Moderation Policies and Strategies of Four News Sites for User Generated Content

Permanent Link: http://ufdc.ufl.edu/UFE0045599/00001

Material Information

Title: Moderation Policies and Strategies of Four News Sites for User Generated Content
Physical Description: 1 online resource (83 p.)
Language: english
Creator: Rollins, Antionette Oshay
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2013


Subjects / Keywords: commenting -- journalism -- moderation -- newspaper
Journalism and Communications -- Dissertations, Academic -- UF
Genre: Mass Communication thesis, M.A.M.C.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Abstract: While traditional forms of interactivity such as letters to the editor were still significant in the 21st Century, the Internet gave the public new ways to communicate with journalists and other readers. One popular interactivity tool was online commenting, which allowed individuals to leave feedback on news stories. However, commenting had drawbacks like incivility, therefore news sites moderated comments using various techniques.   Research was conducted to understand the moderation strategies of online news sites and to determine the issues that moderators considered when reviewing comments. A case study involving interviews with six professionals associated with four newspapers from two media corporations found that the goal of moderators was both to promote civil commenting and to build online communities. News sites simultaneously fostered community and controlled unwanted comments by using crowd-sourced moderation, with all four newspapers letting readers be the first line of eliminating inappropriate comments. Moderators also removed comments that contained personal attacks, hate speech, profanity or libel. These practices served to empower users, promote safe commenting environments and to make comment review a collaborative effort between news organizations and readers.   Although moderators used similar practices to letters editors in terms of determining which comments remained posted, gatekeeping has transformed on the Web. Not only do readers have more control over what comments are presented, but news site moderators said that comments provided reporters with ideas for stories and identified potential sources. These findings can be helpful to educators as they determine ways to incorporate moderation skills into journalism curriculum.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Antionette Oshay Rollins.
Thesis: Thesis (M.A.M.C.)--University of Florida, 2013.
Local: Adviser: Dodd, Julie E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2013
System ID: UFE0045599:00001

Permanent Link: http://ufdc.ufl.edu/UFE0045599/00001

Material Information

Title: Moderation Policies and Strategies of Four News Sites for User Generated Content
Physical Description: 1 online resource (83 p.)
Language: english
Creator: Rollins, Antionette Oshay
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2013


Subjects / Keywords: commenting -- journalism -- moderation -- newspaper
Journalism and Communications -- Dissertations, Academic -- UF
Genre: Mass Communication thesis, M.A.M.C.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Abstract: While traditional forms of interactivity such as letters to the editor were still significant in the 21st Century, the Internet gave the public new ways to communicate with journalists and other readers. One popular interactivity tool was online commenting, which allowed individuals to leave feedback on news stories. However, commenting had drawbacks like incivility, therefore news sites moderated comments using various techniques.   Research was conducted to understand the moderation strategies of online news sites and to determine the issues that moderators considered when reviewing comments. A case study involving interviews with six professionals associated with four newspapers from two media corporations found that the goal of moderators was both to promote civil commenting and to build online communities. News sites simultaneously fostered community and controlled unwanted comments by using crowd-sourced moderation, with all four newspapers letting readers be the first line of eliminating inappropriate comments. Moderators also removed comments that contained personal attacks, hate speech, profanity or libel. These practices served to empower users, promote safe commenting environments and to make comment review a collaborative effort between news organizations and readers.   Although moderators used similar practices to letters editors in terms of determining which comments remained posted, gatekeeping has transformed on the Web. Not only do readers have more control over what comments are presented, but news site moderators said that comments provided reporters with ideas for stories and identified potential sources. These findings can be helpful to educators as they determine ways to incorporate moderation skills into journalism curriculum.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Antionette Oshay Rollins.
Thesis: Thesis (M.A.M.C.)--University of Florida, 2013.
Local: Adviser: Dodd, Julie E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2013
System ID: UFE0045599:00001

This item has the following downloads:

Full Text




2 2013 Antionette Rollins


3 To my parents, thank you fo r all of your love and support


4 ACKNOWLEDGMENTS Although I am the auth are many people who deserve to be acknowledged for their hard work, guidance and support. I would like to thank my Chair Dr. Julie Dodd, who worked tirelessly throughout knowledge and advice helped me tremendously and I have learned so much about research, hard work and perseverance. Thank you for your vision and enthusiasm, Dr. Dodd. My committee members Drs. Judy Robinson and Laurence Alexander also deserve to be acknow ledged. I thank you both for sharing your knowledge and time throughout this process. Your suggestions forced me to think critically and ultimately strengthened this thesis. I would also like to thank Dr. Robinson for contributing to my interest in online journalism through the course Multimedia Blogging. Thank you to all of the interviewees that kindly participated in this research. I learned so much from speaking with each one of you, and I appreciate you for taking time out of your busy schedules to cont ribute. I would also like to thank every professor and classmate that I have had the pleasure of learning from in the College of Journalism and Communications. Thank you for providing spaces that allowed me to think, engage in healthy discourse and expand my ideas about media. Lastly, I would like to thank my parents Yolanda and Charles Williams, my friend Dan Mathis and other family and friends who continuously took the time to listen to my ideas, edit drafts and provide encouragement throughout this pro cess.


5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF FIGURES ................................ ................................ ................................ .......... 7 ABSTRACT ................................ ................................ ................................ ..................... 8 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 10 2 LITERATURE REVIEW ................................ ................................ .......................... 17 Online Comments ................................ ................................ ................................ ... 20 Anonymity in Computer Mediated C ommunication ................................ ................. 22 Comment Moderation ................................ ................................ ............................. 26 Legal Aspect of User Generated Content for News Organizations ......................... 28 Gatekeeping Theory ................................ ................................ ............................... 29 3 METHODOLOGY ................................ ................................ ................................ ... 30 4 FINDINGS ................................ ................................ ................................ ............... 34 Cox Media Group ................................ ................................ ................................ .... 34 The Palm Beach Post ................................ ................................ ............................. 35 User Registration ................................ ................................ .............................. 36 The Moderators and Moderation Strategies ................................ ..................... 37 Moderation Advice and the Future of Online Commenting ............................... 39 The Atlanta Journal Constitution ................................ ................................ ............. 41 User Registration ................................ ................................ .............................. 42 Moderation Advice and the Future of Online Commenting ............................... 43 Dayton Daily News ................................ ................................ ................................ 43 User Registration ................................ ................................ .............................. 44 The Moderators and Moderation Strate gies ................................ ..................... 44 Moderation Advice and the Future of Online Commenting ............................... 45 Tribune Company ................................ ................................ ................................ ... 47 Orlando Sentinel ................................ ................................ ................................ ..... 47 Comment Policy ................................ ................................ ............................... 49 User Registration ................................ ................................ .............................. 49 The Moderators and Moderation Strategies ................................ ..................... 49 Moderation Advice and the Future of Online Commenting ............................... 53 5 DISCUSSION ................................ ................................ ................................ ......... 54 The New Gatekeeper ................................ ................................ .............................. 55


6 Building an Online Community ................................ ................................ ................ 57 Crowd S ourced Moderation ................................ ................................ .................... 59 Anonymity in Online Commenting ................................ ................................ ........... 60 The Role of Social Media in Online Commenting ................................ ................... 61 Suggestions for Future Research ................................ ................................ ........... 62 APPENDIX A COX MEDIA GROUP VISITOR AGREEMENT ................................ ....................... 65 B ORLANDO SENTINE L TERMS OF SERVICE ................................ ....................... 68 C THE PALM BEACH POST LETTERS TO THE EDITOR POLICY .......................... 71 D THE ATLANTA JOURNAL CONSTITUTION LETTERS TO THE EDITOR POLICY ................................ ................................ ................................ ................... 72 E DAYTON DAILY NEWS LETTERS TO THE EDITOR POLICY .............................. 73 F ORLANDO SENTINEL LETTERS TO THE EDITOR POLICY ................................ 74 G INTERVIEW QUESTIONS ................................ ................................ ...................... 75 H SAMPLE EMAIL ................................ ................................ ................................ ..... 76 REFERENCES ................................ ................................ ................................ .............. 77 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 83


7 LIST OF FIGURES Figure page 1 1 Online reader c omment ................................ ................................ ...................... 15 1 2 Online r eader c omments ................................ ................................ .................... 16


8 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Master of Arts in Mass Communication MODERATION POLICIES AND STRATEGIES OF FOUR NEWS SITES FOR USER GENERATED CONTENT By Antionette Rollins May 2013 Chair: Julie Dodd Major: Mass Communication While traditional forms of interactivity such as letters to the editor were still significant in the 21 st Century, the Internet gave the public new ways to communicate with journalists and other readers. One popular interactivity tool was online commenting, which allowed individuals to leave feedback on news stories. However, commenting had drawbacks like incivility, therefore news sites moderated comments using various techniques. Research was conducted to understand the moderation strategies of online news sites and to determine the issues that moderators considered when reviewing comments. A case study involving interviews with six professionals associated with four newspapers from two media corporations found that the goal of moderators was both to promote civil commenting and to build online communities. News sites simultaneously fostered community and controlled unwanted comments by using crowd sourced moderation, with all four newspapers letting readers be the first line of eliminating inappropriate comments. Moderators also removed comments that contained persona l attacks, hate speech, profanity or libel. These practices served to empower users,


9 promote safe commenting environments and to make comment review a collaborative effort between news organizations and readers. Although moderators used similar practices t o letters editors in terms of determining which comments remained posted, gatekeeping has transformed on the Web. Not only do readers have more control over what comments are presented, but news site moderators said that comments provided reporters with id eas for stories and identified potential sources. These findings can be helpful to educators as they determine ways to incorporate moderation skills into journalism curriculum.


10 CHAPTER 1 INTRODUCTION The Internet significantly impacted the daily lives of Americans in the 21 st Century by allowing people to complete activities such as shopping, conducting business, communicating a nd consuming media on the Web. The Internet also made it possible for everyday people to produce, publish and disseminate media with greater ease than previously seen before. This user generated content could be seen in the form of Facebook posts, weblogs and videos. User generated content can be defined as the content that an everyday individual publishes such as comments, videos and photos (Hermida, 2008). The Web also impacted journalism, as an increasing number of newspapers began to have an on line presence where journalists could publish content specifically for the Web, make print stories available on websites and interact with readers in new ways. Interaction between journalists and the public transitioned from established means such as lette rs to the editor and radio phone ins to electronic communication through emails and message boards (Wahl Jorgensen, 2007). One of the most popular ways to promote interactivity on news websites during this time was through the comment section, which allowe d users the opportunity to give feedback to journalists and create discourse with other users. This exchange of ideas and public discourse sometimes developed into an online community where users and journalists felt free to share, communicate and learn fr om each other. Commenting allowed users to contribute their voices to local, national and global news. S tories that garnered a great amount of media attention also received many comments. For example, a May 2012 article posted on C


11 profile Trayvon Martin case, which involved a Black teen who was murdered in Florida by a man who claimed the killing was in self defense, received 8,709 comments (Figure 1.1; Figure 1.2) by February 2013 (Botelho, 2012). Anothe r article about the case posted on website in March 2012 had received 336 comments by February 2013 (Alcindor, Bello & Copeland, 2012). While commenting increased the interaction between users and journalists and thought provoking and healthy d iscourse existed, some members of the online community left hateful, and sometimes irrelevant, remarks. These comments may have also contained libelous language that could put the publication in legal trouble. Many publications adopted moderation policie s in an effort to control the content of user comments. Moderation, which are strategies aimed at combating such comments, took many different forms including pre moderation and post moderation (Diakopoulos & Naaman, 2011). This is similar to newspapers de veloping submission policies for letter writers, however these submission guidelines tend to focus more on procedural rules like word count rather than incivility and profanity (Wahl Jorgenson, 2007). For example, letter submission poli cy states the following : Letters to the editor should only be sent to The Times, and not to other publications. We do not publish open letters or third party letters. Letters for publication should be no longer than 150 words, must refer to an article that has appeared within the last seven days, and must include the writer's address and phone numbers. No attachments, please. We regret we cannot return or acknowledge unpublished letters. Writers of those letters selected for publication will be notified wit hin a week. Letters may be shortened for para. 1 3). While letter submission policies typically address ed these types of rules, letter editors still employ ed een publishable and unpublishable Jorgensen, 2007, p. 68). Despite the medium, these


12 principles were similar to the principles that moderators at news websites used in the 21 st Century. In November 2012, the North Carolina newspaper The Robesonian announced According to editor Donnie Douglas (2012), the website would no longer allow individuals to post racial comments on stories that did not have a racial element. Douglas wrote that those at the paper knew it was time to address the issue of racial comments when they began to receive an influx of complaints from readers. While The Robesonian still allowed comments about race on articles that had a directly racia l component, those comments were required to be related to the story and in good taste. We understand our obligation to report the news, but we are not obligated to provide a forum for bigotry and hatred. We are a member of this commu nity, and allowing this kind of divisiveness does not benefit anyone. Douglas 2012 para. 9). The Robesonian was not the only newspaper to take action regarding comments on their website. The Gazette ( Iowa Ci ty Iowa) turned off commenting on all of their invited to submit pre website. Opinion Page Editor Jeff T ecklenburg (2012) wrote on the s website: Yes, thegazette.com has changed the rules for online commenting again, as announced Sept. 2. For good reason. Over the past few years, Gazette managers have revised the commenting process several times in h opes of encouraging a more civil, informed and broad based online discussion. working on it (para. 1 3).


