• TABLE OF CONTENTS
HIDE
 Introduction
 All dolled up...
 A place to lay my head
 The physic of dream
 Awakening
 Calculus of mist
 A clear head
 Standing in the mist, at dawn






Title: Inscriptions on the mist : digital objects management, University of Florida Experience
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00087244/00001
 Material Information
Title: Inscriptions on the mist : digital objects management, University of Florida Experience
Physical Description: Book
Language: English
Creator: Kesse, Erich J.,, 1959-
Affiliation: University of Florida -- University Libraries -- Digital Library Center
Publisher: Erich Kesse
Publication Date: 1999
 Subjects
Subject: University of Florida.   ( lcsh )
Spatial Coverage: North America -- United States of America -- Florida -- Alachua -- Gainesville -- University of Florida
North America -- United States of America -- Florida
 Record Information
Bibliographic ID: UF00087244
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved, Board of Trustees of the University of Florida

Table of Contents
    Introduction
        Page 1
    All dolled up...
        Page 2
    A place to lay my head
        Page 2
        Page 3
    The physic of dream
        Page 4
        Page 5
        Page 6
    Awakening
        Page 7
        Page 8
    Calculus of mist
        Page 9
    A clear head
        Page 10
        Page 11
        Page 12
    Standing in the mist, at dawn
        Page 13
Full Text




riptions
he Mist





Digital Objects Management
university of Florida Experiences
rch Kesse



Inscriptions on the Mist:

Digital Objects Management,
University of Florida Experience

Erich Kesse
1999 May 21


Eudora Welty, the Southern writer, during a writer's conference at Danville
University (Kentucky) in the late-1970s suggested that the most difficult thing
about writing was asking a character to walk across the room. As a southerner,
she explained, she had the gift of "jaw" and could talk aimlessly about this and
that. Information about the common lives of her characters, she implied, were
merely words hung out upon the air, like inscriptions on the mist.

The challenge of those lives was living them. It was enough, she asserted, to
live out her own life. Living those of her characters was enriching but,
sometimes, they seemed unmanageable.

As the writers of our field, we face the same challenge, managing what often
seems the unmanageable. And, the problem we face, too, is keeping our
characters alive and enabling them for action.

I'll use experiences from the University of Florida to direct my comments.






Kesse,Erich. 1999 May 21.


"All dolled up,..."

Where we are now is someplace akin to a dream state. We find ourselves all
dolled up, outfitted in the accouterments of the "new" technology.

We have somewhere to go. We can see it as clearly as the blue sky over an
oracle. And, we can hear the
"All dolled up,... oracle speak though, from time
S to time, it sounds as intelligible
ELECTRONIC JOURNALS as a howler monkey hidden in
PORTABLE ELECTRONIC the forest beyond the mist.
MEDIA
DIGITIZATION It seems we reside in a house
Eleltronc Rese rves with no walls, a "virtual"
Florida Ecosystems & Geology
Florida HEcosystemory & CuGeology house. And, the glass ceiling
Legislat .lve n itih that, time and again, we rise
r agrint, up against, we believe to be a
pre-Copernican sphere, the
crystalline gates of heaven.
This barrier, too, will pass.

Everything seems nearly perfect.

We are in the process of replacing dread microfilm with electronic serials.

We acquire information in floppy, CD and DVD formats that will reside in our
collections along side print, augmenting if not replacing paper.

And, we make daily offerings of digital "media" created through transfiguration
of our own collections, aptly, among them, newspapers and theology --
collections that remain archived in tightly controlled and well regulated vaults of
master microfilms and special collections.

We proselytize; ourselves, converts to "distance learning." Like Christians among
lions, there is security in numbers and the talents of others.


A Place to Lay My Head

How we came to this dream state is inconsequential. At least, in this narrative.


inscriptions on the Mist






Kesse,Erich. 1999 May 21.


"All dolled up,..."

Where we are now is someplace akin to a dream state. We find ourselves all
dolled up, outfitted in the accouterments of the "new" technology.

