Cybertelecom Federal Internet Law & Policy An Educational Project

Filters

Filters have found themselves in a wide variety of policy scenarios.

• The Children's Internet Protection Act altered the Erate program, conditioning the funds to schools and libraries on the development of an Internet safety plan and the implementation of filters.
• COPA and the Tax Moritorium Act require ISPs to provide notification to subscribers of the availability of filtering software.
• Copyright owners have advocated filtering the Internet for pirated music and videos.
• ISPs have agreed to work with state attorney generals in order to block child pornography.
• Countries have policies to filter the Internet the pornography and filter political dissent.

June 2008: The FCC has initiated an proceeding to consider licensing the Advanced Wireless Service for broadband Internet access with a requirement of Internet filters.

Filters function by filtering of blocking based on some criteria. That criteria can be

• The content itself
• Tags or labels of the content
• Source or origin of the content (such as ip number, domain name)

The filters can be built and maintained by private services, who then market the filtering service as blocking pursuant to the criteria of different communities. While some of these service's databases can be built as a result of human analysis and inspection, humans are not capable of reviewing all of the content that is on or will soon be on the Internet. As a result, filtering services rely upon automated criteria in order to develop the lists of what should and should not be blocked. The automated process of identifying inappropriate content results in false positives (blocking of otherwise appropriate content) and false negatives (failure to block inappropriate content). [NAS 10] [COPA II.B.3].

In addition, the effectiveness of filters are limited by the ability of those filtered to get around or defeat the filters.

Filtering services are private companies that select what material to filter pursuant to their own criteria. Those companies generally consider that criteria and the database of filtered content proprietary. They have been criticized for not being forthcoming about what they are blocking or their selection process, resulting in censorship without accountability or due process. [COPA II.B.3]

It has been found that groups that are critical of Internet filters have been blocked by Internet filters:

• Amnesty International
• American Family Association
• Banned Books On-line
• The Religious Society of Friends
• The Safer Sex Page
• The Web site of Rep. Dick Armey
• Dozens of Web sites of candidates in the last election
• Child Online Protection Act commissioners’ pages that had the word “cum” (as in “magna-cum-laude”) in the bio
• Information on China’s response to the spread of AIDS
• National Association of Women
• People of the Ethical Treatment of Animals (PETA)
• The American Association of University Women Maryland.

An early useful report concerning Internet filters is the “Electronic Privacy Information Center’s Faulty Filters: How Content Filters Block Access to Kid-Friendly Information on the Internet” (December 1997) (www.epic.org).

Peacefire recently conducted a review of the sites one Internet filter, SurfWatch, blocked of the first 1,000 working .coms. According to Peacefire, out of 1,000 sites tested, 147 were blocked. Of those 147 blocked sites, 96 sites were under construction and did not have content, 42 were non-pornographic and did not have sexually explicit content and nine were pornographic.  According to Peacefire, this constitutes an error rate of 82 percent. A search of the Surfwatch (now SurfControl) Web site revealed press releases acknowledging - but not refuting - the Peacefire study. See also official government reports reviewing the effectiveness of filters.

Papers

• Bambauer, Derek, "Guiding the Censor's Scissors: Assessing Internet Filtering" (June 10, 2008). Available at SSRN: http://ssrn.com/abstract=1143582
• Cheryl B. Preston, Making Family-friendly Internet a Reality: The Internet Community Ports Act, 2007 B.Y.U. L. REV. 1471 (2007).
• Mary Minow, Lawfully surfing the Net: Disabling public library Internet filters to avoid more lawsuits in the United States, First Monday 4/30/2004
• Documentation on Internet Filtering Worldwide, Harvard Law
• REPORT: See No Evil: How Internet Filters Affect the Search for Online Health Information, KFF 12/13/02
  • Internet Filters' Affect on Online Health Information, Kaiser 12/13/02
  • Study: Web Filters Block Health Information, NYT 12/13/02
  • Web filters blocking health sites, CNET 12/13/02
• Internet Filtering Software Wrongly Blocks Many Sites, EFF 9/18/02
• EFF: School Communities Give Internet Filtering Law Failing Grade, EFF 9/18/02
• Children's Online Protection Act Commission Report 2000
  • Section II.B. Filtering/Blocking
• Peacefire, Amnesty Intercepted: Global human rights groups blocked by Web censoring software December 12, 2000
  • Blind Ballots: Web Sites of U.S. Political Candidates Censored by Censorware  November 7, 2000
  • Sites blocked by Cyber Sentinel 8/2/2000
  • SurfWatch error rate for first 1,000 .com domains August 2000
  • CYBERsitter: Where Do We Not Want You To Go Today?
• EPIC:  Faulty Filters: How Content Filters Block Access to Kid-Friendly Information on the Internet  December 1997
• ACLU:  Censorship In a Box: Why Blocking Software is  Wrong for Public Libraries 1998
  • Fahrenheit 451.2: Is Cyberspace Burning? How Rating and Blocking Proposals  May Torch Free Speech on the Internet
  • Alternatives to Blocking Software