13 National newspapers such as The Washington Post and The New York Times also re vised pubic discussion (Perez Pena, 2010). In addition to moderating comments, publication s could also disallow or limit commenting on certain topics like The Robesonian and Gazette or use tactics, like requiring those who wanted to post comments to register on the news site or installing social media plugins that required those who commented to first sign on to a social media acc ount that they had, such as Facebook As the 21 st Century progressed news sites experimented with different approaches that would simultaneously decrease the occurrences of inappropriate comments and promote community engagement. A case study examining th e moderation practices of four widely circulated publications was completed to gain a better understanding of how online news sites moderate user generated comments. This case study included interviews with people who moderated comments and manage d moderators. While there has been an increase in research on user generated content and online comments, at the time of this study literature was lacking that detailed the process that professionals at news sites in the United States undertook in an attempt to handle comments In fact, as of February 2013, neither the Society of Professional Journalists nor the American Society of Newspaper Editors had established guidelines or best practices for addressing online moderation of comm ents on news sites However, the American Society of Newspaper Editors did survey newsrooms in 2009 and found that 87.6% of the 267 surveyed newsrooms invited readers to post online comments on their news sites ( Ke yes, 2009).


14 While past research has explored the substance of user generated content, one purpose of this study is to present journalism educators with methods that their students may use in their future jobs as editors, moderators and community managers. This case study sought to answer the following questions: R1: How do news websites develop comment moderation policies? R2 : What are the components of online comments policies developed by news organizations? R3 : What strategies do news organiz ations use to moderate user generated comments? R4 : Who is involved in moderating comments? R5 : What issues do journalists consider while moderating comments?


15 Figure 1 1 Online reader c omment


16 Figure 1 2 Online r eader c omment s


17 CHAPTER 2 LITERATURE REVIEW st Century cannot be understated. As a testament to this impact, most news publications by 2009 had an online presence (Gunter Campbell & Touri, 2009). Along with new ways of storytelling and distributing news 1 the Web gave the public more opportunities to interact with online news media, n ews sites are increasingly becoming places for communities to Naaman, 2011, p. 1). While the Web ushered in new opportunities for interactivity, the press has always had a connectio n to community discourse. dependency with the public, who are their audiences, their s Jorgensen, 2007, p. 4). Wahl Jorgensen studied this relationship and examined letters to the editor in her book Journalists and the Public (2007). In many ways, letters to the editor can be seen as a precursor to online comments, as both letters to the editor and online comments allow members of the public to discuss articles with and voice opinions to journalists and other citizens. Wahl ne of a few arenas for public discussion to have survived throughout a large period of the history of 1 New storytelling approaches and ways t o distribute news emerged in the 21 st Century, particularly due to the rise in social media. For example, journalists used social media sites such as Twitter to disseminate stories they had written (Holcomb, Gross & Mitchell, 2011).


18 (2007). Although letters to the editor give readers the opportunity to com ment on news, it is important to recognize that these letters typically do not represent the total readership of a publication or population of its community. L etters to t he editor are moderated, as the letters that appear in the printed news publicati on are selected and published by section shape public debate, and their decisions and work practices can therefore tell us Jorgensen, 2007, p. 55). Wahl Jorgensen (2007) interviewed 23 San Francisco Bay editors and found that these professionals viewed the letters s ection as something that should be an open 154). However, these editors also said that they sometimes needed to reject certain violated principles of civility. Uncivil letters that contained libelous personal attacks were most likely to be rejected by these San Francisco Bay editors (Wahl Jorgensen, 2007, p. 154,). Despite editor s having to moderate some uncivil letters, Wahl Jorgensen found that while letters to the editor are subjec tive it was not because of these moderation practices. Instead, letters to the editor may be biased because they are not representative of the popul ation (Wahl Jorgensen, 2007). Hart (2001) found that letter


19 found that letter writers wer American (p. 55). Wahl useful framework for evaluating how decisions are made regarding moderation of comments made t o online news stories. While journalism has a long standing relationship with public participation, the Internet has allowed people to interact with media and each other in new ways. Rafaeli relate d construct about Interactivity places shared interpretive contexts in the primary role. Interactivity describes and prescribes the manner in which conversational interaction as an iter ative process leads to jointly produced meaning. Interactivity merges speaking with listening And it is a general enough concept to encompass both intimate, person to person, face to face communication and other forums and forms (Rafaeli & Sudweeks, 1997, para. 10). Many forms of interactivity could be found throughout the Web in the early part of the 21 st Century. One form was the bulletin board, or message board, which was an on line gathering place that allowed people to have a con versation even if they were not in the same place at the same time (Bishop, 2009, p. 6). Interactivity also occurred through weblogs 2 social media 3 and Wikis 4 (Chen, Xu & Whinston, 2011). 2 A weblog, or blog, i 3 based services that allow individuals to (1) construct a public or semi public profile within a bounded system, (2) articulate a list of other users with whom they share a connection, and (3) view and traverse their


20 Online Comments One of the most popular interactivity tools that publications were usin g in the 21 st Century was the comment section ( Domingo et al. 2008). One of the first American newspapers to allow user generated commen ts on same page articles was The Rocky Mountain News (Denver, Colorado) in 1998 (Santana, 2011). Following The Rocky M decision to allow online comments the number of news sites with comment sections grew tremendously. In 2010, 92% of the top 150 newspapers in the U.S. allowed users to post online comments (Santana, 2011). Faridani, Bitton, Ryokai and Goldbe of websites increasingly provide d feedback usually in the form of comments (p. 1175). Through comments, the public voiced their opinions on an article, created discussions with o ther users and even interacted with journalists. Many sites allowed the public to leave comments using an online form at the bottom of an article, which may or may not have required users to register with the site (Hermida & Thurman, 2008). Comment section s were also extremely popular among user s In 2011 The Huffington Post, a popular political blog, received approximately three million comments per month (Diakopoulos & Naaman, 2011). Overall, 25% of Internet users reported that they had commented on eithe r a news story or blog post (Purcell, Rainie, Mitchell, Rosenstiel & Olmstead, 2010). list of connections and those made by others within the system. The nature and nomenclature of these connec 4 editing tool in which users may add or edit content directly through their we


21 Although these comment sections promoted public discourse, commenting had flaws. Faridani et al. (2010) explained some of the common problems: First, thoughtful moderates are often shouted down by extremists. Online discussions, conducted through threaded lists of comments, often end in data can be overwhelming. News stories and blog posts often gen erate hundreds or thousands of comments. As the number of comments grows, presenting them in a chronological list is simply not a scalable interface for browsing and skimming. Third, many websites tend to attract people with like minded viewpoints, which c an reinforce biases and produce Discussing the moderation of comments requires determining the quality of intelligence and normatively include s notions of accuracy, reliability, validity, currency, Low quality comments can be in the form of attacks on the publication and other users, hate speech and libelous statemen ts. Singer (2011) found that while journalists have a n unenthusiastic view of user comments because of their undesirable content, many journalists find comments useful as online comments 2011, p. 76). However, because of the vast amount of comments that a particular article or topic may receive, journalists may have a difficult time separating quality comments from low quality ones (Braun & Gillespie, 2010). Singer and Ashman (2009) stated that while interactivity can be viewed as positive, the increase of user generated content has changed the process of news, giving journalists less control. The researchers also pointed out that user generated content brings in a host of practical and ethical questions that practicing journalists must address. Such questions include the following :


22 What might an optimal relationship between journalists and users/contributors look like, and what are the challenges to achi eving it? If the content space is shared, is responsibility for the content itself also shared? Who decides what is credible, true, or even newsworthy in the first place? What happens to the prized journalistic norm of autonomy in this environment? (p. 4). Anonymity in Computer Mediated Communication Another question that journalists had to answer in this digital age was how to deal with the low Naaman (2 011). One reason that people could feel inclined to post low quality or uncivil comments is because they believe that other users will not be able to identify them because their comments were made anonymously. others to large social context, such as a crowd, or in smaller context, such as two person 3040). Previous rese arch on computer mediated communication (CMC), communication over the Internet, has shown that it differs from face to face (FtF) communication because of the lack of verbal and nonverbal social cues (Coffey & Woolworth, 2004; Walther, 1992). One obvious appearance appearance is obvious, it is assumed that one could potentially be completely free of these physical cues in CMC. It is a well established finding that physical appearance is an important cue in social interactions (Christopherson, 2007, p. 3045). P hysical cues include body language that signifies discomfort or verbal responses (Coffey & Woolworth, 2004). Coffey and Woolwort h (2004) found that the lack of social cues coupled with virtual distance and anonymity may result in computer


23 mediated communication that is more angry, extreme and even racist than face to face communication. Although computer mediated communication has some drawbacks, it allows people to communicate with others that are separated by space and time, create and maintain social and professional contacts and encourages a greater sense of commonality between people than face to face communication (Hardacker, 2010, p. 223). Even with these positives, anonymity on the Internet makes ensuring quality comments a challenge ( Chen et al 2011, p. 238). Santana (2012) conducted a study that tested the effects of anonymity o n civility in user comments made on online n ewspapers. Through a content analysis of both anonymous and non anonymous comments made in response to a racially charged topic (immigration) and to a non racial, yet controversial topic (the U.S. Tea Party movement), Santana found that a story with a raci al aspect a non racialized one and that removing anonymity elevates the level of d (Santana, 2012, p. v). C ommenters posting about immigration were significantly more likely to write an un civil comment than commenters posting about the Tea Party (Santana, 2012, p. 72). Of the 450 analyzed anonymous comments regarding immigration, 53.3% were area were comments that invoked disparaging sentiments on the basis of race/ethnicity (16.1%), xenophobia (15.4%), name calling (14.8%), racist or bigoted sentiments majority of th ese uncivil commenters made emotional appeals rather than posting facts