We have somewhere to go. We can see it as clearly as the blue sky over an
oracle. And, we can hear the
"All dolled up,... oracle speak though, from time
S to time, it sounds as intelligible
ELECTRONIC JOURNALS as a howler monkey hidden in
PORTABLE ELECTRONIC the forest beyond the mist.
MEDIA
DIGITIZATION It seems we reside in a house
Eleltronc Rese rves with no walls, a "virtual"
Florida Ecosystems & Geology
Florida HEcosystemory & CuGeology house. And, the glass ceiling
Legislat .lve n itih that, time and again, we rise
r agrint, up against, we believe to be a
pre-Copernican sphere, the
crystalline gates of heaven.
This barrier, too, will pass.

Everything seems nearly perfect.

We are in the process of replacing dread microfilm with electronic serials.

We acquire information in floppy, CD and DVD formats that will reside in our
collections along side print, augmenting if not replacing paper.

And, we make daily offerings of digital "media" created through transfiguration
of our own collections, aptly, among them, newspapers and theology --
collections that remain archived in tightly controlled and well regulated vaults of
master microfilms and special collections.

We proselytize; ourselves, converts to "distance learning." Like Christians among
lions, there is security in numbers and the talents of others.


A Place to Lay My Head

How we came to this dream state is inconsequential. At least, in this narrative.


inscriptions on the Mist






Kesse, Erich. 1999 May 21.


To paraphrase James Cargill, "It's the bed!" the circumstances outside the
dream state that really matter. Making sense of the dream and use of the digital
objects we acquire or create depends -- if you'll allow me to mix my metaphors -
- on what's in the primordial soup. So much depends upon:

Who's in bed;
What kind of cover we have; and
What kind of cushioning we have for our heads.

For those of you who may not
have realized, I'm reading A place to lay my head
figuratively from Gay Walker's STAFF TOOLS RESOURCES
words on preservation Archives; stems (OS) Storage Media
Preservation; 4
management, which appeared a Systems i ications Storage Facility
are
in print -- published and Florida Canter. Data Recover
for Libra hardware & Services & Vendors
unpublished throughout the Automation Devices Daon Facilles
1980s. These factors read so regio Digilization &






Our staff, if not also levels of staffing, together with their training and
simply without the metaphor. Ele"cVr conorJpn
If policy, like politics, is the art
of what is possible with what
we've got, then surely we must
consider:
Our staff, if not also levels of staffing, together with their training and
ability to learn;
The tools at our disposal for working with or creating digital objects
and electronic artifacts; and
Backup and ability to archive and restore.

Working with digital objects, interpreting the dream, if you will, is much easier if
we can assert a context within which to work.

At the University of Florida Libraries, staff responsible for digital objects include
archivists and collection managers who acquire them, preservation staff who
create and attempt to manage them, systems staff who provide access to them,
and reference staff who assist others in their use.

Policy setting can be a family affair, complicated by a relationship with the
Florida Center for Library Automation (FCLA), which provides multi-institutional
access to bibliographic records and shared digital files; and in turn, on their
relationship with Florida's Northeast Regional Data Center (NERDC) to deliver
and archive these files.


Inscriptions on the Mist






Kesse, Erich. 1999 May 21.


Together we oversee a wide array of resources facilities and services and
tools, both software and hardware.


The Physic of Dream

The metaphor of dream state begins to atomize as it's content comes into view.
Nevertheless, it is cohesive enough to have a familiarity about it.

Fueled by first national recognition in the late 1970s and followed by funding in
the early 1980s, Preservation staff in libraries began to deal with a culture of
paper that generated artifacts with no regard for their longevity, with plans for
their storage but no plan for their maintenance. The culture with which they
dealt was one that celebrated history as of the past rather than in the making.