• The Censorware Project:  Blacklisted by Cyber Patrol:  From Ada to Yoyo - A report from The Censorware Project
  • Censored Internet Access in Utah Schools and Libraries (PDF format) (finding that the following had been blocked by filtering software:  The Bible, the US Constitution, the Declaration of Independence, anti-drug info, Shakespeare, Sherlock Holmes, and the Koran)
• Internet Free Expression Alliance:
• CPSR's Filtering FAQ
  • CPSR Press Release Computer Professionals Question Internet Filtering Agreement July 18, 1997
• Global Internet Campaign's website on rating systems and filters
  • GILC Member Statement on "Impact of Self-Regulation and Filtering on Human Rights to Freedom of Expression" March 1998
• Cyber-Rights & Cyber-Liberties (UK) Report,  ‘Who Watches the Watchmen: Internet Content Rating Systems, and Privatized Censorship’ Nov 1997
• Electronic Frontier Foundation
• EFF: Internet Blocking & Censorware
  • Public Interest Principles for Online Filtration, Ratings and Labeling Systems, Feb 28, 1997.
  • "Blacklisting Bytes", by Seth Finkelstein and Lee Tien
• People For the American Way Foundation Written Testimony of People For the American Way Foundation to the U.S. National Commission on Libraries and Information Science on "Kids and the Internet: The Promise and the Perils"  December 10, 1998
• Seth Finkelstein's Anticensorware Investigations
• TODAY'S INTERNET BLOCKING AND FILTERING TECHNOLOGY CAPABLE OF MEETING NEEDS OF MOST EDUCATIONAL INSTITUTIONS, NTIA REPORT SAYS, NTIA 8/20/0

Statistics

Beginning in 2001, the FRSS surveys on Internet access asked whether public schools used any technologies or procedures to prevent student access to inappropriate material on the Internet, the types of technologies or procedures used, and whether such technologies were used on all computers with Internet access used by students. The 2002 and 2003 surveys also asked about the methods used to disseminate information about the technologies or procedures to students and parents.

• In 2003, almost all public schools with Internet access (97 percent) used various technologies or procedures to control student access to inappropriate material on the Internet (table 17). Across all school characteristics, between 96 and 100 percent 20 of schools reported using these technologies or procedures. In addition, 99 percent of these schools used at least one of these technologies or procedures on all Internet-connected computers used by students.
• Among schools using technologies or procedures to prevent student access to inappropriate material on the Internet in 2003, 96 percent used blocking or filtering software (table 18). Ninety-three percent of schools reported that teachers or other staff members monitored student Internet access, 83 percent had a written contract that parents have to sign, 76 percent had a contract that students have to sign, 57 percent used monitoring software, 45 percent had honor codes, and 39 percent allowed access only to their intranet. 21 Most of the schools (97 percent) used more than one procedure or technology as part of their Internet use policy (not shown in tables).
• Ninety-five percent of public schools using technologies or procedures to prevent student access to inappropriate material on the Internet indicated that they disseminated the information about these technologies or other procedures via their school policies or rules distributed to students and parents (table 19). Sixty-six percent did so with a special notice to parents, 58 percent used their newsletters to disseminate this information, 31 percent posted a message on the school website or web page, 25 percent had a notice on a bulletin board at the school, 17 percent had a pop-up message at computer or Internet log on, and 5 percent used a method other than the ones listed above.

Source: Internet Access to US Public Schools and Classrooms, National Center for Education Statistics, p. 12

Technology Measures and Filtering Programs

• Bess The Internet Retriever
• Cyberpatrol
• Cybersitter
• Family Click
• Internet Content Rating Association
• NetNanny
• Safesurf
• Websense

News & Blogs

• That Didn't Take Long At All: $89 Million Australian Internet Filters Called A Failure, Techdirt 2/21/2008
• OV council to decide on filters at library that would limit Internet access to porn, Arizona Daily Star 11/17/2006
• ACLU sues rural libraries over Internet filtering policies, Seattle Post 11/17/2006
• 'Access denied': Popular Web sites running afoul of content filters, IHT 3/9/2006
• Hot market for censorship tools, IHT 10/18/2005
• Censorware software fails to cut it, Register 8/30/2005
• Protecting Teens Online, Pew 3/18/2005
• More Parents Use Filters to Control Teen Web Use, clickz 3/18/2005
• Sharp split on library filters, Mlive 3/1/2004
• Zittrain and The OpenNet Initiative Take on Internet Filtering, Berkman 3/1/2004
• Technology can block porn but costs high, fix temporary, USA Today 1/8/2004
• Empirical Analysis of Google SafeSearch, Harvard 4/17/03
• Google filter blocks innocuous sites, MSNBC 4/11/03
• EFA Comments on Mandatory and Blocking by ISPs, EFA 3/24/03
• Curious teenagers need to be informed about sex, not controlled, Sydney 3/5/03
• Sex, the Constitution and the Net, CNET 3/5/03
• Aussies chew over enforced Net filters, Register 3/5/03
• Net porn controls useless: study, CNN 3/5/03
• Library Porn Filter Law Hits, Internet News 3/5/03
• What Would Dewey Do? Libraries Grapple With Internet, NYT 12/2/02
• Most Parents Don't Use Web Filters, Electronicnews 9/26/02
• Report cites possible religious bias in school web filters, eschool news 3/4/02
• USA - Net filter use jumps in libraries, CNET 1/11/02
• Filtering company SurfControl gets patent, CNET 5/11/01
• Army deploys porn-blocking software, USAToday 5/1/01
• Court won't force library to filter Net content, CNET 3/12/01
• Internet Content Filtering Comes Of Age, Techweb 3/9/01
• Web filters far from perfect, Consumer Reports says, Nando 2/15/01
• Net filters strain to block sites, CNET 2/15/01
• Consumer Reports on Internet Filters, CR 2/15/01
• Report: Net filters don't work, USAToday 2/15/01
• Filtering Software Is Not Up to Snuff, IN 2/15/01
• Civil Liberties Groups Urge FCC To Stay Out Of Net Filtering, Washtech 2/16/01
• Pr. William Libraries To Keep Web Filters, Wash Post 6/10/02