24 and figures. Of the anonymous comments about the Tea Party Movement, Santana found that 35.6% of the 450 comments were uncivil. In terms of non anonymous comments, findings showed that of the 450 analyzed comments on immigration, 44% of the comments were civil while 28.7% were uncivil and 27.3% were unclear. Santana (2012) found that anonymous commenters were significantly more likely to post uncivil comments than non anonymous commente rs. Civility in online discussion is significant because it may affect the way information is interpreted. Hwang (2008) conducted an experiment using a simulated online discussion to examine incivility in online chatting. Participants in this experiment b elieved that they were in an online discussion session. Results showed that discursive mindedness and had an overall unfavorable attitude of ot hers who did not share their views (Hwang, 2008, p. 90). Uncivil comments may also have an effect on the way a journalist gathers information. While interviewing journalists about civility in online comments, Diakopoulos and Naaman (2011) found that some r eporters were having a difficult time obtaining sources for their stories as some individuals feared that online commenters might criticize them if they were identified as a source in a story. While commenters may be uncivil or post low quality comments because they truly believe in the validity of their opinion, others may post simply to start an argument or make others uncomfortable, an Internet phenomenon known as trolling entails luring others into pointless and time (Herring, Job


25 Sluder, Scheckler & Barab, 2002). Hardaker (2010) found that trolling is primarily made up of aggression, deception, disruption and success. It seems clear that part of the human condition is to find a degree of entertainment in conflict, wh ether in the form of high risk sports, action films, violent computer games, or linguistic aggression in television pro grams (Culpeper 1996, 2005; Culpeper et al. 2003; Bousfield 2008). However, unlike these situations where the individual typically only watches or simulates conflict, online, with the protection of anonymity and distance, CMC users can exercise aggression against other real humans, with little risk of being identified or held accountable for their actions (Hardaker, 2010, p. 238). Another issue that journalists had to confront on news websites was arcia Molina, 2005, p. 1). Spam could be posted with ( Rajendran & Pandey, 2012, para 2). With the increase of blogs and other websites that encouraged user commenting, a new type of spam known as comment spam had emerged ( Bhattarai, R us & Dasgupta, 2009). I nature to facilitate commentators to write their opinions about the piece of writing. Along with comments, people would also like to link the article with other posts in the blogosphere to express their opinions in relation to these posts. This results in links betw een pages of a blog or different blogs. While this kind of link can be a legitimate hyperlink to relate two different posts, spammers exploit this concept to increase their link weights by posting random comments and links in blogs (Bhattarai, R us & Dasgup ta, 2009 p. 37 ). In attempts to address negatives such as incivility, trolling and spam, many journalists began to call for the end of ano nymous online commenting toward the end of the first decade of the 21 st Century (Santana, 2011). The policy of requir ing those


26 posting comments to in some way register or identify themselves follows print journalistic practices, as anonymous letters to the editor and the use of anonymous sources is generally avoided (Santanta, 2011). One increasingly popular process used to limit anonymity was a plugin created by the social network Facebook that allowed other sites to install the Facebook plugin and required users to log into their Facebook accounts before posting a comment. Other news websites will not allow a user to co mment unless the user registers for an account with the online news site using his or her real name (Santana, 2012). Comment Moderation Whether allowing anonymous comments or not, many news organizations by 2013 started to review user content in an effort to prevent and discourage low quality comments. In a study examining how various United Kingdom news sites handle d user generated content, Hermida and Thurman (2008) found that some editors were concerned about the potential that user generated content ha d to damage the brand of their organizations; therefore, these editors moderated comments. Comment moderation can take many forms. Diakopoulos and Naaman (2011) identified two strategies that news organizations employed when addressing comment quality. P ublications can moderate by asking users to report (also referred to as flag ) comments that violate website guidelines for posting comments or comments that were abusive. If a particular comment received many flags then that comment could be deleted by the publication. A moderator could then leave a note to other users that the comment had been removed due to flags it received (Santana, 2012). This type of


27 crowd sourced moderation 5 empowers users by giving them a sense of ownership in the online community a nd makes moderation a collective activity (Briggs, 2010, p. 49). News websites can also moderate comments, either before or after they are posted, in house or by outsourcing (Diakopoulos & Naaman, 2011). While outsourcing can be beneficial, especially on controversial articles that may generate a high volume of user generated comments, Diakopoulos & Naaman found that some editors might not trust the outside company to implement journalistic standards (2011). While the moderation of comments varies by news paper, comments often post immediately, though auto filters generally disallow vulgar language. In this way, unmoderated comment boards might tend to have more acerbic language than those cleaned up and closed by moderators. Still, even after the boards ar e moderated and closed, entire discussion sections can and do disappear; at many Gannett newspapers, online stories and the accompanying comment forums 30 days (Santana, 2012, p. 3). News organizations may be encouraged to moderate comments because moderation has been found to have an effect on the quality of user generated comments. Chen, Xu and Whinston (2011) found that moderation systems directly affect ed ion and could improve the quality of comments (p. 237). Furthermore, those who post comments that are being moderated may contribute quality comments in order to build up a positive reputation. The researchers found that moderation can have an impact on po frequency of moderation is critical and should be properly chosen for optimal 5 self moderation (crowd sourced moderation) is seen on


28 Legal Aspect of User Generated Content for News Organizations While this research does not have a legal focus, it is important to mention the law as it pertains to online comment moderation. The United States Congress passed the Communications Decency Act (CDA) in 1996 in an attempt to regulate obscenity on the Internet (Ehrlich, 2002). Though m ost of the CDA was ruled unconstitutional because it immunity from suit to ISPs [Internet Service Providers] and other interactive computer services for content origin Section 230 protects websites from liability if a user posts a comment that is defamatory or illegal, a moderator may be considered a content provider if a comment is edited in any way, in which c ase the moderator may not be protected from suit (Rich, 2012). Many online publications post the legal issues that are related to user generated content on their site, usually accessible through a link at the bottom of each page (Ruiz, Domingo, Mic, Daz Noci, Meso & Masipp, 2011). The text is a contract with the corporate entity publishing the news portal that users implicitly accept when they access the website. The fact that the legal rules are public, as the participation guidelines are, means that u sers cannot argue they were not aware of them ( Ruiz et al., 2011, p.474). Wahl Jorgensen (2007) found that letters editors also considered certain Jorgensen, 2007, p. 70). In addition, Wahl Jorgensen (2007) found that letter editors rejected letters that may have resulted in libel suits (p. 87).


29 Gatekeeping Th eory When media professionals at online publications moderate user comments, they are acting as gatekeepers. Gatekeeping theory, as it relates to mass communication, refers to the process that information goes through before it is presented to the public. continued or halted as they pass along news channel from source to reporter to a series The analogy of the ga tekeeper in mass communication was originally introduced by social scientist Kurt Lewin, who suggested that news items first travel through certain and the role of journal ists as gatekeepers were then popularized by David Manning White, who conducted a case study of the news selection of a newspaper editor in 1950 (Shoemaker et al., 2001). This study will use gatekeeping theory to explain how moderators at news websites p lay an important role in the public discourse on comment sections. Specifically, those involved in implementing comment moderation policies select which voices get heard and determine which voices are silenced. User generated content may go through a serie s of filters before it is posted online for the public, while content that is deemed uncivil or low quality may be deleted before or after it is posted.


30 CHAPTER 3 METHODOLOGY In order to determine the answers to the research questions, the researcher decided to conduct interviews with individuals at news org anizations who were involved in the development of and implementation of comment moderation policies. An interview approach was selected for collecting information about comment moderation as o pposed to a survey, because the interview process would al low for more explanation of how and why certain procedures are used by news sites. The interview questions ( Appendix C) were info rmed by prior research on interactivity, user generated content and moderation and developed after an examination of the community, or user, guidelines for all newspapers. This questionnaire consisted of 11 primary questions and additional follow up questi ons based on responses Q uestions were tailored for each interviewee based on their job description and role in moderation. Each participant was originally contact ed through an email letter ( Appendix D), which included information about the researcher, th e study and details about how the study would be conducted. Some participants responded indicating that they were willi ng to be interviewed, while some responded referring someone else who would be better suited to participate. The referral system was also used after some of the initial interviews were conducted, as some interviewees passed along information about this study to others who were eventually interviewed. This resulted in the study being expanded to newspapers in Georgia and Ohio that were part of the Cox Media Group


31 This study initially focused on three newspapers in Florida that were part of regional or national news organizations. The goal was to select newspapers that had posted online user policies and that would be identified with a certai n geographical community. One of the purposes of this research was to learn about moderation policies that could be an aid to news organizations that are developing moderation policies. T herefore it was determined that the size, community and resources of these newspapers would be more relatable than a larger news site that received more d aily traffic and user comments, like USA Today or The New York Times This study originally focused on newspapers in Florida, as the research was taking place in Florida a nd the researcher had contact information for journalists at several Florida newspapers. An application for the Institutional Review Board was completed to contact the aforementioned journalists, however the board determined that approval was not needed to conduct this research. The interview process initially started with The Palm Beach Post The Orlando Sentinel and one other major newspaper in Florida. The interview with the online community manager at The Palm Beach Post led to an interview with Cox M edia Group which led to interviews with The Atlanta Journal Constitution and Dayton Daily News Multiple attempts, through email and phone calls, to schedule interviews with potential sources at the third major newspaper were unsuccessful, so that newspap er was eliminated from the study. All interviews were conducted by telephone between January 2, 2013, and February 5, 2013. Each interview was audio recorded and later transcribed. Additional


32 follow up questions were asked through telephone and email commu nication. In total six individuals were interviewed. Limitations One factor that limited this study was the lack of institu tional memory. Because some of those interviewed were new to their positions or had only been with the news organization for a sho rt period of time, they were unable to answer c ertain questions about comment moderation Notably, there was an interviewee who had come into the position two weeks before the interview was conducted and another interviewee who joined the company five mont When Another limitation was anonymity as some participants did not want have their names published because they could potentially receive negative responses from their readership. Four interviewees did not wish to have their names mentioned in this research, so every participant is listed by job title for the sake of consistency. One participant in this study did not want to inform readers about specific moderation procedures, therefore responses about certain processes were not detailed. Additionally, the interviews conducted fo r this study were not uniform in terms of asking all the pre planned questions and, in some cases, not as in depth because of the time limitations of the individuals interviewed. Factors included sched uling conflicts and illnesses. This study originally sought to unde rstand how comment policies were developed. However, requests for interviews from individuals who developed the


33 policies received no response. Therefore, the researcher relied on information from those who were interviewed. Because of this some important q


34 CHAPTER 4 FINDINGS In order to determine the policies comments and the process used in moderating comments, six individuals were interested who had responsibility for online moderating Those individuals were employed at The Palm Beach Post The Atlanta Journal Constitution, The Dayton Da ily News and The Orlando Sentinel at the time of the interviews, which took place in January and February 2013. Cox Media Group Cox Media Group, Inc., a Cox Enterprises media company, owned television broadcast stations, a cable news channel, radio sta tions, digital services and newspapers in 2013. In 2011 the company owned eight daily newspapers and 16 non daily papers. Cox owned newspapers included The Palm Beach Post The Atlanta Journal Constitution and T he Dayton Daily News Former Ohio governor and United States presidential nominee James A. Cox founded Cox Enterpr ises, Inc. in 1898 in Dayton, Ohio acquisition of several newspapers, radio stations and cable systems, Cox Enterprises became one of the lar gest broadband communications companies in the United States during the 21 st Cox developed the visitor agreement what they called their commenting policy, which 2013. The Social Med ia Manager/Strategist, who had been in this position for over three years, provided information about the Cox comment policy. The Social Media Manager/Strategist helped build the content management system that Cox websites


35 used for commenting in 2013 and a dvised Cox properties about ways to utilize social media and engage the online community. The Social Media Manager/Strategist previously worked as an Internet producer at The Palm Beach Post and worked in news for approximately nine years. Comment Policy A ccording to the Social Media Manager/Strategist, Cox developed the visitor agreement that was used by all the Cox properties and posted on Cox property websites in 2013. Cox Cox affiliate had adopted the terms of the user agree m ent ( Appendix A). Cox developed the agreement, but moderation of the sites was left to the discretion of the i ndividual newspapers. Individual Cox newspapers also had their own letters to the editor policies ( Appendix C, Appendix D, Appendix E and Appendi x F). A ll Cox properties ran o n the same commenting system in 2013 The Social Media Manager/ Strategist at Cox said that C ox was centralized in the sense that all of the news sites use the same commenting system and those at the corporation encouraged its news properties to allow commenting and moderating but c ommunity management, which involves overseeing user participation (Braun & Gillespie, 2010), was something that individuals at the newspapers performed themselves. The Palm Beach Post In 2013, The Palm Beach Post was a Cox owned newspaper based in West Palm Beach, Florida The paper was founded in 1908 as a weekly originally named The Palm Beach County. In 1916 The Palm Beach County became a daily known as The Palm Beach Post