Before there was a U.S. The Physic of Dream
Newspaper Program, the
news was rarely such after HISTOi a E PAST HISTORY IN THE MAKING
more than a day. With the I ARN NOTYET BORN
analog airwaves shortening Offer incentive option Establish a uniform standard for
ategy production
the average attention to Planl ategy Support across platforms
Mi standard t anandards based
span no more than a ue &verify Non-proprietary
minute, a half-minute and procedures Supported with documentation
copies Documented history of migration
finally a sound bite, was it ks b mne
k* Monitor, aisor th, orjo th
any wonder Andy Warhol, v rocas V
offering reproduction as the
sincerest form of flattery,
extended by five the ten
minutes fame E.M. Forster
earlier had attributed to each individual? We hope we can extend the news
much longer on microfilm than in paper, but in the relative scheme of things any
measure of time is limited. It is occupational hazard, perhaps, to think that the
even shorter life expectancy of electronic media and, now, the much shorter life
of electronic technologies can also be managed.

Extending the life expectancy of artifacts (material, analog or digital) is
honorable, but the truth is that preservation reproduction is reactive and not the
best of models for policy setting, at least, not without establishing control over
artifacts which, as yet, have not come into being. To put this colloquially, the
magician will never be caught in the act if he can distract you with what happens


Inscriptions on the Mist






Kesse, Erich. 1999 May 21.


on stage. But, the lesson may be drawn more clearly from medical economics;
pre-natal (that is, at conception, not simply during creation) care in construction
of digital objects should be more cost effective than acquiring such objects,
together with their attendant ills, "already born".

In this regard, the University of Florida's Smathers Libraries internally promotes
creation of relatively uniform digital objects. Uniformity, in some measure, we
hope should afford us efficiency of treatments, including any future migration,
and mitigation of training costs associated with maintenance.

This uniformity goes unnoticed if not taken for granted by most staff, as the
digital objects they create are bounded by a single software suite, the use of
which is imposed library-wide. The recent adoption of a University computing
standard and base hardware/software requirements together with economic
incentives derived from block buying power drive this policy objective forward. -
We continue to see, however, some platform incompatibilities as faculty,
researchers and students come on board. Fortunately, our new standard is
constructed with some support across platforms. Emulation is still somewhat
toward the foggier edge of the dreamscape.

I wish I could say that the selection of the suite was driven by preservation per
se. It was driven, instead, by Internet software developments, functionality and
price. I'd rather not discuss the advantages and disadvantages of supporting the
Empire of Washington. I would note simply that viability of a format and the
projection of vigor in the marketplace are factors in selection for the exchange of
files within a large community such as the University of Florida.

Separate from mention of the Empire, then, we encourage the adoption of
formats based in standards vetted through bodies such as:
the American and international standards organizations (e.g., ANSI,
NISO, and ISO);
the International Telegraph and Telephone Consultative Committee
(CCITT);
the Joint Photographic Experts Group (JPEG); or
World Wide Web Consortium (W3C)
Or, tested by bodies such as the Digital Library Federation (DLF) or its
members and "allies."
We are clear but not unequivocal in support of non-proprietary standards.

Another factor in the selection of software and formats is the documentation
supporting them. In "recommendation" of proprietary formats, especially, we


Inscriptions on the Mist






Kesse,Erich. 1999 May 21.


also prefer to see a documented history of migration and the free availability of
programming to migrate forward or faithfully read previous versions and
alternate formats. If Adobe, for example, had not provided a free reader or
demonstrated the ability to easily migrate Acrobat PDFs forward from previous
versions, we might not have "purchased" the Elsevier electronic science journals.

In practical terms, we rely upon and have had mixed success migrating to
standard formats. Incentives based in pricing and facilitated communication
have helped. With our own conversion yet underway, we have not attempted to
proselytize beyond our campus.

For official digital objects that we create, state records legislation seeds an
archive strategy. Digital together with paper files are inventoried by type, listed
in a retention schedule, and those marked for permanent retention archived
either on location or in the University Archives. The lack of space in both
locations, the convenience of use in electronic versions, combined with short-
staffing, and a lack of understanding about digital files and retention policy make
this problematic however.

In any case, policy requires the precautions of back up and storage.
Additionally, when databases and graphical file formats are used, administrative
and structural metadata should be documented and saved files verified. I regret
to say, among the Libraries' staff, these policies were not always carried out. It
became routine, therefore, for the Systems Department to encourage staff use of
file servers, which they backup on a daily basis, using a rotation of digital
tapes/cassettes (DAT). Across campus, where file servers administered by a
certified (primarily, on Microsoft or NOVELL) systems officer are unavailable, the
Northeast Regional Data Center (NERDC), which supports campus computing
needs, offers similar encouragement and executes backup routines, likewise,
centrally administered.

