CONTENTS

EXECUTIVE SUMMARY

INTRODUCTION

OVER- AND UNDER-BLOCKING BY INTERNET FILTERS

• America Online Parental Controls
• BESS
• ClickSafe
• Cyber Patrol
• Cyber Sentinel
• CYBERsitter
• Family Click
• I-Gear
• Internet Guard Dog
• Net Nanny
• Net Shepherd
• Norton Internet Security
• Safe Server
• Safe Surf
• Smart Filter
• SurfWatch
• We-Blocker
• WebSENSE
• X-Stop

BIBLIOGRAPHY OF TESTS AND STUDIES

APPENDIX A: Blocked Sites by Subject: Artistic and Literary; Sexuality Education; Gay and Lesbian Information; Political Topics/Human Rights; Censorship

APPENDIX B: Blocking Categories for Different Filters Defined

EXECUTIVE SUMMARY

In the spring and summer of 2001, the Free Expression Policy Project of the National Coalition Against Censorship surveyed all of the studies and tests that it was able to locate describing the actual operation of 19 products or software programs that are commonly used to filter out World Wide Web sites and other communications on the Internet. This report summarizes the results of that survey. Its purpose is to provide a resource for policymakers and the general public as they grapple with the difficult, often hotly contested issues raised by the now-widespread use of Internet filters.

The existing studies and tests vary widely. They range from anecdotal accounts to extensive tests applying social-science methodologies. In some instances, we located only one or two test reports; in other cases-for example, Cyber Patrol, SmartFilter, and X-Stop-we found a great many. Most tests simply describe the actual sites that a particular product blocked when Web searches were conducted. Nearly every one, however, revealed massive over-blocking by filtering software.

This problem stems from the very nature of filtering, which must, because of the sheer number of Internet sites, rely to a large extent on mindless mechanical blocking through identification of key words and phrases. Where human judgment does come into play, filtering decisions are based on different companies' broad and varying concepts of offensiveness, "inappropriateness," or disagreement with the political viewpoint of the manufacturer. A few examples of over-blocking from the more than 70 studies or tests summarized in this report are:

• BESS blocked the home pages of the Traditional Values Coalition and Massachusetts Congressman Edward Markey.
• Cyber Patrol blocked MIT's League for Programming Freedom, part of the City of Hiroshima Web site, Georgia O'Keeffe and Vincent Van Gogh sites, and the monogamy-advocating Society for the Promotion of Unconditional Relationships.
• CYBERsitter blocked virtually all gay and lesbian sites and, after detecting the phrase "least 21," blocked a news item on the Amnesty International Web site (the offending sentence read, "Reports of shootings in Irian Jaya bring to at least 21 the number of people in Indonesia and East Timor killed or wounded").
• I-Gear blocked an essay on "Indecency on the Internet: Lessons from the Art World," the United Nations report "HIV/AIDS: The Global Epidemic," and the home pages of four photography galleries.
• Net Nanny, SurfWatch, Cybersitter, and BESS, among other products, blocked House Majority Leader Richard "Dick" Armey's official Web site upon detecting the word "dick."
• SafeSurf blocked the home pages of the Wisconsin Civil Liberties Union and the National Coalition Against Censorship.
• SmartFilter blocked the Declaration of Independence, Shakespeare's complete plays, Moby Dick, and Marijuana: Facts for Teens, a brochure published by the National Institute on Drug Abuse (a division of the National Institutes of Health).
• SurfWatch blocked such human-rights sites as the Commissioner of the Council of the Baltic Sea States and Algeria Watch, as well as the University of Kansas's Archie R. Dykes Medical Library (upon detecting the word "dykes").
• WebSENSE blocked the Jewish Teens page and the Canine Molecular Genetics Project at Michigan State University.
• X-Stop blocked the National Journal of Sexual Orientation Law, Carnegie Mellon University's Banned Books page, "Let's Have an Affair" catering company, and, through its "foul word" function, searches for Bastard Out of Carolina and "The Owl and the Pussy Cat."

INTRODUCTION

The still new, revolutionary medium of the Internet contains a wealth of information, images, and ideas-as the U.S. Supreme Court observed in 1997, "the content on the Internet is as diverse as human thought."¹ Unsurprisingly, not all of this online expression is accurate, pleasant, or inoffensive. Virtually since the arrival of the Internet, concerns have been expressed about minors' access to online pornography, about the proliferation of Web sites advocating racial hatred, and about other online content deemed to be offensive or dangerous. Congress and the states responded in the late 1990s with censorship laws, but most of these have been struck down by the courts. Partly as a result, individual parents, employers, school districts, and other government entities have turned with increasing frequency to privately manufactured Internet rating and filtering programs.

Early Internet filtering was based on either "self-rating" by those who published online communi-cations; or "third-party rating" by filter manufacturers. Because of the Internet's explosive growth (now more than a billion Web sites, many of which change daily), and the consequent inability of filtering companies to review and evaluate even a fraction of it, third-party rating had to rely largely on mechanical blocking by key words or phrases such as "over 18," "breast," "sex," or "pussy." The results were not difficult to predict: large quantities of valuable information and literature, particularly about sexuality, feminism, gay and lesbian issues, civil rights, and other politically important subjects, were blocked.

Even where company employees did review Web sites, there arose massive problems of subjectivity. The political attitudes of the different filter manufacturers were reflected in blocking decisions, particularly with respect to such subjects as homosexuality, human rights, and criticism of filtering software. The alternative, self-rating, did not suffer these disadvantages, but it proved impossible to persuade the great majority of online speakers to self-rate their sites. Online news organizations, for example, are among those that steadfastly refused to reduce their content to decontextualized, simplistic letters or codes through self-rating.

Third-party rating and filtering systems have thus become the industry standard, at least in the United States. Private software companies actively market such products as SurfWatch and Cyber Patrol, which contain multiple categories of potentially offensive, "inappropriate," or "objectionable" material. Internet service providers such as America Online provide "parental control" options that block Web sites based on technological word or phrase identification, augmented by the company's-or its subcontractor's-judgments about age-appropriateness. Some manufacturers market products that essentially block all of the Internet, with only a few hundred or thousand preselected sites accessible (so-called whitelists). One company-later the subject of a First Amendment lawsuit-erroneously claimed that its "X-Stop" software was able to identify and block only "illegal" obscenity and child pornography: an impossible task, since legal judgments in both categories are subjective, and under the Supreme Court's three-part obscenity test, determinations of legality vary depending on different communities' standards of "prurience" and "patent offensiveness."²

The late 1990s saw political battles in many communities over the use of filtering products in public libraries. New groups such as Family Friendly Libraries attacked the American Library Association (ALA) for adhering to a no-censorship and no-filtering policy, even for minors. (The ALA and other champions of intellectual freedom objected to the over-blocking propensities of filtering software, and advocated noncensorial approaches such as privacy screens and "acceptable use" policies.) Online anti-censorship groups such as the Censorware Project and Peacefire began to publish reports documenting the blocking of numerous valuable, educational sites by different filters. In December 2000, Congress passed the Children's Internet Protection Act ("CIPA"), mandating filters in all schools and libraries that receive federal financial assistance through the E-rate or "universal service" program, or through the Library Services and Technology Act.³ This amounted to about 60% of the nation's libraries and public schools.

Thus, although initially promoted as a voluntary alternative to coercive government censorship, Internet filtering is now embraced by government at both the federal and local levels. Reports of over-blocking, of vague and subjective standards, and of politically biased blocking decisions continue, while industry spokespersons assert that their methodologies are improving and that new software programs designed to distinguish between acceptable and unacceptable material will soon be on the market. But no filtering technology, no matter how sophisticated, can make contextualized judgments about the value, offensiveness, or age-appropriateness of online expression.

Internet filtering has thus become a major public policy issue, and is likely to remain so. In the interests of advancing informed debate on this important issue, the Free Expression Policy Project has collected and summarized all of the studies and tests that it has been able to locate on the actual operation of Internet filters. The report presents this information in one place and in readily accessible form, so that the ongoing policy debate will be better informed about what Internet filters actually do, and their ultimate impact on free expression.

The report is organized by filtering product. Necessarily, there is some overlap, since many studies have sampled more than one product. A bibliography of all the studies is included, along with an appendix listing blocked sites according to subject: artistic and literary sites; sexuality education; gay and lesbian information; political topics; and sites relating to censorship itself. (Another appendix, describing the blocking categories used by different products, is available in the online version of this report.)