36 Cox purchased a numb er of Palm Beach and West Palm Beach, Florida, newspapers in 1969. A decade later, the newspaper The Palm Beach Times was renamed The Evening Times and in 1987 merged with The Post forming The Palm Beach Post The information about T he Pal m moderation of online comments was obtained through a telephone interview with the Digital Manager, who held this responsibility in 2013 was overseein g T he Pal m online content. The The Palm Beach Post included Online Editorial Director and Assistant Online Director. The Digital Manager had over 15 years of news experience. As of 2012, The Pal m Beach Post reached an average of 142, 679 readers with its Sunday edition, 88,231 readers Monday through Friday and a readership of 89,335 with its Saturday edition ( Alliance for Audited Media, 2012) According to the Digital Manager at The Palm Beach Po st The Palm Beach Post started to allow commenting shortly after the website was launched in the early 2000s. The Digital Manager was unable to recall the exact date. User Registration In order for someone to comment on website dur ing the time of this study, he or she must have registered an account on The Palm Beach website, using a valid email address and agreed to the visitor agreement One also had to register an account in order to report a comment, which meant that a reader found a comment posted by another reader to be inappropriate. Those reported comments were the ones that the Digital Manager would review to determine if the


37 comment should be removed. By requiring an individual to register in order to report a comm ent, the same individual could not report the same comment multiple times. Howeve r, the Social Media Manager/ Strategist acknowledged that a drawback of the system was that someone could create multiple accounts with different email addresses. Site registra tion rules had changed six months before this research was conducted. Prior to June 2012, users did not have to register an account in order to comment on a news story or report a comment that violated the visitor agreement. The Social Media Manager/ Strate gist said that prior to June 2012 there was no way to trace what user report ed what comment Because there was no way to verify an email address, The Palm Beach Post decided to disable commenting on crime stories. Digital Manger: egistered; you could just pop in there anonymously and leave comments. So on a lot of crime stories we would not offer commenting at all by default because those would tend to turn into ugly discussions. Once we put the registration requirement in place, n ow we have the ability that if somebody is them and they would have to reregister with a different email address, and it becomes harder for them to do that on a consistent basis. The Digital the implementation of the registration requirement. The Moderators and Moderation Strategies The Palm Beach Post did not have full time moderators at the time of this study. The Digital Man ager explained that moderating comments was a shared task that any staff member performed during their shift members who work on the website would review flagged comments when there was free time during their shif t.


38 According to the Digital Manager, The Palm Beach Post paper did not pre screen comments every user comment automatically appeared on the site when posted by the reader The paper used a type of crowd sourced moderation in which users reported, or flagged, comments that they believed violated the visitor agreement Every comment on the to the paper For example, a user could flag a comment that he or she thought was offensive. Once a comment was deleted, a message appeared in place of the original still be present on the site Community moderation helped those at the newspaper better handle the vast amount of comments the site received. The staff of The Palm Beach who reviewed the online comments referred to the visitor agreement to determine whether a flagged comment should b e removed or not. Those at the paper found that some people would report a comment simply because they did not agree with it, not because was objectionable. This is the reason that once a comment was reported it stayed visible on the site. I t was also sen t to a queue at the administration end of the site where moderators could review it and remove it from the site if it was judged to have violated the visitor agreement If the comment was deemed appropriate then the comment it stayed on the site. While a moderator reviewed queued comments periodically throughout the day, there could be times when comments could not be reviewed immediately. In 2013, The Palm Beach had a system in place where comments that received five individual flags were automatically re moved from the site. Once a certain comment was removed from


39 posting on the website, moderators were still able to review it to determine if it violated t he visitor agreement. If the comment was determined by the moderators to not violate the visitor agree ment, the comment could be republished. de scribing unwanted comments ( Appendix A). Although moderators did not receive any official moderation training, the Digital Manger said that det ermining whether a comment violated the visitor agreement was often simple. These types of comments included those that contained profanity, personal attacks or were overtly racist. In incidents where a moderator was unable to make a determination about wh ether or not a comment violated the visitor agreement, the moderator would go to the Digital Manager or another editor for advice. For example, The Palm Beach Post received a large number comments on various stories that expressed disapproval over immigrat ion laws and a moderator may have had difficulty in determining whether the language in these comments violated the commenting policy. The Digital Manager said having a debate about immigration on the site was appropriate but racist language was not allo wed. Digital Manager: an editor and sort of bounce it against us and ask for our opinion. The guidelines have kind of been shaped in that way over time; for what can pass and what cannot. In general they will come to an editor for a judgment call if it s Moderation Advice and the Future of Online Commenting The Digital Manager said that any news organization that u ses cr owd sourced moderation should never go half of the day without reviewing flagged comments, as the conversation cold turn from the news article to the lack of moderation. The Digital Manager said that newsroom budgets might not allow funding to allocate som eone to


40 constantly moderate comments, but the Digital Manger emphasized that user comments are something that must be addressed regularly throughout the day. Both the Digital Manager and t he Social Media Manger and Strategist had a favorable view of reade rs commenting and the comm unity environment that reader comments brought to the site. Digital Manger : You can even get sources out of the comments or news tips out of read something in the c omments that leads them to another ending to that story or follow wind up keeping for a long time through the comments. The Digital Manager also said, doe The Social Media Manager/ Str ategist and Digital Manager both supported the use of online anonymity. Both interviewees said that a user may have legitimate or reasonable view s, but may not want to express those v iews in a public forum using a real name. Examples given were business owners not wanting customers to know their political views or an employee not feeling comfortable with his or her employer reading certain opinions. The Social Media Manager/ Strategist also said that certain members of society, particularly women, young people and those with safety issues, may feel vulnerable if others on the Web knew their real identities. The advantages associated with anonymity was one of the reasons cited for why Th e Palm Beach Post had not adopted the Facebook plugin for its site as of 2013. Another reason stated was the functionality of the plugin, which interviewees said did Digital Manger: We want people to be able to sign into our site and sign up for newsletters, comment on stories and all these different things


41 native to our system, and then if you thrust in the plugin from Facebook it kind of pulls them out of that environment. The Social Me d ia Manager/Strategist said The Atlanta Journal Constitution Th e Atlanta Journal Constitution (Atl anta, Georgia ) was a Cox owned newspaper with a total average circulation of 402,606 for its Sunday issue in March 2012 ( Lulofs 2012 ). The paper was the result of a merger between The Atlanta Constitution founded in 1868, and The Atlanta Journal which was founded in 1883. In 1939 James A. Cox purchased The Atlanta Journal and later purchased The Atlanta Constitution in 1950. Cox both The Atlanta Journal and The Atlanta Constitution merged in 1982 and became The Atlanta Journal Constitution en online since 1998. The site did not allow commenting on news articles until the summer of 2012. The Editor of AJC.com was interviewed to determine how another Cox Media Group newspaper handled online comments. The Editor managed the staff that was respo nsible for The Atlanta Journal website in 2013. This included monitoring the website to ensure that the there and collaborating with e. The Editor had been in this position for two weeks at the time of this interview. The Editor was previously the Digital Manager for Strategic and


42 Subscriber Initiatives at The Atlanta Journal Constitution and had over a decade of journalism experience. As of 2013, a commenting option was automatically attached to every story. However the E ditor said that the paper did not allow commenting on any type of death User Registration A t the time of this study a n individual must have registered an account with the site using an email address Users also had to sign in with that account to comment or to report a comment The Moderators and Moderation Strategies As of 2013, The Atlanta Journal Constitution did no t have full time moderators. Instead the web site was divided into content panels, such as sports, entertainment and news. If a comment was reported in a certain content panel, then a moderator who worked within that panel would be responsible for reviewing it, according to the E ditor. The Atlanta Journal Constitution primarily relied on a crowd source moderation system to moderate comments that users found inappropriate. Every comment on the to report a comment to the paper. A comment would be automatically removed from the site if it received five reports. The Atlanta Journal Constitution used the same process as The Palm Beach Post That is, when a comment was deleted, a message appeared in place of the original comment that in dicated the comment was removed. However, the commenter user name would still be present on the site. The E ditor said that The Atlanta Journal Constitution did not monitor every comment on the website. In some case s the paper would receive an email from a user


43 informing the paper that there was a comment on the site that may be unsuitable. In this instance a moderator would review that comment to determine whether it needed to be removed. Moderators removed comments that contained racial slurs, profanity or threatened other users. The E ditor of the site said that while it was hard to gauge whether a comment is inappropriate, those at the paper would leave comments on the site that were not profane, racist or threaten ing. The editor explained that disagreements in comments were not always removed if the comments contributed to the discourse. but we will police it for the slurs and the bad langua Moderation Advice and the Future of Online Commenting The E of the news site was strengthened by the user comments. She praised had a strong community like environm ent where users were comfortable with policing each other and moderating for inappropriate comments. The E ditor recommended newspapers that are moderating comments encourage community engagement Editor: Try to be as open as possible to provide that engage ment for the users and to allow them to kind of create that community, because engagement level that you might be looking for. Dayton Daily News The third Cox paper in this study, the Dayton Daily News was founded when James A. Cox purchased, and eventually renamed, the Dayton Evening News in 1898. counties in Ohio and reached approximately 555,000 adult rea


44 T he Digital Product Manager was interviewed to find out the moderation strategies used by the Dayton Daily News The Digital Product Manager, who had been in the position for three years, was responsible for t he various social media platforms for the Dayton Daily News an d other Cox properties in Ohio. Previously, the Digital Product Manger worked as a photojournalist for Cox Media Group Ohio. The Dayton Daily News did not allow commenting on its website for a period of time that was not known by the Digital Product Manager, however commenting was enabled approximately six months prior to this research. The Digital Product Manager said that commenting was not allowed on every bsite. Specifically, the paper disabled commenting on crime stories. Digital Product Manager: Usually the topics turn into things that are not productive. of speech, but most of the time on those types of stories time to keep up with all of the different comments, and those conversations tend to be very negative h ave commenting available to them. User Registration In 2013 users had to register an account using a valid email address in order to comment or report a comm Product Manager. The Moderators and Mode ration Strategies The Digital Product Manager said that the Dayton Daily News did not have full time moderators in 2013. Instead the newspaper had a digital department, in which members had various responsibilities such as reviewing comments.


45 The Dayton D aily News followed the visitor agr eement that was developed by Cox According to the Digital Manager, in the cases when commenting was allowed on stories, users were able to flag comments that possibly violated the visitor agreement. Once a comment was fla gged, moderators would often review the entire conversation to understand the context in which the flagged comment was used and determine whether it should be removed from the site. However, moderators did not review entire conversations when removing spam from the website in 2013. Digital Manager: When we get complaints we definitely dig into the story to see commenting was on, then we go in and look at everything and evaluate what comme nts are ok and which ones need to be taken down. The Digital Product Manager said that there was not a great deal of moderation large amount of comments. Moderation Ad vice and the Future of Online Commenting The Digital Product Manager said that the Dayton Daily New s received more comments on its Facebook page than its website in 2013. The Dayton Daily News 6 at the time of this study. The paper posted stories from its site onto the Facebook page and allowed Facebook users to leave comments. The Dayton Digital News used the same comment policy for it s website and Facebook page. The policy was not posted on the Facebook page at the time of this research. The Digital Product M anger said that the user posting rules would be 6


46 extracted from the visitor agreement and posted on the Facebook page shortly after the interview. The Dayton Daily News moderated comment s on its Facebook page in 2013, but th e Digital Product M A moderator would simply remove comments that were inappropriate after the comment was posted on the Dayton Daily Ne ws Facebook page. Moderators removed comments that were considered spam and those that may have containe d libelous statements from the Facebook page. They also removed spam and libelous comments Moderators also removed commen ts that were considered uncivil, which included comments that incited arguments, contained hate speech and personally attacked the conversation and those at the Dayton Dai ly News expected Facebook commenters The Digital Product Manger s aid that the Dayton Daily News may have received more comments on its Facebook page versus its website because users were trained to post on Facebook when the newspaper site did not allow comments. Additionally, the Digital Product Manger said that the Fac ebook page may be receiving a large amount of comments because the Facebook users were very engaged and had created an online community. The Digital Product Manager also interacted with the community by explaining the commenting policy on the Facebook page and knowing several users.