By way of admission, I should note that, at the time I was asked to present
University of Florida policy, one of FCLA's servers was crashing. Hundreds of
megabytes of digitized resources, including Electronic Reserves was lost, not to
the crash perse but to the NERDC systems backup operator who, for whatever
other reasons, decided they weren't necessary, the risk of a crash was small.
But, risk management is an aside I'll address later. I want to note that the
Preservation Department lost no files in this crash. As paranoid people, we
maintain backups for backups just as we would for our microfilm masters.

To promote uniformity from among the campus politic, a committee of systems
administrators, marshaled by a campus information officer, whom we lovingly


inscriptions on the Mist






Kesse, Erich. 1999 May 21.


call "the Information Czar," oversees direction and performance. Within the
Libraries, a similar committee of systems liaisons, directed by the Libraries'
representative to the campus committee, deals with the various issues of file
creation, access and use. The members of this committee form the front lines of
assistance not only dealing with hardware, software, and access issues but it is
in their charge to communicate and implement management policy and
procedure. Yet another committee, at the level of the State University System
works under the umbrella of the Florida Center for Library automation on issues
related to the creation, access and maintenance of digital surrogate files, while
another works on issues of electronic resource acquisition. An overlap in
membership affords, while not ensuring, that one hand alone will not be found
clapping.

Policy within the Libraries, at least, charges committee members with what is
termed "professional development." While the charge does not specifically direct
them to study the maintenance of digital objects, the nature of their positions
allows that staff will monitor standards groups if not become active in them.


Awakening

Nightmare is perhaps the most frightening potential of the dream state.
If populated through "family planning," it is also populated by adoption.

Acquisition of resources already born can be frightfully awakening. Bibliographic,
administrative and structural metadata may not exist and might not be
established in the course of investigation.

Part of the Congressional correspondence of Senator and recently departed
Governor, Lawton Chiles, came to us eleven years ago in electronic formats
peculiar to and since replaced several times over by the information services
aiding Congress. Collections such as these are like the newly adopted child who
refuses new clothes for the insular familiarity of the old. Woven deeply into its
fabric, the information contained in such digital objects can be indecipherable
within and inseparable from its media. With the technology that wrote and read
these objects gone, the objects might have been likened to the last of the
Mohegans.

In the interest of time, I'll generalize to describe a practice, of information
collection for acquired files. The practice is not yet uniform. Aspects are
deployed differently, if at all, from unit to unit. Allow me to call the process


Inscriptions on the Mist






Kesse, Erich. 1999 May 21.


"Resource Discovery;" some theologians might describe it as "Awakening." In its
best, most complete practice it is seen among our Archivists, Collection Managers
and Electronic Serials Acquisition Librarians. Like a religious awakening, it's
intent is the collection of information sufficient to revive or execute the digital
object. This information is collected from the source closest to the object's
creation. Sometimes, as in the case of physical format, the object "self-testifies."

Material, bibliographic, administrative, and structural information is or should be
collected. As we expect with analog audio media, we trust this information will
inform our attempts to "depose" the object (that is, to migrate it during initial
execution for evaluation of the file's content). With magnetic media in particular,
one shot is all we may have. Collected information is recorded most often
separate from or in addition to its residence in a given object's file header; and,
the form of record frequently is a database or tracking system of one design or
another.


In the case of the Lawton
Chiles Senate files, the
Archivist managed to have the
Senate computing center
process the files into a
standard text-based format
with searchable descriptive
headers. The files are now
stored in the Tape Library, and
maintained by the Libraries'
Systems Department. And,
they can be mounted and run
on the NERDC mainframe.


Resource Discovery
Collection of information sufficient to revive or execute the digital object
Material, bibliogr ic, administrative, structural information is collected
Tracking Syst
Record of a digital ob aracteristics
Means of scheduling kl refreshing, migration, etc.
Applicable to objects d and created


.-B A,,lp.



