed in, you need to infer the date or determine a probable date of publication and enter this in brackets with a question mark, e.g. [1923?]. The date will be displayed as Publication date for the item in Textual Collections as well as for sorting the search results according to date ascending, date descending. The date in this field will also be used for statistical purposes. Issue Node Issue node is created as child of the parent volume node. Please note that for annuals or serials that only have volumes, issues node is not required. Node name As with volumes, the node name for the issue should be want to display as part of the title for the item in Textual Collections. It is important that you record all node names for volumes and issues consistently throughout the entire serial run. Recommended best practice is to use either 93

PAGE 94

Number x or No. x and to put the chronology in parenthesis. For instance, Number 1, No. 1 (Spring). The Florida Historical Quarterly volume 1 issue 1 Type of part Useissue. Issue name This is required for a serial when applicable. It must be the numeric designation of the issue. E.g. if this is issue 3, the Issue name must be If there isnt a numeric designation, you have to establish one. Date This is required for a serial when applicable. The date in this field will be used to make a chronology tracing for the date of publication of the original materials. Use the date of publication for the issue in hand. For issues that don't actually have real publication dates printed in, you need to infer the date or determine a probable date of publication and enter this in brackets with a question mark, e.g. [1923?]. The date will be displayed as Publication date for the item in Textual Collections as well as for sorting the search results according to date ascending, date descending. The date in this field will also be used for statistical purposes. Note that the Date recorded for the issue should be the same as the Date used for the volume. Contents of Issues Nodes Issues of serials generally contain material such as letters to the editor, articles, advertisements, etc. Each article or other section should be a child node under the parent node for the Issue. Node name The name of the node (Division name) should be what you want to appear in the online table of contents for the issue in Textual Collections. If the node is an article, you should generally use the title of the article; however, if the title is very long, then use a short form as the node name, and put the complete title in Title. If the node is some section of the issue that does not have a name, make up some descriptive heading, e.g. Letters. Type of part Commonly used values: permission, cover, contents, correspondence, article, abstract, advertisements, editorial, index. 94

PAGE 95

Title Title can usually be left blank. Its used only when a title is very long. Anything entered in Title will be used instead of the Node name in the table of contents displayed to the user. Typing in a long title as the node name can make the tree look messy and hard to read. Instead, type an abbreviated form of the title as the node name, and type the complete title in Title. Creator If the Division node is for an article with one or more named authors, you should enter the name of the author in this element. File level information File nodes should be children nodes under the parent nodes for contents of issue divisions. File nodes generally contain page images. There are several required elements at the file level, but most of them can be supplied by the system automatically. Type of Part Usually use Page. File name Automatically supplied when you browse and select the file or use the drag method to create the file node. It is recommended that if you have different file formats for the same page image, they should use the same filename, but with different file extension. File format Automatically supplied by the system if it can get this from the extension on the filename. It is required that if you have different file formats for the same page image, you have to create sibling file nodes for them continuously. For example, if you have both cover.tif and cover.jpg for the Cover page image, the file nodes for cover.tif and cover.jpg should come in pair consecutively. File size Automatically supplied by the system when the information is present in the image header. 95

PAGE 96

An example of a serial showing many sections as contents. 96