47 Although staff at the Dayton Daily News interacted with Facebook users, the Digital Product Manager said that newspapers must ensure that they are engaging lead to arguments or that do not promote the topic at hand tal Product Manger said. The Digital Product Manger said that the next step in online commenting after 2013 will depend on the digital tools available. While the Dayton Daily News received more comments on its Facebook than its website in 2013, the Digita l Product Manager as users would have more accountability when posting through their Facebook accounts and their Facebook network would be able to view their comments. Tr ibune Company The second news organization that was included in the research study was The Tribune Company. In 2013 Tribune Company was a medi a business based in Chicago, Illinois The Tribune Company was founded in 1847, which is the same year that the c Chicago Tribune websites and newspapers that include the Los Angeles Times The Baltimore Sun and the Orlando Sentinel Orlando Sentinel In 2013 the Orlando Sentinel was a newspaper based in Orlando, F lorida that was the result of the 1973 merger of The Orlando Morning Sentinel and the Reporter


48 Star The History Of The Orlando Se ), Tribune Company acquired the The paper changed its name from the Sentinel Star to the Orlando Sentinel in 1982. The newspaper became available on the Internet in 1995 through the compa ny America Online and launched its own website in 1996 The History Of The Orlando ). In 2012 the Orlando Sentinel had a circulation of 271,824 for its Sunday edition, 185,262 for its Saturday edition and 162,636 for its Monday through Fri day issues (Alliance for Audited Media, 2012). The Online Comment Manager and the Primary Moderator were interviewed for this research. The Online Content Manager had been at The Orlando Sentinel for five months at the time that this research was conduct ed. The Online Content Manger supervised the team of staff members who managed the Orlando Sentinel homepage. The Online Content Manager was also responsible for reviewing comments when the Primary M oderator was not available. The Primary Moderator was responsible for reviewing a flagged comment to determine whether the comment violated the terms of service which is what Tribune Company called their commenting policy. The Primary Moderator, who held this position for three years and was employed by The Orlando Sentinel for over a decade, also worked on search optimization and online audience development. The Primary Moderator had over 20 years of journalism experience and received training in chat hosting from another company.


49 Comment Policy The Orlando Sentinel implemented the terms of service that was developed by The Tribune Company for all of its properties. The policy and commenting system was Primary M oderator at T he Orlando Sentinel described the p very specific guidelines regarding commenting. Online Content Manger: The idea is that we did want to encourage freedom of speech; that we did want to encourage an active participation and en gagement for our users. We want them to respond to our stories, to our online content. But yet we want to make sure that people who are commenting are attacking anyone or spreading any malicious or falsehoods that could be defamatory or could be considered to be essentially attacking other users. User Registration The Primary M oderator and Online Content Manager said that i n order for an individual to comment or report another comment he or she was required to sign into the site with an account that was register ed using a valid email address. The Moderators and Moderation Strategies According to the Primary M oderator, at the time of this study the news site had a team of individua ls responsible for online activities, which included selecting stories that appeared on the site, managing social media and moderating user commenting. Within this team was the Primary M oderator who determine d whether a flagged comment violated the terms of service. Another member of the team would take on these responsibilities when the Primary M oderator was out of the office. The Primary Moderator would train anyone covering the shift. This training consisted of an overview


50 of the procedure used to revi ew comments. However, those taking over the shift would not ban users. T hey could remove a comment, but left that task to the Primary M oderator As of 2013, T he Orlando Sentinel had filters in place on its website to reduce the amount of inappropriate comm ents that were posted. These filters mainly blocked comments containing profanity and racial slurs, but the team refined the filter list by adding offensive words that they encountered while moderating. Primary Moderator: Sometimes people will use words l boards to be a safe and comfortable place for people to n and be heard without being attacked or having somebody make fun of them for other reasons. The team also added to the filter list profane words that users have spelled incorrectly to At the time of this study the Orlando Sentinel did not monitor every comment on every story posted on orlandosentinel.com, instead relying on users to report comments If a reported or flagged comment was reviewed by the moderator and did not violate t he terms, the comment would stay active on the site. If the moderator determined that a flagged comment violated the terms of service, then the comment would be unpublished from the site with no message that alerted other users. In addition, the moderator would send an email to the user stating that he or she has user would receive a warning explaining that he or she would be banned from


51 commenting on the site if a nother inappropriate comment was posted. If the user had already received a warning, then the moderator would explain that the user had been banned from the site for repeated violation of the terms of service. In 2013, the Orlando Senti nel had certain functions incorporated into their system that helped the team identify who had been banned from commenting, which decreased the likelihood of a banned user reregistering with a new account. A user who posted spam on the website would automa tically be banned without warning, as the team had found that those posts often were not from authentic users. When a potentially inappropriate comment made it through the filter and was subsequently flagged, the moderator was responsible for determining whether that comment violated the terms of service. If a moderator could not make the decision to remove the comment then he or she would ask coworkers or a supervisor. Online Content Manager: times w e will have an internal discussion about whether that comment is appropriate. If someone flags something that the site violates those terms, a nd if our primary moderator or someone who is moderating that post is unsure, then they can send it up the chain of command and get more eyes on The Primary M oderator explained the process of rev iewing a comment. Primary Moderator: I also will keep in mind how the user communicates on the little ambiguous, but we can see their other comment s and Chances are if we think in that particular case it might be inappropriate comments.


52 The Primary M oderator and Online Content Manager said that many flagged comments were fairly easy to review because they blatantly violated the terms of service. For example, the moderators at the Orlando Sentinel had zero tolerance for comments that attacked other users on the basis of race, gender, eth nicity, or sexual preference. The Primary M reviewed flagged comment. Each flagged comment was reviewed on a case by case unt. For example, if a user informed those at the Orlando Sentinel that he or she should not be banned because they never received a warning, the moderator would give the user the benefit of the doubt and reinstate the account. The moderator also recognize d that the paper covered topics that have received national readership, so a new user to the site may not be familiar with the way that the paper operated its website. The Primary M oderator also said that the team is se nsitive to cultural differences. Prim ary Moderator: You have examples of people who are from different parts of the nation or from different parts of the world, and they may more acceptable in their culture. So sometimes you can ki nd of get that feeling from somebody and you jut tell them, At the time of this study the Primary M oderator checked flagged comments at least every t wo h ours throughout the work shift, which was from 9 a.m. until 5 p.m. on weekdays. The Primary M oderator and if the Primary M oderator was not at work then someone else from the team reviewed flagged comment s.


53 Moderation Advice and the Future of Online Commenting The Online Content Manger at T he Orlando Sentinel recommended that any media organization implementing comment moderation policies should use the commenting option as a way of engaging their audien ce and should ask users to actively participate. However, while the manager advised organizations to giver users a wide latitude with commenting, moderators should not allow bullying or malicious attacks on the site. The Primary M oderato r advised media organizations to set up a clear, spe cific terms of service, place the policy somewhere visible on the website and ensure that every person on the moderating team is on one accord. The Online Content Manager explained that social media plu gins would be the future of online commenting in the years following 2013 and said that users may be more even tempered when commenting from their social media profiles. The Primary M oderator said that websites using social media plugins may see a decrease in the number of comments as users may not want to post certain opinions, such as political stances, using their real names out of fear that someone, like an employer, may read it. Primary Moderator : I think the determination would be that you might try what the Internet is all about. In my opinion it would be great as a moderator, but I do think that the comments would drop off and we would get fe wer comments on stories if people had to sign in with their real names.


54 CHAPTER 5 DISCUSSION This research set out to determine how online news sites moderate user comments and what issues journalists consider when moderating comments. Through a series of interviews with six professionals associated with four newspapers, this study found that news sites relied on crowd sourced moderation to both monitor comments and encourage community. Users would flag comments that potentially violated the commenting Moderators would then review flagged comments and typically remove ones that contained personal attacks, profanity and hate speech, as these comments did not contribute to the constructive com munity environment. These processes are consistent with how journalists reviewed letters to the editor and suggest that certain journalistic practices for promoting public dialogue and interactivity have simply shifted to an online medium. gnificance grew in the 21 st Century, the medium also continued to impact journalism. Not only did an increasing number of newspapers have an online presence, the Internet gave members of the public new ways to interact with journalists and each other. In f act, in 2012 the Pew Research Center for People and the Press found that there was an increasing number of Americans who received news online, or news headlines on a soc One interactivity feature, the online comment section, was popular among news organizations and readers alike. Commenting increased t he interaction between the


55 public and journalists, gave commenters the opport unity to leave feedback on news, However, allowing online commenting forced news organizations to deal with comments that contain ed hate speech, libelous language, spam, off topic dialogue and incivility. These aforementioned features of online commenting led professionals at news role as a gat ekeeper. The New Gatekeeper The gatekeeping theory in mass communication refers to the series of filters that content, such as news stories, letters to the editor, and photographs, go through before they are presented to the public. This process includes a n item of information passing from the source, to the reporter, to various editors and finally to the public (Shoemaker, Eichholz, Kim & Wrigley, 2001). Although gatekeeping traditionally meant that journa lists had a role in what news was prese nted to the public and how it was presented, journalists still hold on to the gatekeeping role when they moderate comments on the Internet. Corporations and news organizations start the gatekeeping process when the y develop a commenting policy. Individual m oderators a t news sites continue the process when th ey determine what comments stay on the website and what comments are removed. In addition, comments may also go through literal filters before they are even posted on a site. In this instance, a reader who may have a valid comment may be silenced if he or she uses profanity or a racial slur. If a comment is not presented in the manner that the news organizat ion has established as acceptable, that comment may be blocked from being poste d or deleted after it is posted


56 Half of the moderators in this study indicated that it was not always easy to determine whether a comment should be unpublished from the news site. However, most comments that they said were removed had posting policie s. Comments that contained profanity, racial slurs and spam were removed. Personal attacks and libelous statements were also removed from the websites almost automatically. If a particular moderator could not decide whether to remove a comment or not, then he or should would discuss it with coworkers or supervisors to reach a consensus. This gatekeeping process is similar to the way that editors determined which letters to the editor would be published in newspapers. Wahl Jorgensen (2007) found that editor inte raction may change, the gatekeeping practices that journalists use stay in tact. While gatekeeping practices of a moderator may be similar to that of a letters editor, t he findings of this study also suggest the emergence of a new trend that can be viewed as reverse gatekeeping, in which user comments affect how journalists perform their jobs and the stories that are reported. For example, the Digital Manager at The Palm Beach Post stated that reporters at the newspaper would often read comments and discove r tips, new story angles and future sources. This sort of reverse gatekeeping is not unique to The Palm Beach Post as other report ers also found online comments useful because they could and what is not (Santana, 2011). In reverse gat ekeeping the online community of