^^^*yw


Tracking systems are used to manage the object into the future, both as a
record of its characteristics and the systems required to mount and run them and
as a means of scheduling cleaning, refreshing, migration, etc. The system in use
by the University of Florida Libraries' Preservation Department for digital objects
created through the Libraries' digital initiatives is one derived from another
system that serves to inventory, describe, and manage our large microfilm
holdings.

Beyond the obvious differences of the tracked format, the only differences of
consequence are atomized level (i.e., page, sub-part or character) control and a
much advanced maintenance schedule. The number of data-elements or facets
of information maintained for those digital objects is in no way small. But, in


Inscriptions on the Mist






Kesse, Erich. 1999 May 21.


terms of performance and use, the digital objects tracking system merely seems
to have more descriptive elements of information than does the microfilm
tracking system. While treatments (i.e., storing, inspecting, maintaining,
refreshing, migrating, etc.) of digital objects may be more complex than
treatments of microfilms, the management of treatments does not appear to be
more complex. Systems queue treatments in much the same way; record use
and sales in nearly the same way; and so on.


Calculus of Mist

I will say what should already be obvious, that as the life expectancy of optimally
stored electronic media is shorter than that of optimally stored microfilm, there
are that many more tasks to be performed over time.

Let's quickly look at the mathematics, microfilm compared to digital media. This
comparison could be judged to be "simple." For simplicity, let's say that both
contains a printed volume.

The microfilm, of course, is a series of images on 35mm, gelatin-based, silver-
halide emulsion negative film, post-processed with SilverLock polysulfiding
treatment; and, it has measured density and a variety of other characteristics.


Calculus of Mist
MEASURE MICROFILM CD-ROM MULTIPLE
Life 300 10 +30
c lancec .ears )ears
es

or One Life Time
-ROFI LM
S' Migration = (a 15 minulep assuming bulk processes
CD-ROMn
SMigration = 7 minutes qsuming bulk processes
action over LE or Microfilm = 7 x 30 = 3.5 hours
At


(Aside) I was taught
music on a church
organ. And, the
Gregorian music that
I learned to play was
taught by number,
each a position on a
scale. To this day,
when I hear music, I
see numbers. For
those of you who
may have some
USMARC cataloging
skills, I trust you see,
in this description, a
symphony of
USMARC tagging.


Inscriptions on the Mist






Kesse, Erich. 1999 May 21.


The digital object is
SCalc lus f Mist a directory of
MEASURE MICROFILM CD-ROM MULTIPLE uncompressed, TIFF
tion 56 z-3 21 archive images and
Years ear a subdirectory of
Sof One LifeTime JPEG images (of a
OMNLI1 given DPI, bit-
S ration = (. 15 minutes assuming bulk processes x ge p b-
(300/6)= 12.5hours p depth, color-space,
CD-ROM compression, etc.)
a 7 minutes assuming bulk processes abound within XML-
tiqg over LE of hcrofl 24 x 3, o encoded TEI
S.....markup, together
with all manner of
metadata, on gold-
based CD-ROM,
written at 4x speed,
for Intel-based
architecture. Let us also say that this metadata, for both the microfilm and the
CD-ROM, also is maintained separately from the media for purposes of tracking
and object management.

These descriptions should serve to briefly illustrate the type and nature of
technical information that we are recording.

Let's establish the life expectancy of the media under optimal conditions. Again,
this is illustrative rather than precise. The tables belabor the point. While the
select management processes may take less time for digital objects than for
microfilm, because they must be undertaken more frequently with digital objects,
they ultimately consume more time. Efficiencies in the management process
have to be identified. The University of Florida, traditionally, has done that
through information and, specifically, the automation of information. What the
tables should illustrate is that we expend far more resources now than we ever
have on the maintenance and treatment of objects we hope to preserve.


A Clear Head

Awaking to the dream state, then, we find we have more work than can
seemingly be managed. We find that automation of management processes
build efficiencies that afford us some relief.


Inscriptions on the Mist






Kesse, Erich. 1999 May 21.