Where the study gives Web addresses or URLs, we have included these and checked their accuracy whenever possible. (Some Web addresses are now obsolete.) If we have not given Web addresses, it is because they were not supplied in the underlying report.

We hope that Internet Filters: A Public Policy Report will prove a useful resource for policymakers, parents, teachers, librarians, and all others concerned with the Internet, intellectual freedom, or the education of youth. Internet filtering is popular, despite its unreliability, because many parents, political leaders, and educators feel that the alternative-unfettered Internet access-is even worse. But to make these policy choices, it is necessary to have complete and accurate information about what filters actually do. Ultimately, less censorial approaches such as media literacy, sexuality education, and Internet acceptable-use training may be better policy choices than Internet filters in addressing concerns about young people's access to "inappropriate" content or disturbing ideas.

OVER- AND UNDER-BLOCKING BY INTERNET FILTERS

America Online Parental Controls

AOL offers three levels of Parental Controls: "Kids Only," for children aged 12 and under, "Young Teen," for ages 13-15, and "Mature Teen," for ages 16-17, which allows access to "all content on AOL and the Internet, except certain sites deemed for an adult (18+) audience." At one time AOL employed Cyber Patrol's block list; at another point it stated it was using SurfWatch. While as of 2001 the Parental Controls information page provided no specific information as to its filtering categories or methodology other than its use of a user-recommended database of sites, on May 2, 2001, AOL announced that Parental Controls had integrated the RuleSpace Company's "Contexion Services," which identifies "objectionable" sites "by analyzing both the words on a page and the context in which they are used."⁴

Access Denied, Version 2.0: The Continuing Threat Against Internet Access and Privacy and its Impact on the Lesbian, Gay, Bisexual and Transgender Community, Gay and Lesbian Alliance Against Defamation (GLAAD), 1999.
This 1999 report was a follow-up to GLAAD's 1997 report Access Denied: The Impact of Internet Filtering Software on the Lesbian and Gay Community, which described the potential defects of various filtering products without identifying particular blocked sites. Access Denied, Version 2.0 addressed AOL Parental Controls only in its introduction, where it reported that AOL's "Kids Only" setting blocked the Web site of Children of Lesbians and Gays Everywhere (COLAGE), as well as a number of "family, youth and national organization Web sites with lesbian and gay content," none of which were specifically named or described in the report.

Brian Livingston, "AOL's 'youth filters' protect kids from Democrats," CNet News.com, Apr. 24, 2000.
This news report described Livingston's investigation of AOL's blocking decisions for signs of political bias. He found that the "Kids Only" setting blocked the Web sites of the Democratic National Committee (www.democrats.org), the Green Party (www.greens.org), and Ross Perot's Reform Party (www.reformparty.org), but not those of the Republican National Committee (www.rnc.org) and the conservative Constitution (www.ustaxpayers.org) and Libertarian (www.lp.org) parties. Livingston also reported that AOL's "Young Teen" setting blocked the home pages of the Coalition to Stop Gun Violence (www.gunfree.org), Safer Guns Now (www.safergunsnow.org), and the Million Mom March (www.millionmommarch.com), but neither the NRA site (www.nra.org) nor the commercial sites for Colt (www.colt.com) and Browning (www.browning.com) firearms.

"AOL Parental Controls error rate for the first 1,000 .com domains," Peacefire, Oct. 23, 2000.
Peacefire Webmaster Bennett Haselton selected 1,000 dot-com domains he had compiled for a similar test of SurfWatch 2 months earlier (see p. 39), and attempted to access each site on AOL 5.0 adjusted to its "Mature Teen" setting. Five of the 1,000 working domains were blocked, including a-aji.com, a site on which vinegar and seasonings were sold. Haselton decided the 4 others were "pornographic" and thus accurately blocked. This produced an "error rate" of 20%, the lowest, by Peacefire's calculation, of the 5 filters it tested. AOL also "blocked far fewer pornographic sites than any of the other programs," however. Haselton stated that 5 blocked domains was an insufficient sample to gauge the efficacy of AOL Parental Controls accurately, and that the true error rate could fall anywhere between 5 and 75%.

"Digital Chaperones for Kids," Consumer Reports, Mar. 2001.
Consumer Reports published its assessments of AOL's "Young Teen" and "Mature Teen" settings in this review of various filtering technologies. Through each, Consumer Reports attempted to access 86 Web sites it deemed objectionable because they contained "sexually explicit content or violently graphic images" or promoted "drugs, tobacco, crime, or bigotry," and 53 it deemed legitimate because they "featured serious content on controversial subjects." The "Mature Teen" setting left 30% of the "objectionable" sites unblocked; the "Young Teen" filter failed to block 14%-the lowest such error rate of all products reviewed as far as underinclusive filtering was concerned. But "Young Teen" also blocked 63% of the "legitimate" sites, including Peacefire.org, Lesbian.org, an online guide to lesbian politics, history, arts, and culture, the Web sites of the Citizens' Committee for the Right to Keep and Bear Arms and the Southern Poverty Law Center, and SEX, Etc., a sex-education site written by and for teenagers, and hosted by Rutgers University.

Miscellaneous Reports
— In "BabelFish blocked by censorware" (Feb. 27, 2001), Peacefire reported that AOL's "Mature Teen" setting barred access to Babel-ish (babelfish.altavista.com), AltaVista's foreign-language translation service.

BESS

BESS, manufactured by N2H2, provides its Internet-filtering services in one of 2 ways: either as a proxy server, whereby each Web request is passed through a server located at N2H2 itself, or in the form of a dedicated server called the "Internet Filtering Manager," installed on a local computer or system. Dedicated-server administrators can enable or disable any of BESS's blocking categories, as well as BESS's keyword filtering features; users on BESS proxy servers cannot. In both scenarios, BESS provides 29 categories of blocked content under its "Typical School Filtering" setting, ranging from "Adults Only" and "Alcohol" to "Gambling," "Lingerie," "Personals," and "Tasteless/Gross." (See appendix B for a complete list.) N2H2 states that 4 of the 29 classifications-"History," "Medical," "Moderated," and "Text/Spoken Only"-are designed to distinguish between sites falling squarely into BESS's blocking categories and those that may contain sexually oriented, violent, or other question-able content but also some educational merit, such as the Starr report to Congress on President Clinton's sexual transgressions.

Under the "Maximum Filtering" setting, all 29 categories, as well as employment sites, message and bulletin boards, investment-related sites, images of individuals wearing swimsuits, and all Web searches are blocked. Configured for "Minimal Filtering," N2H2's Internet Filtering Manager blocks sites falling into the categories of "Adults Only," "Hate/ Discrimination," "Illegal," "Pornography," "Sex," and "Violence."

Karen Schneider, A Practical Guide to Internet Filters, 1997.
From April to September 1997, Karen Schneider supervised a nationwide team of librarians in testing 13 filtering technologies, including BESS. The results of the Internet Filter Assessment Project, or TIFAP, were published later that year in Schneider's Practical Guide to Internet Filters.

The researchers began by seeking answers to some 100 common research queries on the Web, on both unfiltered computers and ones equipped with BESS (and the various other filters) configured for maximum blocking, including keyword blocking. Each query fell into one of 11 categories: "sex and pornography," "anatomy," "drugs, alcohol, and tobacco," "gay issues," "crimes (including pedophilia and child pornography)," "obscene or 'racy' language," "culture and religion," "women's issues," "gambling," "hate groups and intolerance," and "politics." The queries were purposely devised to gauge filters' handling of controversial issues-for instance, "I'd like some information on safe sex"; "I want to do some research on Robert Mapplethorpe"; "I want information on the legalization of marijuana"; "I want information on PFLAG" [Parents and Friends of Lesbians and Gays]; "Is the Aryan Nation the same thing as Nazis?"; and "Who are the founders of the Electronic Frontier Foundation and what does it stand for?" In some cases, the queries contained potentially provocative terms "intended to trip up keyword-blocking mechanisms" such as "How do beavers make their dams?"; "Can you find me some pictures from Babes in Toyland?"; "I need some information and a picture of the Enola Gay"; "I'm a farmer and want to research rape-the plant used to make canola oil"; and "I'm trying to find out about the Paul Newman movie The Hustler."

Schneider used Web sites, blocked and unblocked, that arose from these searches to construct her testing sample of 240 URLs. Researchers tested these URLs against a version of BESS configured for "Maximum Filtering," but with keyword filtering disabled. TIFAP found that "several" (Schnieder did not say how many) nonpornographic sites were blocked, including a page discussing X-rated videos but not containing any pornographic imagery, and an informational page on trichomaniasis, a vaginal disease. Upon notification and review, BESS later unblocked the trichomaniasis site. A Practical Guide included neither the names nor the Web addresses of the blocked sites.

Passing Porn, Banning the Bible: N2H2's Bess in Public Schools, Censorware Project, 2000.
From July 23-26, 2000, the Censorware Project tested "thousands" of URLs against 10 BESS proxy servers, 7 of which were in use in various public schools across the United States. Among the blocked Web sites were Friends of Lulu (friends-lulu.org), a site promoting comic books to girls, a page from Mother Jones magazine's site (bsd.mojones.com/hellraiser_central), the Institute of Australasian Psychiatry (www.iap.org.au); the nonprofit effort Stop Prisoner Rape (www.spr.org), and a portion of the Columbia University Health Education Program site on which users are invited to submit "questions about relationships; sexuality; sexual health; emotional health; fitness; nutrition; alcohol, nicotine, and other drugs; and general health" (www.goaskalice.columbia.edu). BESS also blocked several sites opposing censorship, including the Web site of the United Kingdom-based Feminists Against Censorship (www.fiawd.demon.co.uk/FAC), the personal site of a librarian opposing Internet-filter use in libraries (burn.ucsd.edu/~mail/library), and Time magazine's "Netly News," which has reported, positively and negatively, on filtering software.

The report noted that BESS does not (as is implied in its published filtering criteria) review home pages hosted by such free site providers as Angelfire, Geocities, and Tripod (owing, it seems, to their sheer number). Instead, users must configure the software to block none or all of these sites; some schools opt for the latter, thus prohibiting access to such sites as The Jefferson Bible (www. angelfire.com/co/JeffersonBible), a com-pendium of Biblical passages selected by Thomas Jefferson, and the Web site of the Eustis Panthers (www.angelfire.com/fl/eustispanthers), a high-school baseball team. Though each proxy was configured to filter out pornography to the highest degree, Censorware was able to access "hundreds" of pornographic Web sites, of which 46 are listed in Passing Porn. Of the total unblocked pornographic URLs, some 285 were listed on Yahoo.com, and of these, 28 were accessible through all 7 of the proxies in use in public schools.

"'BESS, the Internet Retriever' Examined," Peacefire, 2000.
This report consists of a list of 15 sites that Peacefire deemed inappropriately blocked by BESS during the first half of 2000. These included Peacefire.org itself, which was blocked for "Profanity" when the word "piss" appeared on the site (within a quotation from a letter written by Brian Milburn, president of CYBERsitter's manufacturer, Solid Oak Software, to journalist Brock Meeks). Also blocked were two portions of the Web site of Princeton University's Office of Population Research, both resources on contraception methods (information on emergency contraception pills was found on opr.princeton.edu/ec/ecp.html; information on IUDs on opr.princeton.edu/ec/eciud.html); the Safer Sex page (www.safersex.org); 5 gay-interest sites, including the home page of the Illinois Federation for Human Rights (www.suba.com/~ifhr/ifhr.html), which "works to preserve the equal rights of lesbian and gay Illinoisians," and 2 online magazines devoted to gay topics (www.youth.org/loco/quirx/index.html and www.oasismag.com); 2 Web sites providing resources on eating disorders (members.aol.com/edapinc/home.html and www.stud.unit.no/studorg/ ikstrh/ed); and 3 sites discussing breast cancer (www.undnj.edu/univhosp/d163.html, members.aol.com/BCLEGIS/index.html, and www.blooberry.com/bformfaq).⁵

"Mandated Mediocrity: Blocking Software Gets a Failing Grade," Peacefire & Elec-tronic Privacy Information Center (EPIC), Oct. 2000.
"Mandated Mediocrity" describes another 23 Web sites inappropriately blocked by BESS. The URLs were tested against an N2H2 proxy as well as a trial copy of the N2H2 Internet Filtering Manager set to "Typi-cal School Filtering." Among the blocked sites were the home page of the Traditional Values Coalition (www.traditionalvalues.org); Hillary for President (www.hillary4president.org); The Smoking Gun (www.smokinggun .com), an online selection of primary documents relating to current events ("obtained from government and law enforcement sources, via Freedom of Information requests, and from court files nationwide"); a selection of travel photographs of Utah's national parks (meltingpot.fortunecity.com/offord/719/herbsutah.htm); "What Is Memorial Day?" (members.aol.com/ceharger/what_is_memorial_day.htm), an essay lamenting the "capitalistic American" conception of the holiday as nothing more than an occasion for a 3-day weekend; the home page of "American Government and Politics," a course at St. John's University (users.aol.com/drblw/homepage.htm); and the Circumcision Information and Research Pages (www.cirp.org), a site that contained no nudity and was designated a "Select Parenting Site" by ParenthoodWeb.com.

"BESS error rate for 1,000 .com domains," Peacefire, Oct. 23, 2000.
This October 2000 test involving a sample of 1,000 active dot-com domains has already been described (see p. 6). As N2H2 evidently reviewed Peacefire's earlier SurfWatch report and prepared for a similar test of its own software by unblocking any of the 1,000 sites inappropriately filtered by BESS,⁶ Peacefire selected the second 1,000 dot-com domains for testing against a BESS proxy server in use at a school where Peacefire had found a student to help test BESS's performance.

The filter was configured to block sites in the categories of "Adults Only," "Alcohol," "Chat," "Drugs," "Free Pages," "Gambling," "Hate/Discrimination," "Illegal," "Lingerie," "Nudity," "Personals," "Personal Information," "Porn Site," "Profanity," "School Cheating Info," "Sex," "Suicide/Murder," "Tasteless/Gross," "Tobacco," "Violence," and "Weapons." The program's keyword-blocking features were also enabled. The BESS proxy blocked 176 of the 1,000 domains; among these, 150 were "under construction." Of the remaining 26 sites, Peacefire deemed 7 wrongly blocked: a-celebrity.com, a-csecurite.com, a-desk.com, a-eda.com, a-gordon.com, a-h-e.com, and a-intec.com.

The report said the resulting "error rate" of 27% was unreliable given how small a sample was examined; the true error rate "could be as low as 15%." Peacefire's Bennett Haselton also noted that the dot-com domains tested here were "more likely to contain commercial pornography than, say, .org domains. ... [W]e should expect the error rate to be even higher for .org sites" (Haselton's emphasis), and added that the results called into question N2H2 CEO Peter Nickerson's claim, in 1998 testimony before the House Subcommittee on Telecommunications, Trade, and Consumer Protection, that "[a]ll sites that are blocked are reviewed by N2H2 staff before being added to the block lists."⁷

"Blind Ballots: Web Sites of U.S. Political Candidates Censored by Censorware," Peacefire, Nov. 7, 2000.
"Blind Ballots" was published on Election Day, 2000. Peacefire obtained a random sample of U.S. political candidates' Web sites from NetElection.org, a site providing information on political campaigns nationwide, and set out to see which sites BESS's (and Cyber Patrol's) "Typical School Filtering" would allow users to access. (Around the start of the 2000 school year, BESS and Cyber Patrol asserted that together they were providing filtered Internet access to more than 30,000 schools nationwide.⁸)

BESS's wholesale blocking of free Webpage hosting services caused the sites of one Democratic candidate, 5 Republicans, 6 Libertarians (as well as the entire Missouri Libertarian Party site), and 13 other third-party candidates to be blocked. Report coauthor Bennett Haselton commented that, as "many of our political candidates run their campaigns on a shoestring, and use free-hosting services to save money," BESS's barring of such hosts leads it to an inadvertent bias toward wealthy or established politicians' sites. Congressional incumbent Edward Markey (a Democrat from Massachusetts), also had his site (www.edmarkey.org) blocked-unlike the others, it was not hosted by Geocities or Tripod, but was blocked because BESS categorized its content as "Hate, Illegal, Pornography, and/or Violence." "While blocking software com-panies often justify their errors by pointing out that they are quickly corrected," Haselton wrote, "this does not help any of the candidates listed above. . . . [C]orrections made after Election Day do not help them at all."

"Amnesty Intercepted: Global Human Rights Groups Blocked by Web Censoring Software," Peacefire, Dec. 12, 2000.
In response to complaints from students barred from the Amnesty International Web page, among others, at their school computer stations, Peacefire undertook an examination of various filters' treatment of human rights sites. Peacefire found that BESS's "Typical School Filtering" blocked the home pages of the International Coptic Congress (www.copts.com), which tracked human rights violations against Coptic Christians living in Egypt; and Friends of Sean Sellers (www.seansellers.com), which contained links to the works of the Multiple Personality Disorder-afflicted writer who was executed in 1999 for murders he had committed as a 16-year-old (the site opposed capital punishment). "Typical school filtering" also denied access to the official sites of recording artists Suzanne Vega (www.vega.net) and the Art Dogs (www.artdogs.com); both contained statements that portions of their proceeds would be donated to Amnesty International. Peacefire also reported that BESS's "Minimal Filtering" configuration blocked the Web sites of Human Rights & Tamil People (www.tamilrights.org), which tracks govern-ment and police violence against Hindu Tamils in Sri Lanka, and Casa Alianza (www.casa-alianza.org), which documents the condition of homeless children in the cities of Central America.

Miscellaneous Reports
— In its survey of "Winners of the Foil the Filter Contest" (Sept. 28, 2000), the Digital Freedom Network reported that BESS blocked House Majority Leader Richard "Dick" Armey's official Web site (armey.house.gov) upon detecting the word "dick."
— Peacefire reported, in "BabelFish blocked by censorware" (Feb. 27, 2001), that BESS blocked the URL-translation site BabelFish (babelfish.altavista.com).
— In "Teen Health Sites Praised in Article, Blocked by Censorware" (Mar. 23, 2001), Peacefire noted that BESS blocked portions of TeenGrowth (www.teengrowth.com), a teen-oriented health education site that was recognized by the New York Times in the recent article, "Teenagers Find Health Answers with a Click."⁹

ClickSafe

Rather than relying on lists of objectionable URLs, ClickSafe is designed to review each requested page in real time. According to company cofounder Richard Schwartz's outline for testimony submitted to the commission created by the 1998 Child Online Protection Act (the COPA Commission) in 2000, ClickSafe "uses state-of-the-art, content-based filtering software that combines cutting edge graphic, word and phrase-recognition technology to achieve extra-ordinarily high rates of accuracy in filtering pornographic content," and "can precisely distinguish between appropriate and inappropriate sites."

"Sites blocked by ClickSafe," Peacefire, July 2000.
Upon learning that ClickSafe blocked the home page of cyberlaw scholar Lawrence Lessig (www.thestandard.com/people/display.0,1157,1739,00.html), who was to testify before the COPA Commission, Peace-fire attempted to access various pages on the COPA Commission site, as well as the Web sites of organizations and companies with which the commissioners were affiliated, through a computer equipped with ClickSafe. On the COPA Commission's site, ClickSafe blocked the Frequently Asked Questions page (www.copacommission.org/commission/faq.shtml); the biographies of commission members Stephen Balkam (www.copacommission.org/commission/balkam .shtml), Donna Rice Hughes (www.copacommission.org/commission/hughes.shtml), and John Bastian (www.copacommission.org/commission/bastian.shtml); a list of "technologies and methods within the scope" of the commis-sion's inquiry (www.copacommission.org/ commission/technologies.shtml); the commission's Scope and Timeline Proposal (www.copacommission.org/commission/scope.shtml); and two versions of the statute itself (www.copacommission.org/commission/original.shtml and www.copacommission.org/commission/amended.shtml).

As for groups with representatives on the commission, Peacefire found that ClickSafe blocked several organizations' and companies' sites, at least partially: Network Solutions (www.networksolutions.com/legal/service-agreement.html); the Internet Content Rating Association (www.icra.org/about.html); Security Software's information page on its signature filtering product, Cyber Sentinel (www.securitysoft.com/cyber-page.html); FamilyConnect (www.familyconnect.com/block.html), a brand of blocking software-the page blocked was one on which users could submit URLs to be reviewed as potential blocks or unblocks; the National Law Center for Children and Families (www.nationallawcenter.org/federal.htm); the Christian site Crosswalk.com; and the Center for Democracy and Technology (www.cdt.org). In addition to the CDT, ClickSafe blocked the home pages of the ACLU (www.aclu.org), the Electronic Frontier Foundation (www.eff.org), and the American Family Association (www.afa.net), as well as part of the official site of Donna Rice Hughes's book, Kids Online: Protecting Your Children in Cyberspace (www.protectkids.com/app.a.html).

Cyber Patrol

Cyber Patrol, currently owned by Surf-Control, operates with 12 default blocking categories, including "Partial Nudity," "Intoler-ance," "Drugs/Drug Culture," and "Sex Education." (See appendix B.) According to the manufacturer's Web site, "Cyber Patrol employs a team of professional researchers at least 21 years of age including parents and teachers" to determine whether sites are to be blocked. Any page that "contains more than 3 instances in 100 messages or any easily accessible pages with graphics, text or audio that fall within the definition" of any of the 12 categories "will be considered sufficient to place the source in that category." As with most filtering products, Cyber Patrol's list of prohibited sites is not made public, but SurfControl offers the CyberNOT search engine, a feature on its Web site through which users can enter URLs and receive immediate responses as to whether or not those pages are on the filter's block list. SurfControl adds, "Internet sites that contain information or software programs designed to hack into filtering software, including Cyber Patrol, are added to the CyberNOT list in ALL categories as a measure of protection for the parents, educators and businesses that rely on Cyber Patrol to screen Internet content."

Brock Meeks and Declan McCullagh, "Jacking in from the 'Keys to the Kingdom' Port," CyberWire Dispatch, July 3, 1996.
The first evaluation of Cyber Patrol appeared in this early report on the problems of Internet filtering by journalists Brock Meeks and Declan McCullagh. Meeks and McCullagh viewed a decrypted version of Cyber Patrol's block list (along with those of CYBERsitter and Net Nanny), and noticed that Cyber Patrol stored the Web addresses it blocked only partially, cutting off all but the first 3 characters at the end of a URL. For instance, the software was meant to block loiosh.andrew.cmu.edu/~shawn, a Carnegie Mellon student home page containing information on the occult; yet on its block list Cyber Patrol recorded only loiosh.andrew.cmu.edu/ ~sha, thereby blocking every site beginning with that URL segment and leaving, at the time of the report's publication, 23 unrelated sites on the CMU server blocked.

The authors also found that with all default categories enabled, Cyber Patrol barred multiple sites concerning cyberliberties-the Electronic Frontier Foundation's censorship archive, for example, and the home page of MIT's League for Programming Freedom. Also blocked were the Queer Resources Directory, which counts among its resources information from the Centers for Disease Control and Prevention, the AIDS Book Review Journal, and AIDS Treatment News. Cyber Patrol also blocked a number of newsgroups dealing with homosexuality and gender issues, such as alt.journalism.gay-press, soc.support.youth.gay-lesbian-bi, alt.feminism, and soc.feminism, as well as soc.support.fat-acceptance.

Karen Schneider, A Practical Guide to Internet Filters, 1997.
The Internet Filter Assessment Project tested Cyber Patrol configured to block only "Full Nudity" and "Sexual Acts." Schneider reported that the software "blocked 'good sites' 5-10% of the time, depending on the tester, and pornographic sites slipped through about 10% of the time." One of the "good sites" was www.disinfo.com, described by Schneider as a site "devoted to debunking propaganda."

"Cyber Patrol: The Friendly Censor," Censorware Project, Nov. 22, 1997.
Censorware Project member Jonathan Wallace tested his personal collection of approximately 270 Web sites on ethics, politics, and law-all "containing controversial speech but no obscenity or illegal material"-against the CyberNOT search engine after learning that the Web pages of Sex, Laws, and Cyberspace, the 1996 book he co-authored with Mark Mangan, were blocked by Cyber Patrol. Wallace found 12 of his chosen sites were barred, including Deja News (www.dejanews.com), a searchable archive of USENET materials,¹⁰ and the Web page of the Society for the Promotion of Unconditional Relationships (dspace.dial.pipex.com/ town/estate/xgv92/spur2.htm), an organization advocating monogamy whose site includes such articles as "The Role of Faith in Relationships."

Wallace reported that Cyber Patrol also blocked several sites featuring politically loaded content, such as the Flag Burning Page (formerly www.indirect.com/user/warren/flag.html; now www.esquilax.com/flag), which examines the issue of flag burning from a constitutional perspective; Interactivism (www.interactivism.com), a site inviting users to engage in political activism by corresponding with politicians on issues such as campaign-finance reform and Tibetan independence; Newtwatch (no longer active; formerly www.cais.com/newtwatch), a Democratic Party-funded page that consisted of reports and satires on the former Speaker of the House; Dr. Bonzo, another now-inactive page (www.iglou.com/drbonzo/anathema .htm), which featured "satirical essays on religious matters"¹¹; and the Web site of the Second Amendment Foundation (www.saf.org)-though, as Wallace noted, Cyber Patrol did not block other gun-related sites, such as that of the National Rifle Association.

"Gay sites netted in Cyber Patrol sting," press release, GLAAD, Dec. 19, 1997.
Cyber Patrol's evident bias against homosexuals was reported by the Gay and Lesbian Alliance Against Defamation (GLAAD) in a Dec. 1997 press release stating that Cyber Patrol was blocking the entire "WestHollywood" subdirectory of Geocities. WestHollywood, at that time, was home to more than 20,000 gay- and lesbian-interest sites, such as that of the National Black Lesbian and Gay Leadership Forum's Young Adult Program. When contacted, Cyber Patrol's then-manufacturer Microsystems Software cited, by way of explanation, the high potential for WestHollywood sites to contain nudity or pornographic imagery. GLAAD's press release pointed out, however, that Geocities expressly prohibited "nudity and pornographic material of any kind" on its server.

Microsystems CEO Dick Gorgens re-sponded to further inquiry with the admission that GLAAD was "absolutely correct in [its] assessment that the subdirectory block on WestHollywood is prejudicial to the Gay and Lesbian Geocities community. . . . Over the next week the problem will be corrected." Yet according to the press release, after a week had passed, the block on WestHollywood remained.

Blacklisted by Cyber Patrol: From Ada to Yoyo, Censorware Project, Dec. 22, 1997.
This report documented a number of sites that the Censorware Project considered wrongly blocked in the "Full Nudity" and "Sexual Acts" categories. Among them were Creature's Comfort Pet Service (www.creaturescomfort.com/main.html); Air Penny (BeInMyPoster.com), a Nike site devoted to basketball player Penny Hardaway; the MIT Project on Mathematics and Computation (www-swiss.ai.mit .edu); AAA Whole-sale Nutrition (pas-fitness.com); the site of the National Academy of Clinical Biochemistry (nacb.org); the online edition of Explore Underwater magazine (www.exploreuw.com); the computer science department of England's Queen Mary and Westfield College (murphys.dcs.qmw.ac.uk); and the Web site of the United States Army Corps of Engineers Construction Engineering Research Laboratories (pandora.cecer.army .mil). The report took its title from 2 additional sites blocked for "Full Nudity" and "Sexual Acts": "We, the People of Ada" (www.ada-twp.org), an Ada, Michigan, committee devoted to "bring[ing] about a change for a more honest, fiscally responsible and knowledgeable township government," and Yoyo (yoyo.cc.monash.edu.au), a server of Melbourne's Monash University.

Blacklisted also catalogued 4 nonpornographic sites "oriented toward alternative sexuality" that were blocked for "Full Nudity" and "Sexual Acts": Gay Daze (gaydaze.com), "a sort of online soap opera" centered around 6 homosexual characters; Gay Mart, an online catalog merchant selling such items as gay-themed baseball caps and cookie jars; the home page of a West Hollywood coffee dealer called Stonewall, Inc. (www.stonewallinc.com); and, as was reported 3 days earlier by GLAAD, the WestHollywood subdirectory of Geocities. In addition, Blacklisted reported that every site hosted by the free Web page provider Tripod (members.tripod.com) was barred, not only for nudity or sexually explicit content, but also for "Violence/Profanity," "Gross Depictions," "Intolerance," "Satanic/Cult," "Drugs/Drug Culture," "Militant/Extreme," "Questionable/ Illegal & Gambling," and "Alcohol & Tobacco." Tripod was home, at the time of the report, to 1.4 million distinct pages, but smaller servers and service providers were also blocked in their entirety-Blacklisted lists 40 of them. Another section of the report lists hundreds of blocked newsgroups, including alt.atheism, alt.adoption, alt.censorship, alt.journalism, rec.games.bridge (for bridge enthusiasts), and support.soc.depression.misc (on depression and mood disorders).

The day after Blacklisted was published, Microsystems Software unblocked 55 of the 67 URLs and domains the report had cited. Yet 8 of the remaining 12, according to the Censorware Project, were still wrongly blocked: Nike's Penny Hardaway site, the National Academy of Biochemistry sites, 4 Internet s ervice providers (phantom.datamg.com, www.dada.it, www.ctsserver.com, and thorgal.globalxs.nl), Tripod, and a site-in-progress for a software company (www.rotw.com). This last site, at the time of Censorware's Dec. 25, 1997, update to Blacklisted, contained very little content, but did contain the words "HOT WEB LINKS"-which was "apparently enough for Cyber Patrol to continue to block it as pornography through a second review." Of the 4 other sites left blocked, 2, Censorware acknowledged, fell within the Microsystems Software's blocking criteria and "shouldn't have been listed as wrongful blocks originally."

Christopher Hunter, "Filtering the Future?: Software Filters, Porn, PICS, and the Internet Content Conundrum," July 1999.
In June 1999, Christopher Hunter of the University of Pennsylvania's Annenberg School for Communication tested 200 URLs against 4 widely used Internet filters, including Cyber Patrol. Contending that existing reports on blocked sites applied "largely unscientific methods" (that is, they did not attempt to assess overall percentages of "wrongly" blocked sites), Hunter tested Cyber Patrol, CYBERsitter, Net Nanny, and SurfWatch by "social science methods of randomization and content analysis."

Hunter intended half of his testing sample to approximate an average Internet user's surfing habits. Thus, the first 100 sites consisted of 50 "randomly generated" by Webcrawler's random links feature and 50 others Hunter compiled through Altavista searches for the 5 most frequently requested search terms as of April 1999: "yahoo," "warez" (commercial software products made obtainable for illegal download), "hotmail," "sex," and "MP3"¹². Hunter gathered the first 10 matches from each of these 5 searches.

For the other 100 sites, Hunter focused on material often identified as bases of controversial blocks. He therefore added to his testing sample the Web sites of the 36 plaintiff organizations in ACLU v. Reno and ACLU v. Reno II, the American Civil Liberties Union's challenges to the 1997 Communications Decency Act and the 1998 Child Online Protection Act, respectively. Hunter then conducted Yahoo searches for sites pertaining to Internet portals, political issues, feminism, hate speech, gambling, religion, gay pride and homosexuality, alcohol, tobacco, and drugs, pornography, news, violent computer games, safe sex, and abortion. From each of the first 12 of these 13 searches, Hunter chose 5 of the resulting matches for his sample, and then selected 4 abortion-related sites (2 pro- and 2 anti-) in order to arrive at a total of 100 URLs.

Hunter evaluated the first page of each site using the Internet rating system devised by the Recreational Software Advisory Council (called RSACi). Under RSACi's 4 categories (violence, nudity, sex, and language) and 5 grades within each category, a site with a rating of zero in the "sex" category, for example, would contain no sexual content or else only "innocent kissing; romance," while a site with a "sex" rating of 4 might contain "explicit sexual acts or sex crimes." Using these categories, Hunter made his own judgments as to whether a filtering product erroneously blocked or failed to block a site, characterizing a site whose highest RSACi rating he thought would be zero or one as nonobjectionable, while determining that any site with a rating of 2, 3, or 4 in at least one RSACi category should have been blocked.

Hunter concluded that Cyber Patrol blocked 20, or 55.6%, of the sites containing material he deemed objectionable according to RSACi standards, and 15, or 9.1%, of sites he deemed innocuous. Among these 15 sites were the feminist literary site RiotGrrl (www.riotgrrl.com); the home page of Stop Prisoner Rape (www.igc.apc.org/spr); the Qworld contents page (www.qworld.org/TOC.html), a collection of links to online gay-interest resources; an article on "Promoting with Pride" on the Queer Living page (www.qmondo.com/queerliving); the Web site of the Coalition for Positive Sexuality, or CPS (www.positive.org/Home/index.html), which promotes "complete and honest sex education"; SIECUS, the Sexuality Information and Education Council of the United States (www.siecus.org); and Gay Wired Presents Wildcat Press (www.gaywired.com/wildcat/index), a page devoted to an award-winning independent press. Five of the other sites Hunter deemed over-blocked, however, were alcohol- and tobacco-related promotional sites, and thus did fall within Cyber Patrol's filtering criteria. On the other hand, similar alcohol and tobacco sites were left unblocked.

In Feb. 2000, filtering advocate David Burt responded to Hunter's study with a press release citing potential sources of error.¹³ Burt argued that "200 sites is far too small to adequately represent the breadth of the entire world wide web" and charged that all but the 50 randomly generated URLs constituted a skewed sample, containing content "instantly recognizable as likely to trigger filters" and "not represented in the sample proportionately to the entire Internet," thus giving rise to "much higher-than-normal error rates." A more serious problem, however, is that in attempting to arrive at "scientific" estimates of percentages of wrongly blocked sites, Hunter relied on his own subjective judgments of appropriateness.

Youth Access to Alcohol and Tobacco Web Marketing: The Filtering and Rating Debate, Center for Media Education, Oct. 1999.
In April 1999, Center for Media Education researchers tested Cyber Patrol (along with 5 other filters) for under-inclusive blocking of alcohol and tobacco marketing materials. They first selected the official sites of 10 beer manufacturers (such as www.budweiser.com) and 10 liquor companies (such as www.absolutvodka.com) that are currently popular and "[have] elements that appeal to youth." They added 10 other sites pertaining to alcohol-discussing drinking games or containing cocktail-making instructions, for example-and 14 sites promoting smoking. (As major U.S. cigarette brands are not advertised online, CME chose the home pages of such magazines as Cigar Aficionado [cigaraficionado.com] and Smoke Magazine [smokemag.com].) Cyber Patrol blocked only 43% of the promotional sites.

The CME also conducted Web searches on three popular search engines-Yahoo, Go/InfoSeek, and Excite-for the alcohol- and tobacco-related terms "beer," "Budweiser lizards," "cigarettes," "cigars," "drinking games," "home brewing," "Joe Camel," "liquor," and "mixed drinks." It then attempted to access the first 5 sites returned in each search. Cyber Patrol blocked 30% of the result pages, allowing, for example, cigarettes4u.com, tobaccotraders.com, and homebrewshop.com, which, according to the report, "not only promoted the use of alcohol and tobacco, but also sold products and accessories related to their consumption."

To test blocking of educational and public-health information on alcohol and tobacco, the CME added to its sample 10 sites relating to alcohol consumption-www. alcoholismhelp.com, for instance, and the Mothers Against Drunk Driving (www.madd.org) and National Organization on Fetal Alcohol Syndrome (www.nofas.org) sites-and 10 anti-smoking sites, including www.tobaccofreekids.org, www.wholetruth.com, and the site of the American Cancer Society (www3.cancer.org). Cyber Patrol did not block any of the sites in this group. Nor did it block most sites returned by the 3 search engines when terms like "alcohol," "alcoholism," "fetal alcohol syndrome," "lung cancer," or "substance abuse" were entered. Cyber Patrol allowed access to an average of 4.8 of the top 5 search results in each case; CME deemed an average of 4.1 of these contained important educational information.

Eddy Jansson and Matthew Skala, The Breaking of Cyber Patrol ® 4, Mar. 11, 2000.
Jansson and Skala decrypted Cyber Patrol's block list and found questionable blocking of Peacefire's site, for example, as well as a number of anonymizer and foreign-language Web translation services on the grounds that they fell under all of Cyber Patrol's default filtering categories. Blocked under every category but "Sex Education" was the Church of the SubGenius site, which parodies the Christian church and corporate and consumer culture, and 2 sites with information pertaining to pirated software.

Also on the block list, for "Intolerance," were a personal home page on which the word "voodoo" appeared-in a mention of voodoo-cycles.com-and the Web archives of Declan McCullagh's Justice on Campus Project (joc.mit.edu/cornell), which worked "to preserve free expression and due process at universities." Blocked in the "Satanic/Cults" category were webdevils.com (a site of multimedia Net-art projects) and Mega's Metal Asylum, a page devoted to heavy metal; the latter site was also branded "Militant/Extremist." Also blocked as "Militant/Extremist," as well as "Violence/Profanity" and "Questionable/Illegal & Gambling," were a portion of the Nuclear Control Institute site; a personal page dedicated, in part, to raising awareness of neo-Nazi activity; multiple editorials opposing nuclear arms from Washington State's Tri-City Herald; part of the City of Hiroshima site; the former Web site of the American Airpower Heritage Museum in Midland, Texas; an Illinois Mathematics and Science Academy student's personal home page, which at the time of Jansson and Skala's report consisted only of the student's résumé; and the Web site of a sheet-music publisher. The "Marston Family Home Page," a personal site, was also blocked under the "Militant/Extremist" and "Questionable/Illegal & Gambling" categories-presumably, according to the report, because one of the children wrote, "[T]his new law the Communications Decency Act totally defys [sic] all that the Constitution was. Fight the system, take the power back. . . ."

"Cyber Patrol error rate for first 1,000 .com domains," Peacefire, Oct. 23, 2000.
Peacefire tested Cyber Patrol's average rate of error, using the same 1,000 dot-com domains as a test sample that it used for an identical investigation of SurfWatch. In accordance with its Oct. 21, 2000, block list, Cyber Patrol blocked 121 Web sites for portrayals of "Partial Nudity," "Full Nudity," or "Sexual Acts." Of the 121 sites blocked, Peacefire eliminated 100 that were "under construction," and assessed the remaining 21. It considered 17 wrongly blocked, including a-actionhomeinspection.com, a-1bonded.com (a locksmith's site), a-1janitorial.com, a-1radiatorservice.com, and a-attorney-virginia.com. Peacefire deemed 4 sites appropriately blocked under Cyber Patrol's definition of sexually explicit content, for an error rate of 81%. Peacefire's Haselton wrote that Cyber Patrol's actual error rate was anywhere between 65 and 95%, but was unlikely to be "less than 60% across all domains," and as with BESS, that the results may have been skewed in Cyber Patrol's favor owing to the test's focus on dot-com domains, which "are more likely to contain commercial pornography than, say, .org domains."

"Blind Ballots: Web Sites of U.S. Political Candidates Censored by Censorware," Peacefire, Nov. 7, 2000.
In its Election Day report, Peacefire revealed that Cyber Patrol, configured to block "Partial Nudity," "Full Nudity," and "Sexual Acts," blocked the Web sites of 4 Republican candidates, 4 Democrats, and one Libertarian. The Web site of one additional Democratic candidate, Lloyd Doggett (www.doggettforcongress.com), was blocked under Cyber Patrol's "Questionable/Illegal/Gambling" category. The day after Peacefire published these findings, ZDNet News reporter Lisa Bowman contacted Cyber Patrol's current manufacturer, SurfControl. A company spokesperson directed Bowman to the CyberNOT search engine, which indicated that none of the URLs were actually prohi-bited. But later the same day, after down-loading Cyber Patrol's most recent block list, Bowman attempted to access each site, and found that the software did indeed bar her from the candidate sites in question. Hasel-ton noted that Bowman's test suggested the unreliability of the CyberNOT engine.¹⁴

"Amnesty Intercepted: Global Human Rights Groups Blocked by Web Censoring Software," Peacefire, Dec. 12, 2000.
"Amnesty Intercepted" reported the following organizations' sites blocked by Cyber Patrol on account of "sexually explicit" content: Amnesty International Israel (www.amnesty.org.il); the Canadian Labour Congress (www.clc-ctc.ca); the American Kurdish Information Network (www.kurdistan.org), which tracks human rights violations against Kurds in Iran, Iraq, Syria, and Turkey; the Mumia Solidaritäts Index (www.mumia.de), which is devoted to protesting the death sentence against Mumia Abu Jamal; the Milarepa Fund (www.milarepa.org), a Tibetan-interest group; Peace Magazine (www.peacemagazine.org); the Bonn International Center for Conversion (www.bicc.de), which promotes the transfer of human, industrial, and economic resources away from the defense sector; the Canada Asia Pacific Resource Network (www.caprn.bc.ca), whose stated "mission is to promote regional solidarity among trade unions and NGOs in the Asia Pacific" region; the Sisterhood Is Global Institute (www.sigi.org), an organization opposing violations of the human rights of women worldwide; the Metro Network for Social Justice (www.mnsj.org); the Society for Peace, Unity, and Human Rights for Sri Lanka (www.spur.asn.au); and the International Coptic Congress (www.copts.com).

"Digital Chaperones for Kids," Consumer Reports, Mar. 2001.
Consumer Reports found that Cyber Patrol failed to block 23% of the magazine's chosen 86 "easily located web sites that contain sexually explicit content or violently graphic images, or that promote drugs, tobacco, crime, or bigotry." Yet it did, Consumer Reports noted, block the home page of Operation Rescue (which the authors of the article classified as objectionable on account of its graphic images of aborted fetuses). The filter also blocked such nonobjectionable sites as Peacefire (www.peacefire.org) and Lesbian.org.

Kieren McCarthy, "Cyber Patrol bans The Register," The Register, Mar. 5, 2001; Drew Cullen, "Cyber Patrol unblocks The Register," The Register, Mar. 9, 2001.
Days after the Consumer Reports article appeared, the British newspaper The Register received word that its online edition was blocked by Cyber Patrol. The publication was notified by an employee of Citrix Systems, an application server software provider, that he had been unable to access the Register site from his office computer, on which the company had installed Cyber Patrol. SurfControl unblocked the site within days, with the exception of a page containing the Dec. 12, 2000, article that was the basis of the initial block: a piece by Register staff reporter John Leyden on Peacefire's recently introduced filter-disabling program. A SurfControl representative explained, "The Register published an article written by Peacefire containing information on how to access inappropriate sites specifically blocked by Cyber Patrol. Given [the] irresponsible nature of the article, apparently encouraging users to over-ride Cyber Patrol's filtering mechanism, we took the decision to block The Register-upholding our first obligation to customers by preventing children or pupils from being able to surf websites containing sexually explicit, racist or inflammatory material." Cullen responded that there was no "sexually explicit, racist," or "inflammatory" material in the article, which "merely describes peacefire.exe and provides a link to the Peacefire.org Web site. . . . To say The Register in any way enables the children of SurfControl customers to access a cracking utility is, quite simply, false."¹⁵

Miscellaneous reports
— In "How well does Internet filtering software protect students?" (Jan. 1998), Electronic School columnist Lars Kongshem wrote that Cyber Patrol denied users access to a page on tobacco use prevention that appeared on a site maintained by Maryland's Department of Health and Mental Hygiene.
— In his expert witness report for the defendants in Mainstream Loudoun, et al. v. Board of Trustees of the Loudoun County Library (July 14, 1998), David Burt reported that his comparative testing of Cyber Patrol, I-Gear, SurfWatch, and X-Stop had revealed that Cyber Patrol blocked 40% of sites Burt had selected as nonobscene, including the sex-information sites Di Que Si (www.webcom.com/~cps/DiQueSi/index.html), All About Sex (www.allaboutsex.org), New Male Sexuality (www.thriveonline.com/sex/malesex.intro.html), and Internet Sex Radio (www.radiosex.com/tool.html).
— In the New York Times article "Library Grapples with Internet Freedom" (Oct. 15, 1998), Katie Hafner reported that Cyber Patrol blocked Web searches for Georgia O'Keeffe and Vincent van Gogh, while hits from searches for "toys" included sites selling sex toys.
— In Feb. 2000, filteReality, a Web site maintained by former librarian Brian Smith, found that Cyber Patrol blocked Internet Trash (www.internettrash.com), a free Web-page hosting service; the site of the Adult Children of Heterosexuals (www.tiac.net/users/danam/acoh.html); and the home page of a plastic surgeon (www.psurg.com) that contained some "images showing examples of the doctor's genital enhancement work." Smith noted that "Cyber Patrol's description of its criteria explicitly states that 'non-prurient' images of nudity are not supposed to be blocked."
— Peacefire reported, in "BabelFish blocked by censorware" (Feb. 27, 2001), that Cyber Patrol blocked the foreign-language Web-page-translation service, featured on AltaVista (babelfish.altavista.com), in all 12 filtering categories, from "Violence/Profanity" to "Alcohol & Tobacco."
— In the article "Teen Health Sites Praised in Article, Blocked by Censorware" (Mar. 23, 2001), Peacefire reported that Cyber Patrol blocked ZapHealth (www.zaphealth.com), a health-education site that, according to Bonnie Rothman Morris's New York Times article published 3 days earlier, contains "[a]rticles about issues of pressing interest to a teenage audience."

Cyber Sentinel

Rather than maintaining and updating a list of sites to be blocked, or designating forbidden categories, Cyber Sentinel scans each requested + for certain keywords and phrases in its various databases, or "libraries." Its "child predator library," for instance, contains such phrases and "do you have a pic" and "can I call you." Promotional text on Cyber Sentinel's Web site (www.securitysoft.com/cyber-page.html) claims it is "the most advanced Internet filtering software package available today."

Youth Access to Alcohol and Tobacco Web Marketing: The Filtering and Rating Debate, Center for Media Education, Oct. 1999.
The CME's June 1999 study (see p. 17) found Cyber Sentinel ineffective in screening out material promoting alcohol and tobacco use. It blocked only 11% of the promotional sites selected by the CME, allowing users to access an average of 39 of the 44 pages, and blocked just 3% of the pages resulting from the researchers' searches for alcohol- and tobacco-related promotional material.

"Sites blocked by Cyber Sentinel," Peacefire, Aug. 2, 2000.
Having conducted "about an hour of ad-hoc experimentation," Peacefire found that Cyber Sentinel blocked CNN (www.cnn.com), because, as system log files revealed, the word "erotic" appeared on the front page-in the title of an article, "Naples museum exposes public to ancient erotica"); a result page for a search of the word "censorship" on Wired magazine's site, for one of the results contained the word "porn" in the title ("Feds Try Odd Anti-Porn Approach"); result pages for searches of the term "COPA" on the Wired and other news sites, also on account of article titles containing the word "porn" (for instance, "Appeals court rules against Net porn law"); and a portion of the Web site of the Ontario Center for Religious Tolerance (www.religioustolerance.org/sci_rel.htm), containing an essay on collisions between science and religion throughout history.

Cyber Sentinel also blocked sites associated with both sides of the civil liberties and Internet censorship debates: an ACLU press release, "Calls for Arrest of Openly Gay GOP Convention Speaker Reveal Danger of Sodomy Laws Nationwide" (www.aclu.org/ news/2000/n073100b.html), because of the term "anal sex"; the home page of the American Family Association (www.afa.net), because of the word "porn" ("The current administration and the Justice Department have been good to the porn industry"); on account of the word "cum," the biographies of COPA Commission members Stephen Balkam (www.copacommission.org/commission/balkam.shtml) and Donna Rice Hughes (www.copacommission.org/commission/hughes.shtml)-for both graduated magna cum laude; the COPA Commission's list of research papers (www.copacommission.org/papers), because the word "porn" appeared in the title of filtering advocate David Burt's report, "Dangerous Access, 2000 Edition: Uncovering Internet Porn in America's Libraries"; and the home page for Donna Rice Hughes's book, Kids Online: Protecting Your Children in Cyberspace (www.protectkids.com), an appendix of which is titled "Porn on the Net."

CYBERsitter

Before 1999, CYBERsitter, in addition to blocking entire sites and searches for terms on its block list, would excise terms it deemed objectionable or leave blank spaces where they would otherwise appear. This procedure led to some early notoriety for the product, such as the instance in which it deleted the word "homosexual" from the sentence, "The Catholic Church opposes homosexual marriage"-and left Web users reading "The Catholic Church opposes marriage."

In 1999, CYBERsitter modified its system and established 7 default settings, including "PICS Rating adult topics," which "[c]overs all topics not suitable for children under the age of 13," "sites promoting the gay and lesbian life style," and "[s]ites advocating illegal/ radical activities." Its total list of blocking categories grew to 22 (see appendix B). Users could, as they can with the most recent versions of the software, enable or disable any specific category.¹⁶

Brock N. Meeks and Declan McCullagh, "Jacking in from the 'Keys to the Kingdom' Port," CyberWire Dispatch, July 3, 1996.
Meeks and McCullagh reported that CYBERsitter blocked a newsgroup devoted to gay issues (alt.politics.homosexual), the Queer Resources Directory (qrd.org), and the home page of the National Organization for Women (www.now.org). CYBERsitter's list of prohibited word combinations included "[gay, queer,bisexual] [male,men,boy,group,rights, community,activities]" and "[gay,queer, homosexual,lesbian,bisexual] [society, culture]." According to the report, Brian Milburn, president of CYBERsitter's manufacturer, Solid Oak Software, responded, "We have not and will not bow to pressure from any organization that disagrees with our philosophy. We don't simply block pornography. That's not the intention of our product. The majority of our customers are strong family-oriented people with traditional family values. I wouldn't even care to debate the issues if gay and lesbian issues are suitable for teenagers.' We filter anything that has to do with sex. Sexual orientation [is about sex] by virtue of the fact that it has sex in the name."

"CYBERsitter: Where do we not want you to go today?" Peacefire, Nov. 5-Dec. 11, 1996.
Peacefire's Bennett Haselton reported that among CYBERsitter's blocked domains were, in addition to Peacefire.org itself, the "online communities" Echo Communications (www.echonyc.org) and the Whole Earth 'Lectronic Link (www.well.com); the Web site of Community ConneXion (www.c2.org), which manufactured an anonymous-surfing program; the ISP CRIS.com; and the home page of the National Organization for Women (www.now.org). CYBERsitter also barred any user from conducting a Yahoo search for the term "gay rights."

"Cybersitter Blocks The Ethical Spectacle," Ethical Spectacle, press release, Jan. 19, 1997.
In early 1997, CYBERsitter blocked the Ethical Spectacle, an online magazine "examining the intersection of ethics, law and politics in our society," after editor Jonathan Wallace added a link to the site titled "Don't Buy Cybersitter," which directed users to Peacefire's report "CYBERsitter: Where do we not want you to go today?" Wallace wrote to Milburn and Solid Oak technical support "demanding an explanation. I pointed out that The Spectacle does not fit any of their published criteria for blocking a site. I received mail in return demanding that I cease writing to them and calling my mail 'harassment'-with a copy to the postmaster at my ISP."

Karen G. Schneider, A Practical Guide to Internet Filters, 1997.
Schneider's Internet Filter Assessment Project reported that unlike other filtering products, CYBERsitter does not permit its keyword-blocking feature to be disabled. Regarding CYBERsitter's claim that it "looks at how the word or phrase is used in context," Schneider quoted one TIFAP tester: "[N]othing could be further from the truth." Tthe filter deleted the word "queer," for example, from Robert Frost's "Stopping by the Woods on a Snowy Evening" ("My little horse must think it queer / To stop without a farmhouse near"). CYBERsitter did not block sites containing instructions for the growing of marijuana but did block a news item on the legislation surrounding it.

Marie José Klaver, What does Cybersitter block? June 23, 1998.
In June 1998, Marie-José Klaver decrypted and published CYBERsitter's full list of blocked words, strings, sites, and domains. Among the domains on the block list were servers of the University of Chicago (student-www.uchicago.edu), the University of Virginia's Information Technology and Communication division (watt.seas.uvirginia.edu), Georgia State University (panther.gsu.edu), the University of Michigan's engineering department (www-personal.engin.umich.edu), and Rutgers University (quartz.rutgers.edu); several large Dutch domains, including euronet.nl, huizen.dds.nl, and worldaccess.nl; and the phrases "bennetthaselton," "peacefire," and "dontbuycybersitter."

Christopher Hunter, "Filtering the Future," July 1999.
Though Christopher Hunter's study (see p. 16) concluded that CYBERsitter was the most reliable filter in his test in terms of screening out "objectionable" sites (it blocked 25, or 69.4%, of such sites in his sample), he also noted that the software performed well below the 90-95% rate of accuracy boasted by the manufacturer. CYBERsitter fared worst in its treatment of "nonobjectionable" material, blocking 24, or 14.6%, of the sites to which Hunter assigned RSACi ratings no higher than one. Among these were Sharktagger (www.sharktagger.com), a site promoting responsible shark fishing and conservation through the practice of tagging; a listing of local events posted on Yahoo (localevents.yahoo.com); RiotGrrl (www.riotgrrl.com); Planned Parenthood (www.plannedparenthood.org); Stop Prisoner Rape (www.igc.apc.org/spr); the National Organization for Women (www.now.org); the feminist performance-art and activist troupe Guerrilla Girls (www.guerrillagirls.com); the Church of Scientology (www.scientology.org); The Body (www.thebody.com/index.shtml), an informational site on AIDS and HIV; Williams College's information page on safe sex (wso.williams.edu/orgs/peerh/sex/safesex); the Coalition for Positive Sexuality (CPS) (www.positive.org/Home/index.html), SIECUS (www.siecus.org), and Pro-Life America (www.prolife.org).

CYBERsitter proved particularly likely to deny access to nonpornographic sites relating to homosexuality, blocking the QWorld contents page (www.qworld.org/TOC.html); the gay Internet communities Planet Out (www.planetout.com), PrideNet (www. pridenet.com), and the Queer Zone (www.queerzone.com); A Different Light Bookstore (www.adlbooks.com), which specializes in gay and lesbian literature; Gay Wired Presents Wildcat Press (www.gaywired.com/wildcat/index); and Queer Living's "Promoting with Pride" page (www.qmondo.com/queerliving). (These sites, while not falling under RSAC's definition of unacceptability, do fall within CYBERsitter's default filtering category of "sites promoting the gay and lesbian life style.")

Youth Access to Alcohol and Tobacco Web Marketing, Center for Media Education, Oct. 1999.
CME charged CYBERsitter with under- and overinclusive filtering of alcohol- and tobacco-related material, as it blocked only 19% of the promotional sites in CME's sample-leaving unblocked beer sites such as heineken.com, and sites on which tobacco products were sold, such as lylessmokeshop .com. While performing better than most other filters in its response to searches for promotional content-CYBERsitter prohibited searches for "beer," "cigarettes," "cigars," and "liquor"-it subsequently blocked just 3% of the result pages (from the allowed searches) that the CME testers attempted to view. CYBERsitter also blocked 13% of the CME's chosen educational and public-health sites, including alcoholism.miningco.com, al-anon.alateen.org, and health.org, and prohibited testers from conducting searches for "alcohol," "alcoholism," "fetal alcohol syndrome," "tobacco," and "tobacco settlement."

"CYBERsitter Examined," Peacefire, 2000.
Peacefire reported that CYBERsitter blocked a number of nonprofit sites, including Peacefire.org itself; the Penal Lexicon (www.penlex.org.uk), an British online project documenting prison conditions worldwide, and the home page of the Human Awareness Institute (www.hai.org.uk), an organization that "[a]ims to create a world where people live in dignity, respect, understanding, truth, kindness, honesty, compassion and love." CYBERsitter also denied access to various educational servers, including those of the Department of Astronomy at Smith College (earth.ast.smith.edu), the Computer Animation Laboratory at the California Institute of the Arts (itchy.calarts.edu), the University of Oregon (darkwing.uoregon.edu), and the College of Humanities and Social Sciences at Carnegie Mellon University (hss.cmu.edu).

CYBERsitter blocked the Village Voice Web site (www.villagevoice.com); Calweb, an Internet service provider (www.calweb.com); Internex Online (www.io.org), another ISP, which hosts the Web site of the National Organization of People Attacking Sales of Tobacco to Youth (www.io.org/~pete/ccSmok3.html); and Pathfinder (pathfinder.com), which at one time published a search engine on which users could check URLs against CYBERsitter's block list.

"Amnesty Intercepted," Peacefire, Dec. 12, 2000.
CYBERsitter blocked a number of pages on the Amnesty International site because of its keyword-filtering mechanism. A news item, for instance (www.amnesty.org/news/1998/32107198.htm), containing the sentence, "Reports of shootings in Irian Jaya bring to at least 21 the number of people in Indonesia and East Timor killed or wounded," was prohibited for its "sexually explicit" content. Peacefire's review of the system log revealed that CYBERsitter had blocked the site after detecting the words "least 21." The filter blocked another human rights page, which noted that the United Nations Convention on the Rights of the Child "defines all individuals below the age of 18 years as children," for the words "age of 18."

"Digital Chaperones for Kids," Consumer Reports, Mar. 2001.
While failing to block 22% of sites that Consumer Reports deemed objectionable because of "sexually explicit content or violently graphic images" or promotion of "drugs, tobacco, crime, or bigotry," CYBERsitter blocked "nearly one in five" of the sites the authors considered inoffensive, including Lesbian.org, the Citizens' Committee for the Right to Keep and Bear Arms, and the Southern Poverty Law Center.

Miscellaneous reports
— In a review of "Filtering Utilities" (Apr. 8, 1997), PC Magazine noted that CYBERsitter blocked an engineering site with "BourbonStreet" in its URL.
— According to the Digital Freedom Network's article on "Winners of the Foil the Filter Contest" (Sept. 28, 2000), CYBERsitter blocked House Majority Leader Richard "Dick" Armey's official Web site (armey.house.gov) upon detecting the word "dick," and Focus on the Family's Pure Intimacy page (www.pureintimacy.org), which protests pornogaphy and is geared toward individuals "struggling with sexual temptations."
— In "