57 readers and commenters ha ve more control in Web content when compared to their level of control in print publications. Building an Online Community The findings of this r esearch sugges t that creating their own online community where users feel comfortable to interact is important to online news sites. Many decisions that moderators made were done to foster a safe environment for commenters and promote communication. This included removi ng comments that contained personal attacks and hate speech. In addition, moderators also policies. In the case of The Orlando Sentinel a moderator emailed a user who violated the acceptable use policy. In the email, the moderator explained the policy and warned the individual that a second unacceptable posting would ban the user or, in the case of a user who had a second violation, would explain why the user was being banned. This communication between moderators and commenters serves to strengthen the online community. Not only do repeat commenters become familiar with each other as an online community grows, moderators also learn the user names and commenting habits of the c ommunity members. Fostering an online community can result in several positive outcomes for a publication. In this digital age where members of the public can get news from a variety of outlets including national news websites, blogs, cable news network s, magazines and social media, a local newspaper has to find a way to attract and maintain readership. A newspaper with an established Web presence and engaged online community may be inviting to readers who want to interact. An online community can


58 also f oster a sense of loyalty that makes commenters want to come back to a particular news site. Another p ositive related to repeat commenters is familiarity. Members of online ay be less likely to violate it. Furthermore, if a member of the community does violate the policy then he or she should not be surprised when the inappropriate comment is removed. This may lead to more even tempered responses to comments being unpublished a nd users being warned or banned because the user is expecting it. Promoting community engagement also involved being conservative when removing comments. Several of t he individual interviewed as part of this study said that it was important to not to s uppress engagement by over moderating. This is also similar to the approach letters editors used. Like those interviewed in this study, Wahl posting history, the context of the conversation and cultural differences into account when reviewing flagg ed comments to ensure that that the comment was truly unconstructive. Future journalists and others implement ing online moderation policies can foster community by encouraging comments, as online comments give users an opportunity for dialogue. Those want ing to build an online community should also promote freedom of expression and diverse opinions in commenting sections. However, comments should not discourage other users from participating because th e posted comments


59 contain uncivil language, as language that is blatantly sexist, racist, homophobic or personally attacks someone does not contribute to a productive discourse. Crowd S ourced Moderation All five of the newspapers in this study used crowd sourced moderation, in which users could report comment s that they felt violated the commenting policy. This crowd sourced moderation strategy required users to register an account with a valid, working email address before they could comment or report comments. User registration allowed moderators to keep tra ck of users who violated the commenting policy and users who reported comments. The Orlando Sentinel used this information to email a warning to users who violated the terms of service and inform repeat offenders that they would be banned from the website. This crowd sourcing strategy also increased user engagement and gave readers more power to determine what content appeared on news sites As Briggs (2010) stated, the collective experience of crowd sourced moderation gives users a sense of ownership in t heir online communities. This was evident in the sports section of The Atlanta Journal website, in which users formed a community and were not afraid to self police within that community. This type of moderation makes it possible for news o rganizations to moderate comments without the news staff being responsible for reviewing every single comment that is posted on the site. Crowd sourcing moderation can be particularly useful for news sites that may receive too many comments for moderators review In a time where newsrooms may be lacking fundi ng for adequate staffing, crowd sourced moderation can also be beneficial to sites that do not employ full time moderators. All of the


60 newspapers in this study employed moderators who were also responsi ble for other duties. While the newspapers who used crowd sourced moderation did not review every comment posted on their website, the moderators at The Palm Beach Post and the Orlando Sentinel said that it was essential to check the flagged comments queu e often. This reinstates moderation has a significant affect on the performance of an online community (p. 261). Users may begin to feel uncomfortable if a flagged comment has not been reviewed o r removed in a timely manner. In addition, the conversation may shift from the article Anonymity in Online Commenting While users had to create accounts with valid email addresses on all of the newspa per sites in this study, a user did not have to register with his or her legal name. user names could be a pseudonym Although previous research has indicated that anonymous comments tend to be more uncivil, three interviewees in thi s study mentioned that anonymous commenting could be positive. Interviewees from Cox Media Group The Palm Beach Post and the Orlando Sentinel said that user pseudonyms permitted users the freedom to express valid views and opinions that they may not be co mfortable with other people, such as employers, associating with them. Anonymity can be particularly beneficial when users are contributing to political discussions. Although anonymity was mentioned as a positive feature, some of those interviewed said t hat the future of online commenting would involve news organizations using social media plugins, which would make online commenting less anonymous as


61 The Facebook plugin w as gaining popularity among newspaper sites (Santana, 2012) and those who mentioned Facebook pointed to that particular plugin because Facebook was the largest social network at the time of this research. Not only would commenting through Facebook add anot her layer of user accountability, it would also give newspapers the opportunity to engage audiences on multiple social platforms. However, some interviewees said that that anonymity was important enough to not adopt the Facebook plugin and the plugin may e ven result in a decrease in comments. The Role of Social Media in Online Commenting While some moderators said that a social media plugin might be the future of online commenting, one newspaper had already made the transition to Facebook. The Dayton Dail y News posted stories from its website on to its Facebook page for the over 20,000 followers to read, like and comment on. Instead of devoting a large amount of newspaper focused more on moderating comments and engaging the audience on its Facebook page. This type of interactivity could indicate a future in which newspaper commenting and social media has merged in a way that goes beyond a plugin. Meeting users where they are alrea dy participating in pubic discourse may increase interactivity between newspapers and its readers. In essence, journalists could engage with an audience who is already engaged. This social media interactivity also has the potential for community building. Additionally, newspapers may observe more civil discussions when users comment using social media accounts. A newspaper may even gain new readers with an active social media presence.


62 While social media has great potential for community engagement, certa in platforms may also harm an online community. Social media plugins installed on news websit e rather than a newspaper site. This was the reason given by the Digital Manager at The Palm Beach Post who said that a Facebook plugin would pull readers out of the Social media networks can be online communities, but it is important for news sites to develop their own, individual online communities. Thus, technology similar to the Facebook plugin may be counterproductive to that goal. From a technical aspect, a social media plugin may not be compatible with a oderation system. In addition, moderators may not be able to filter comments, contact commenter s or ban policy violators from their site if they are using a plugin. This loss of control could have adverse e community, as users could repeatedly leave uncivil comments that could potentially discourage other users from interacting. Suggestions for Future Research This study found that moderators at newspapers try to foster a community environment on websites. Futu re research should explore whether the level of community engagement correlates with civility in online comments. For example, this research would test whether engagement practices, such as the moderator commenting on stories or emailing users directly, re sulted in more civil comments. The study could also examine whether a strong online community resulted in a wider on line audience. Findings of such a study would aid journalists in the quest to create commenting sections where users feel comfortable enough to contribute to public


63 discourse without the fear of being personally attacked. The se finding s would also be significant to journalism educators who in teach ing online management will teach students how to manage online communities. While past research addressed who wrote letters to the editor, feature research should study who typically posts comments to news sites. Findings of the research would be valuable, as they would inform newsrooms about what portion of the population engages in this form of int eractivity. These findings could also be compared to the demographics of typical letter writers. All five of the newspapers included in this study followed moderation policies that were created by parent companies. Future research should examine how moder ation policies are developed to determine whether this is a common practice. Findings of this study would be significant to newsrooms and websites who are developing moderation policies, as well as educators who are teaching journalism or law students how user terms are developed and implemented. Research studying the inclusion of comment moderation in journalism programs would also be important. A study examining how curricula has changed and what techniques educators are using to teach comment moderation would be valuable to journalism schools that wish to incorporate the teaching of these important skills into their programs. A future study might address the high job turnover within the journalism industry and whether the lack of institutional knowledge has an affect on how comments are moderated. Such research would be useful for newsrooms in determining whether there


64 should be a set training process for new hires in order to ensure consistent moderation practices. The findings of this research are u seful in that they may assist educators in preparing journalism students for careers as moderators, online editors and community managers some of the new roles in the evolving media landscape. Online commenting has become an ever present feature on news we bsites during the 21 st Century; therefore it is becoming more likely that future journalists will need to possess moderation skills. This study has presented the process by which newspapers simultaneously moderate user generated content and foster online c ommunities.


65 APPENDIX A COX MEDIA GROUP VISITOR AGREEMENT REGISTRATION To obtain access to certain services on our Service, you may be required to register with us. Children under the age of 13 may not register for the Service. You agree that the informa tion you supply during that registration process will be accurate and complete and that you will not register under the name of, nor attempt to use this Service under the name of, another person. We reserve the right to reject or terminate any user name th at, in our judgment, we deem offensive. You will be responsible for preserving the confidentiality of your password and will notify us of any known or suspected unauthorized use of your account. USER PROVIDED CONTENT Your License to Us By submitting ma terial (including, but not limited to, any text, photos, video or other content) to us, you are representing that you are the owner of the material, or are making your submission with the express consent of the owner. By submitting any materials via this S ervice, you grant us, and anyone authorized by us, including, without limitation, our Affiliates, a perpetual, irrevocable, royalty free, unlimited, worldwide, transferable, non exclusive and unrestricted license to use, reproduce, modify, archive, publish sell, exploit, display, create derivative works from, publicly perform, and otherwise distribute such material in any medium (whether now known or hereafter developed), in any manner we see fit, and for any purpose that we choose. The foregoing grant inc ludes the right to exploit any proprietary rights in materials you submit to this Service, including, but not limited to, rights under copyright, trademark or patent laws that exist throughout the world. Without limiting the generality of the previous sen tence, you agree that we may use, distribute, share or otherwise provide such material under any terms we see fit to any third party without the requirement of providing you any form of compensation. You also agree that we, and anyone authorized by us, may identify you as the author of any of your postings by name, email address or screen name, as we or they deem appropriate. We also reserve the right (but assume no obligation) to delete, move, or edit any postings that come to our attention that we conside r unacceptable or inappropriate, whether for legal or other reasons. You understand that the technical processing and transmission of the Service,


66 including content submitted by you, may involve transmissions over various networks, and may involve changes to the content to conform and adapt it to technical requirements of connecting networks or devices. USE OF COMMUNICATIONS SERVICES Specific Prohibited Uses Without limiting the foregoing, we may immediately terminate your use of any Communications Ser vice if you engage in any of the following prohibited activities: Uploading, posting, emailing, transmitting or otherwise making available any content that is unlawful, harmful, threatening, abusive, libelous, or obscene; Impersonating any person or entity or falsely stating or otherwise misrepresenting your affiliation with a person or entity; Forging headers or otherwise manipulating identifiers in a manner that disguises the origin of any content you transmit through any Communications Service; Uploadin g, posting, emailing, transmitting or otherwise making available any content that you do not have a right to make available under any law or under any contractual or fiduciary relationship (such as inside information, proprietary and confidential informati on learned or disclosed as part of employment relationships or under nondisclosure agreements); Uploading, posting, emailing, transmitting or otherwise making available any content that infringes any patent, trademark, trade secret, copyright or other prop rietary right of any party; Uploading, posting, emailing, transmitting or otherwise making available any unsolicited or unauthorized advertising, promotional materials, or any other form of solicitation, without our express written approval; Gathering for the purpose of "spamming" any email addresses that users post in our chat rooms, forums and other public posting areas; Uploading, posting, emailing, transmitting or otherwise making available any content or material that contains software viruses, worms o r any other computer code, files or programs designed to interrupt, destroy or limit the functionality of any computer software or hardware or telecommunications or other equipment, or to cause a security breach of such software, hardware or telecommunicat ions or other equipment; Posting fraudulent classified listings;


67 Uploading or posting any off topic or irrelevant material to any chat room or forum; Interfering with or disrupting any servers or networks used to provide the Communications Services, or dis obeying any requirements, procedures, policies or regulations of the networks we use to provide the Communications Services; Violating any applicable local, state, national or international law, including, but not limited to (1) all applicable laws regardi ng the transmission of technical data exported from the United States or the country in which you reside, (2) regulations promulgated by the U.S. Securities and Exchange Commission, and (3) any rules of any national or other securities exchange, including, without limitation, the New York Stock Exchange, the American Stock Exchange or the NASDAQ; "Stalking" or otherwise harassing another; Instigating or encouraging others to commit illegal activities or cause injury or property damage to any person; Collect ing or storing personal data about other users; Gaining unauthorized access to our Service, or any account, computer system, or network connected to this Service, by means such as hacking, password mining or other illicit means; or Obtaining or attempting to obtain any materials or information through any means not intentionally made available through this Service.


68 APPENDIX B ORLANDO SENTINEL TERMS OF SERVICE Registration. Registration is not required to view certain Content. However, you are req uired to register if you wish to post a comment or upload a video, or view certain other Content. If you become a Registered Member of OrlandoSentinel.com, you accept responsibility for all activities that occur under your Registration Account. You agree t o provide true, accurate, complete, and correct information at the time of registration, and to promptly update this information as needed so that it remains true, accurate, complete, and correct. We reserve the right to terminate your access and use of Or landoSentinel.com if individuals from more than one household access OrlandoSentinel.com using any single Registration Account. You are responsible for maintaining the confidentiality of your password and for restricting access to your computer so others o utside your household may not access OrlandoSentinel.com using your name in whole or in part without our permission. If you believe someone has accessed OrlandoSentinel.com using your Registration Account and password without your authorization, e mail us immediately at feedback@orlandosentinel.com User Content Representations and Warranties. By placing material on OrlandoSentinel.com, including but not limited to posting content or communications to an y OrlandoSentinel.com bulletin board, forum, blogspace, message or chat area, or posting text, images, audio files or other audio visual content to the site ("User Content"), you represent and warrant: (1) you own or otherwise have all necessary rights to the User Content you provide and the rights to provide it under these Terms of Service; and, (2) the User Content will not cause injury to any person or entity. Using a name other than your own legal name in association with the submission of User Content is prohibited (except in those specific areas of OrlandoSentinel.com that specifically ask for unique, fictitious names). User Content License. For all User Content you post, upload, or otherwise make available ("Provide") to OrlandoSentinel.com, you gran t Tribune Interactive, Inc. ("TI"), its affiliates and related entities, including OrlandoSentinel.com and its affiliated newspapers, Web sites, and broadcast stations, a worldwide, royalty free, perpetual, irrevocable, non exclusive right and fully sub li censable license to use, copy, reproduce, distribute, publish, publicly perform, publicly display, modify, adapt, translate, archive, store, and create derivative works from such User Content, in any form, format, or medium, of any kind now known or later developed. Without limiting the generality of the previous sentence, you authorize TI to share the User Content across all Web sites, newspapers, and broadcast stations affiliated with Tribune Company, to include the User Content in a searchable format acc essible by users of OrlandoSentinel.com and other TI Web sites, to place advertisements in close proximity to such User Content, and to use your name, likeness and any other information in connection with its use of the material you provide. You waive all moral rights with respect to any User Content you provide to OrlandoSentinel.com. You also grant TI the right to use any material, information, ideas, concepts, know how or techniques contained in any communication you provide or otherwise submit to us for any purpose whatsoever, including but not limited to, commercial purposes, and developing, manufacturing and marketing


69 commercial products using such information. All rights in this paragraph are granted without the need for additional compensation of any sort to you. User Content Screening and Removal. You acknowledge that OrlandoSentinel.com and/or its designees may or may not pre screen User Content, and shall have the right (but not the obligation), in their sole discretion, to move, remove, block, ed it, or refuse any User Content for any reason, including without limitation that such User Content violates these Terms of Service or is otherwise objectionable. User Content Assumption of Risk. OrlandoSentinel.com cannot and does not monitor or manage al l User Content, and does not guarantee the accuracy, integrity, or quality of User Content. All User Content provided to OrlandoSentinel.com is the sole responsibility of the person who provided it. This means that you are entirely responsible for all User Content that you provide To protect your safety, please use your best judgment when using OrlandoSentinel.com forums. We discourage divulging personal phone numbers and addresses or other information that can be used to identify or locate you. You acknow ledge and agree that if you make such disclosures either through posting on any bulletin board, forum, blogspace, message or chat area, or uploading text, images, audio files or other audio visual content, in classified advertising you place or in other in teractive areas, or to third parties in any communication, you do so fully understanding that such information could be used to identify you. User Content Posting Rules. Any decisions as to whether User Content violates any Posting Rule will be made by Or landoSentinel.com in its sole discretion and after we have actual notice of such posting. When you provide User Content, you agree to the following Posting Rules: If the photo or video depicts any children under the age of 13, you affirm that you have wri tten permission from the child's parent or guardian to provide the photo or video. Do not provide User Content that: contains copyrighted or other proprietary material of any kind without the express permission of the owner of that material. contains vu lgar, profane, abusive, racist or hateful language or expressions, epithets or slurs, text, photographs or illustrations in poor taste, inflammatory attacks of a personal, racial or religious nature. is defamatory, threatening, disparaging, grossly inflam matory, false, misleading, fraudulent, inaccurate, unfair, contains gross exaggeration or unsubstantiated claims, violates the privacy rights of any third party, is unreasonably harmful or offensive to any individual or community.


70 violates any right of Or landoSentinel.com or any third party. discriminates on the grounds of race, religion, national origin, gender, age, marital status, sexual orientation or disability, or refers to such matters in any manner prohibited by law. violates or encourages the vi olation of any municipal, state, federal or international law, rule, regulation or ordinance. interferes with any third party's uninterrupted use of OrlandoSentinel.com. advertises, promotes or offers to trade any goods or services, except in areas speci fically designated for such purpose. uses or attempt to use another's Registration Account, password, service or system except as expressly permitted by the Terms of Service. uploads or transmits viruses or other harmful, disruptive or destructive files, material or code. disrupts, interferes with, or otherwise harms or violates the security of OrlandoSentinel.com, or any services, system resources, accounts, passwords, servers or networks connected to or accessible through OrlandoSentinel.com or affilia ted or linked sites. "flames" any individual or entity (e.g., sends repeated messages related to another user and/or makes derogatory or offensive comments about another individual), or repeats prior posting of the same message under multiple threads or s ubjects. WARNING: A VIOLATION OF THESE POSTING RULES MAY BE REFERRED TO LAW ENFORCEMENT AUTHORITIES.


71 APPENDIX C THE PALM BEACH POST LETTERS TO THE EDITOR POLICY Send a letter to The Post We welcome original letters about issues of interest and material t hat has appeared in mail address and daytime phone number.


72 APPENDIX D THE ATLANTA JOURNAL CONSTITUTION LETTERS TO THE EDITOR POLICY Letters to the e ditor section Your letter will be submitted to the Atlanta Journal Constitution for publication in the newspaper. Please include your first and last names (no initials, please) and, for verification purposes only, your home address and both your daytime an d nighttime telephone numbers. Use the link below or send an email to letters@ajc.com Also, we'd appreciate your including a dab of bio information, namely, what you do for a living. This bio info is optional. We hope to hear from you soon.


73 APPENDIX E DAYTON DAILY NEWS LETTERS TO THE EDITOR POLICY Have your say The Dayton Daily News welcomes letters to the editor. Letters should be 250 words or fewer. We need your full name, address and phone number for verification purposes. Due to volume, not all letters can be published. To publish as many letters as possible, they may be edited. No attachments, please. You can send your letter the following ways: > Use the form below > E ma il a letter to the editor > Fax your letter to (937) 225 7302 > Mail: Letters to the editor, 1611 S. Main St., Dayton, OH 45409 > E mail a Speak Up! item to speakup@daytondailynews.com We respect your privacy: Your street address, phone numbers, and email address will not be published. We require that information to verify that you are a real person and to contact you if necessary.


74 APPENDIX F ORLANDO SENTINEL LETTERS TO THE EDITOR POLICY L etters to the Editor Letters must be exclusive to the Orlando Sentinel and include your full name, address and phone number. We publish only your name and home city. Letters should be no more than 250 words. E mail column submissions to insight@orlandosentinel.com


75 APPENDIX G INTERVIEW QUESTIONS 1. Please describe your involvement in implementing the comment moderation policy. 2. When did (publication) launch it s website? 3. When did (publication) start allowing comments on its website? 4. Who moderates the comments on your website? Is moderation a fulltime job? 5. How are moderators trained? 6. Please explain the process used to determine if a comment violates the comment p olicy. 7. What advice would you offer to other news organizations developing comment moderation policies? 8. What do you think is the next step in online comment policies? How will online publications address anonymity? 9. Is there anything else that you would lik e to share about the process of developing or implementing the comment moderation policy?


76 APPENDIX H SAMPLE EMAIL Hello (Potential Interviewee), College of J ournalism and Communications. As a journalist who is particularly interested in online publishing and user generated content, I feel it is important to study the Web and the way it has transformed how we report and interact with the public. I am current ly working on my thesis, which seeks to understand the development and implementation of online comment moderation policies. In order to better understand in developi ng comment moderation policies, and also those who are responsible for moderation. Because comment moderation is becoming increasingly important, educators may need to adjust curriculum to teach this important editing skill to future journalists. One of the goals of this thesis is to help journalism educators in that endeavor. (Referrer) suggested that I speak with you because of the knowledge you could provide on this subject. Although this is research, I feel that this thesis could have positive, pract ical implications. We could conduct the interview, which will be recorded and transcribed, over Skype or phone at your convenience. The interview should take about 30 minutes. I would be extremely appreciative if you were able to participate. Thank you and I look forward to speaking with you, Antionette Rollins


77 REFERENCES About. (n .d.). Retrieved from http://www.coxmediagroup.com/about/ About tribune c ompany. (n.d.). Retrieved f rom http://corporate.tribune.com/pressroom/?page_id=4200 Alcindor, Y., Bello, M., & Copeland, L. (2012, Ma rch 21). In wake of black teen Trayvon M artin's death, America is soul searching. USA Today Retrieved from http://usatoday30.usatoday.com/news/nation/story/2012 03 20/trayvon martin tee n shot stereotypes/53677634/1 Allison, J. E. (2002). Technology, development, and democracy: International conflict and cooperation in the information age. (). Albany: State University of New York Press. Arthur D Santana. (2011). Online readers' commen ts represent new opinion pipeline. Newspaper Research Journal, 32 (3), 66. Banks, J. (2010). Regulating hate speech online. International Review of Law, Computers & Technology, 24 (3), 233 239. doi: 10.1080/13600869.2010.522323 Berkowitz, D. A. (1997). So cial meanings of news: A text reader Thousand Oaks, Calif: Sage Publications. Bhattarai, A., Rus, V., & Dasgupta, D. (2009). Characterizing comment spam in the blogosphere through content analysis. Paper presented at the Computational Intelligence in Cyb er Security, 2009. CICS'09. IEEE Symposium on, 37 44. Bishop, J. (2009). Enhancing the understanding of genres of web based communities: The role of the ecological cognition framework. International Journal of Web Based Communities, 5 (1), 4 17. Blood, R. (2003). Weblogs and journalism: Do they connect? Nieman Reports, 57 (3), 61 63. Retrieved from https://search.ebscohost.com/login.aspx?direct =true&db=ufh&AN=10976577&sit e=ehost live Botelho, G. (2012). What happened the night T rayv on Martin died. Retrieved from http://www.cnn.com/2012/05/18/justic e/florida teen shooting details Boyd, D. m., & Ellison, N. B. (2007). Social network sites: Definition, history, and scholarship. Journal of Computer Mediated Communication, 13 (1), 210 230. doi: 10.1111/j.1083 6101.2007.00393.x Braun, J., & Gillespie, T. (2010). Hosting the public discourse: News organizations, digital intermediaries, and the politics of making newsmedia social. Paper presented at the 11th International Symposium o n Online Journalism, Austin, TX.


78 Briggs, T. (2010). Social media's seco nd act: Toward sustainable brand engagement. Design Management Review, 21 (1), 46 53. doi: 10.1111/j.1948 7169.2010.00050.x Brown, J., Broderick, A. J., & Lee, N. (2007). Word of mouth communication within online communities: Conceptualizing the online so cial network. Journal of Interactive Marketing, 21 (3 ), 2 20. doi: 10.1002/dir.20082. Chen, J., Xu, H., & Whinston, A. B. (2011). Moderated online communities and quality of user generated content. Journal of Management Information Systems, 28 (2), 237 268. doi: 10.2753/MIS0742 1222280209 Christopherson, K. M. (2007). The positive and negative implications of anonymity in e internet, nobody knows y Computers in Human Behavior, 23 (6), 3038 3056. doi: doi:10.10 16/j.chb.2006.09. Citron, D. K., & Norton, H. L. (2011). Intermediaries and hate speech: Fostering digital citizenship for our information age Boston University Law Review, 91 (1435) Coffey, B., & Woolworth, S. (2004). Destroy the scum, and then neuter their families: The web forum as a vehicle for community discourse? The Social Science Journal, 41 (1), 1 14. doi:10.1016/j.soscij.2003.10.001 Cohen, K. R. (2006). A welcome for blogs. Continuum: Journal of Media & Cultural Studies, 20 (2), 161 173. doi: 1 0.1080/10304310600641620 Dan John son. (2000). Anonymity and the I nternet. The Futurist, 34 (4), 12. Daniels, J. (2009). Cloaked websites: Propaganda, cyber racism and epistemology in the digital era. New Media & Society, 11 (5), 659 683. doi: 10.1177/14614 44809105345 Daniels, J. (2010). Cyber racism: White supremacy online and the new attack on civil rights. Journal of Popular Culture, 43 (5), 1137. doi: 10.1111/j.1540 5931.2010.00790_5.x Daniels, J. (2009). Cyb er racism : White supremacy online and the n ew attack on civil rights Lanham: Rowman & Littlefield Publishing Group, Inc. Davis, R. (2009). Typing politics: The role of blogs in American politics Oxford: Oxford University Press. Diakopoulos, N., & Naaman, M. (2011). Towards quality discourse in online news comments. New York, NY. 133 142. doi: 10.1145/1958824.1958844 http://doi.acm.org.lp.hscl.ufl.edu/10.1145/1958824.1958844


79 Dijk, T. A. v. (1992). Racism, el ites, and conversation. Atlantis: Revista De La Asociacin Espaola De Estudios Anglo Norteamericanos, 14 (1 2), 201 257. DiMaggio, P., Hargittai, E., Neuman, W. R., & Robinson, J. P. (2001). Social implications of the internet. Annual Review of Sociology, 27 307 336. doi: 10.1146/annurev.soc.27.1.307 Domingo, D., Quandt, T., Heinonen, A., Paulussen, S., Singer, J. B., & Vujnovic, M. (2008). Participatory journalism practices in the media and beyond. Journalism Practice, 2 (3), 326 342. doi: 10.1080/17512 780802281065. Ehrlich, P. (2002). Communications decency act section 230. (H armful speech regulation). Berkeley Technology Law Journal, 17 (1), 401. Ellis, D., Oldridge, R., & Vasconcelos, A. (2004). Community and virtual community. Annual Review of Inform ation Science and Technology, 38 145 186. Entman, R. M., & Bennett, W. L. (2001). Mediated politics: Communication in the future of democracy Cambridge, UK: Cambridge University Press. Ethics. (n.d.). Retrieved from http://www.spj.org/ethics.asp Faridani, S., Bitton, E., Ryokai, K., & Goldberg, K. (2010). Opinion space: A scalable tool for browsing online comments. Paper presented at the Proceedings of the 28th International Conference on Human Factors in Computing Systems, 1175 1184. Gunter, B., Campbell, V., Touri, M., & Gibson, R. (2009). Blogs, news and credibility. Aslib Proceedings, 61 (2), 185 204. doi: 10.1108/00012530910946929 Gyongyi, Z., & Garcia Molina, H. (2005). Web spam taxonomy. Paper pre sented at the First International Workshop on Adversarial Information Ret rieval on the Web (AIRWeb 2005). Hardaker, C. (2010). Trolling in asynchronous computer mediated communication: From user discussions to academic definitions. Journal of Politeness Re search: Language, Behavior, Culture, 6 (2), 215 242. doi: 10.1515/jplr.2010.011 Hermida, A., & Thurman, N. J. (2009). Comments please: How the British news media are struggling with user generated content. 8th International Symposium on Online Journalism, 2009 (April 3), 1 28. Hermida, A., & Thurman, N. J. (2008). A clash of cultures : The integration of user generated content within professional journalistic frameworks at British newspaper websites. Journalism Practice, 2 (3), 343 356. Herring, S. C., Sche idt, L. A., Wright, E., & Bonus, S. (2005). Weblogs as a bridging genre. Information Technology & People, 18 (2), 142.


80 Herring, S., Job Sluder, K., Scheckler, R., & Barab, S. (2002). Searching for safety online: Managing "trolling" in a feminist forum. Inf ormation Society, 18 (5), 371 384. doi: 10.1080/01972240290108186. Hill, J. H. (2008 ). The everyday language of white racism Wiley Blackwell. History. (n.d.). Retrieved from http://www.coxenterprises.com/about cox/history.aspx#.URsoctCYYb6 Holcomb, J., Gross, K. & Mitchell, A. (2011). How mainstream media outlets use twitter. Retrieved from http://www.journalism.org/analysis_report/how_mainstream_media_outlets_use_t witter How to submit a letter to the editor. (n.d.). Retrieved from http://www.nytimes.com/content/help/site/editorial/letters/letters.html Hurwitz, J., & Peffley, M. (2005). Playing the race card in the post Willie Horton era: The impact of racialized code words on support for puniti ve crime policy. The Public Opinion Quarterly, 69 (1), 99 112. doi: 10.1093/poq/nfi004 Hwang, H. (2008). Why does incivility matter when communicating disagreement?: Examining the psychological process of antagonism in political discussion. ProQuest. In C hanging News Landscape Even Television is Vulnerable ( 2012 ). Retrieved from http://www.people press.org/2012/09/27/in changing news landscape even television is vulnerable/ Josey, C. S. (2010). Hate speech and identity: An analysis of neo racism and the indexing of identity. Discourse & Society, 21 (1), 27 39. doi: 10.1177/0957926509345071 Johnson, K. A ., & Wiedenbeck S (2009). Enhancing perceived credibility of citizen journalism web sites. Journalism & Mass Communication Quarterly, 86 (2), 332 348. doi: 10.1177/107 769900908600205. Keyes, S. (2009). Fiery forums. Retrieved from http://tae.asne.org/StoryContent/tabid/65/id/458/Default.aspx Like. (n.d.). Retrieved from https://www.facebook.com/help/like Lulofs, N. (2012). The top U.S. new spapers for M arch 2012. Retrieved from http://accessabc.wordpress.com/2012/05/01/the top u s newspapers for march 2012/ Leccese M (2009). Online information sources o f political blogs. Journalism and Mass Communication Quarterly, 86 (3), 578.


81 Munksgaard, D. C. (2010). Warblog without end: Online anti islamic discourses as persuadables. (Unpublished PhD). University of Iowa, doi: http://ir.uiowa.edu/etd/715 Nardi, B. A., Schiano, D. J., & Gumbrecht, M. (2004). Blogging as social activity, or, would you let 900 million people read your diary? Proceedings of the 2004 ACM Conference on Computer Supported Cooperative Work (2004), Pp. 222 231, Doi:10.1145/1031607.1031643 Key: Citeulike:77453, 222 231. d oi: doi:10.1145/1031607.1031643. Our history. (n .d.). Retrieved from http://projects.ajc.com/s ervices/info/history/ Perez Pena, R. (2010). News sites rethink anonymous comments. Retrieved January, 2013, from http://www.nytimes.com/2010/04/12/technology /12comments.html?_r=1& Purcell, K., Rainie, L., Mitchell, A., Rosenstiel, T., & Olmstead, K. (2010). Understanding the participatory news consumer: How internet and cell phone users have turned news into a social experience. Pew Research Center, March. Qi Li, Yunhong Zhang, Mi Shi, & Jing Luo. (2010). Can anonymity network increase the utilitarian in personal moral decision? Paper presented at the Web Society (SWS), 2010 IEEE 2nd Symposium on, 181 184. Raeymaeckers, K. (2005). Letters to the editor A f eedback opportunity turned into a marketing tool: An account of selection and editing practices in the flemish daily press. European Journal of Communication, 20 (2), 199 221 doi: 10.1177/0267323105052298. Rafaeli, S., & Sudweeks, F. (1997). Networked inte ractivity. Journal of Computer Mediated Communication, 2 (4). Rajendran, B., & Pandey, A. K. (2012). Contextual strategies for detecting spam in academic portals. (pp. 250 256). Berlin, Heidelberg: Springer Berlin Heidelberg. doi: 10.1007/978 3 642 27308 7_ 26 Reader, B., Stempel, G. H. I., & Daniel, D. K. (2004). Age, wealth, education predict letters to editor. Newspaper Research Journal, 25 (4), 55. Resources page. (n.d.). Retrieved from http://asn e.org/content.asp?contentid=19 Rich, C. (2004). Writing and reporting news: A coaching method Belmont: Thomson Wadsworth. Ruiz, C., Domingo, D., Mic, J. L., Daz Noci, J., Meso, K., & Masip, P. (2011). Public sphere 2.0? the democratic qualities of c itizen debates in online newspapers. The International Journal of Press/Politics, 16 (4), 463 487. doi: 10.1177/1940161211415849


82 Santana, A. D. (2012). Civility, anonymity and the breakdown of a new public sphere. University of Oregon). ProQuest Dissertat ions and Theses, 163. Retrieved from http://search.proquest.com/docview/1038153023?accountid=10920 (1038153023). Shoemaker, P. J. (2001). Individual and routine forces in gate keeping. Journalism and Mass Communication Quarterly, 78 (2), 233 246. doi: 10.1177/107769900107800202 Siapera, E. (2008). The political subject of blogs. Information Polity, 13 (1/2), 51. Singer, J. B., & Ashmanb, I. (2009). "Comment is free, but facts a generated content and ethical constructs at the guardian Journal of Mass Media Ethics, 24 (1), 3 21. doi: http://www.tandfonline.com/d oi/full/10.1080/08900520802644345#tabModule Slocum, F. (2001). White racial attitudes and implicit racial appeals: An experimental Politics & Policy, 29 (4), 650 669. doi: 10.1111/j.1747 1346.2001.tb00609.x Tecklenburg, J. (2012). Online comments: Let's try this again. Retrieved from http://thegazette.com/2012/09/12/online comments lets try again/ The histo r y of th e Orlando S entinel. (2004). Retrieved from http://articles.orlandosentinel.com/2004 01 01/features/04010 20323_1_sentinel communications morning sentinel sentinel star Thomas Paul Bonfiglio, & Jane H Hill. (2011). The everyday language of white racism Cambridge: Cambridge University Press. doi: 10.1017/S0047404511000753 Tribune company history. (n .d.). Retrieved from http://corporate.tribune.com/pressroom/?page_id=2313 Tribune company timeline. (n .d.). Retrieved from http://corporate.tribune.com/pressroom/?page_id=2315 Wahl Jorgensen, K. (2001). Letters to the editor as a forum for public deliberation: Modes of publicity and democratic debate. Critical Studies in Media Communication, 18 (3), 303 320. doi: 10.1080/07393180128085 Wahl Jorgensen, K. (2007). Journalists and the public: Newsroom culture, letters to the editor, and democracy Cresskill, N.J: Hampton Press. Webb, T. J. (2010). Verbal poison Criminalizing hate speech: A comparative analysis and a proposal for the American system. Washburn Law Journal, 50 445 482.


83 BIOGRAPHICAL SKETCH Antionette Rollins received a Master of Arts in Mass Communication from the University of Florida, with a specialization in journalism in the spr ing of 2013 She received a Bachelor of Arts in j ournalism from the University of Georgia in 2010, where she majored in m agazines. Her primary research interests include online journalism and user generated content