At this point, I would like to evoke the medical analogy again as a means of
introducing an approach toward risk management. The analogy is somewhat the
Riddle of the Sphinx yesterday, we referred to the Riddle as "life-cycle
management." I first introduced the concept of pre-natal care, followed by the
ills of the unplanned child. I now add emergency room traumas, though not in
the sense that traumas might be discussed later today.

Early in our history of dealing with and trying to manage digital objects, we
attempted to treat everything. At that time, which I mark by the construction, in
the mid-1980s, of our secure, climate controlled Tape Library administered by
the Libraries' Systems Department and jointly used by the Northeast Regional
Data Center, the formats coming under our control were floppy disk and
magnetic tapes of various sorts. So thorough was the application of preservation
procedure that policy required each digital resource to be "discovered," tracked
for maintenance, and duplicated. Duplicates were frequently passed into service
as they were damaged in use.

Two things stood in our way, forcing change. The first were changes in
copyright legislation and, subsequently, trends toward the licensing of software
and information. We could no longer, legally, duplicate many materials. Even
some migration during acquisition of older, more fragile materials and antiquated
formats fell into doubt. The second change was sheer volume. Publishers had
begun including floppies, eventually CD-ROMs and, now, DVDs in publications.
The Systems Department were overwhelmed. The widely varied nature of the
objects filling out the volume of new acquisitions, also, could not be discounted.
Digital objects were written for mainframe, for Mac, for PC Windows, for various
flavors of UNIX. Systems had become, in a real sense, an emergency room
where treatments based on triage became necessary. The policy of preservation
duplication came to a near sudden death, an exemption granted to select classes
of materials free of copy restrictions. The policy had been slightly misconceived
anyway; collection management policies did not call for this treatment in most
cases. Policies have been revised allowing the application of collection
management policies and the unwritten rule of "uniqueness" for items in archival
collections. Items went into limited circulation or supervised use without
guardian angels to face the curse of the inventive patron.

Death quickly shadowed a policy of Triage. And, death was followed by loss and
mourning. It became policy to replace damaged materials with commercial
replacements where available. But, many damaged materials could not be
saved, revived or replaced. Collection management policy continued to hold
firm. Born commercially digitalis, in many ways, a death sentence. Still, under


Inscriptions on the Mist






Kesse, Erich. 1999 May 21.


a policy of triage, more time
has been freed to afford
A Clear Head concentration on resources
i Collection Management Policy extended we hold more dear.
to electronic media
Co-operative development, Following a chronology of
with centralized management:
access maintenance, and migration events, policy development
shifted to housing. I've
already mentioned the
storage environment of the
Tape Library. It is exacting.
Its temperature is relatively
constant near 60 degrees
Fahrenheit. And, the
relative humidity, controlled
by separate systems, hovers just below 35%. But, for materials that had
entered "circulation," housing issues developed around protective enclosures. It
is now policy, despite the risk of damage and loss, to send digital media
accompanying volumes to the stacks with their volumes. Triage may be a death
sentence, but the bandages applied stave off the inevitable.

(Aside) I encourage you to read between the lines; one of my staff would have
me add that management by theft is a form of management.

Those of us paid to be pessimistic may well be too tired to argue what seems
inevitable. We might rather take interest in "preserving" those few things of
certain value than those that might gain value over time. A volume of Florida
history accompanied by electronic media might be relocated to Special
Collections after 50 years; the volume alone may be readable. This issue
continues to be discussed, as collection managers become spirit guides for the
world of the dreamscape.

Would not like to leave this amble on a dour note, my last words hanging on a
sour breeze. I began describing a policy routed as much in economics as
preservation; it is here that I'd like to end. As book and serial prices increase,
and the prices of electronic media increase with them, the Florida Center for
Library Automation (FCLA) and committees of its member institutions have
planned the purchase, maintenance and migration of jointly owned files. Our
latest policy dictates cooperation and further centralizes the management of
files.


Inscriptions on the Mist






Inscriptions on the Mist


Kesse, Erich. 1999 May 21.


Standing in the Mist, at Dawn

As a body politic, RLG might not deviate from current collection management
strategies. It might continue to divide and conquer, perhaps to a greater degree
of division with responsibility for the "preservation" of particular classes of digital
objects.




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs