Opinion
Civil Action No. 98-5591.
March 22, 2007.
Addison F. Golladay, Benjamin Sahl, Christopher R. Harris, Elan R. Dobbs, Jeroen Van Kwawegen, Joseph B. Widman, Katherine E. Marshall, Paul A. Serritella, Seth L. Friedman, Stefanie E. Laughlin, Latham Watkins, Ann Elizabeth Beeson, Aden J. Fine, Benjamin Elihu Wizner, Catherine Newby Crump, Christopher A. Hansen, American Civil Liberties Union, New York City, David L. Sobel, Washington, DC, Jonathan H. Feinberg, Kairys Rudovsky Messing Feinberg, Mary Catherine Roper, Stefan Presser, ACLU of PA, Philadelphia, PA, for American Civil Liberties Union, American Booksellers Foundation for Free Expression, Addazi, Inc., Electronic Frontier Foundation, Electronic Privacy Information Center, Free Speech Media, Philadelphia Gay News, Powell's Bookstore, Salon Internet, Inc., Heather Corinna Rearick, Nerve.com, Inc., Aaron Peckham, Public Communicators, Inc., Sexual Health Network.
Eric Joseph Beane, Isaac R. Campbell, James D. Todd, Jr., Joel L. McElvain, Kenneth Elliot Sealls, Tamara Ulrich, U.S. Department of Justice, Washington, DC, Richard M. Bernstein, U.S. Attorney's Office, Philadelphia, PA, for Alberto R. Gonzales.
FINAL ADJUDICATION
At issue in this case is the constitutionality of the Child Online Protection Act, 47 U.S.C. § 231 ("COPA") and whether this court should issue a permanent injunction against its enforcement due to its alleged constitutional infirmities. COPA provides both criminal and civil penalties for transmitting sexually explicit materials and communications over the World Wide Web ("Web") which are available to minors and harmful to them. 47 U.S.C. § 231(a). After a trial on the merits, for the reasons that follow, notwithstanding the compelling interest of Congress in protecting children from sexually explicit material on the Web, I conclude today that COPA facially violates the First and Fifth Amendment rights of the plaintiffs because: (1) at least some of the plaintiffs have standing; (2) COPA is not narrowly tailored to Congress' compelling interest; (3) defendant has failed to meet his burden of showing that COPA is the least restrictive, most effective alternative in achieving the compelling interest; and (3) COPA is impermissibly vague and overbroad. As a result, I will issue a permanent injunction against the enforcement of COPA.
TABLE OF CONTENTS
I. PROCEDURAL HISTORY ....................................... 779II. THE RELEVANT LANGUAGE OF COPA AND THE CONSTITUTION ....... 779
III. FINDINGS OF FACT ......................................... 781 A. The Internet .......................................... 781 B. The Parties ........................................... 782 C. The Experts ........................................... 784 1. Plaintiffs' Experts ................................ 784 2. Defendant's Experts ................................ 785 D. Information Regarding Plaintiffs' Web Sites and the Content Thereon and Select Plaintiffs' Fear of Prosecution under COPA ................................ 786 E. Sexually Explicit Materials Available on the Web ...... 788 1. In General ......................................... 788 2. The Amount of Foreign Sexually Explicit Material on the Web ......................................... 789 F. Internet Content Filtering Technology and its Effectiveness ......................................... 789 1. In General ......................................... 789 2. The Availability and Cost of Filters ............... 793 3. Filter Ease of Use and User Satisfaction ........... 793 4. The Effectiveness of Filters ....................... 794 a. In General ...................................... 794 b. Study Results ................................... 795 G. Select Legislative History of COPA and the Limitations of COPA ............................................... 797 H. Statistical Information on Obscenity Prosecutions ..... 799 I. The Affirmative Defenses in COPA and Their Availability and Effectiveness ........................ 799 1. The General Availability of Age Verification Technologies ....................................... 800 2. The Effectiveness of Payment Cards as a Defense and Minors Access Thereto .............................. 800 3. The Effectiveness of Data Verification Services .... 802 4. The Effectiveness of Digital Certificates and Other Reasonable Measures that Are Feasible under Available Technology ......................................... 803 5. The Economic Burdens and Loss of Web Viewership Associated with the Affirmative Defenses ........... 803 6. Web Users' Privacy Concerns and Reluctance to Provide Personal Information ....................... 805 a. Web Users' Privacy Concerns ..................... 805 b. Web Users' Security Concerns .................... 806 J. Geolocation ........................................... 807
IV. CONCLUSIONS OF LAW ....................................... 807 A. Standing .............................................. 807 B. Strict Scrutiny Applies to this Action ................ 809 C. Defendant Has Failed to Meet His Burden of Proof ...... 810 1. Defendant Has Failed to Show that COPA Is Narrowly Tailored to Congress' Compelling Interest ........................................... 810 a. COPA Is Overinclusive ........................... 810 b. COPA Is Underinclusive .......................... 810 c. The Affirmative Defenses in COPA Do Not Aid in Narrowly Tailoring It to Congress' Compelling Interest ........................................ 811 2. Defendant Has Failed to Show that COPA Is the Least Restrictive Alternative for Advancing Congress' Compelling Interest ................................ 813 3. Defendant Has Failed to Show that Other Alternatives Are Not at Least as Effective as COPA .............. 814 D. Vagueness and Overbreadth ............................. 816 1. COPA Is Vague ...................................... 816 2. COPA Is Overbroad .................................. 819
V. CONCLUSIONS .............................................. 820
I. PROCEDURAL HISTORY
The plaintiffs in this action, which include both the individual and institutional plaintiffs listed below, have challenged the constitutionality of COPA under the First and Fifth Amendments. COPA, which was designed to protect minors from exposure to sexually explicit materials on the Web deemed harmful to them, was signed into law on October 21, 1998. COPA is the second attempt by Congress to protect children from such material. The first attempt was the Communications Decency Act of 1996, 47 U.S.C. § 223 ("the CDA") which the Supreme Court held was unconstitutional because it was not narrowly tailored to serve a compelling governmental interest and because less restrictive alternatives were available. See Ashcroft v. ACLU, 542 U.S. 656, 661 (2004) (discussing Reno v. ALCU, 521 U.S. 844 (1997)). COPA was designed to directly address the faults that the Supreme Court found with the CDA. The day after COPA was signed, the plaintiffs filed this suit seeking injunctive relief from its enforcement. On February 1, 1999, after having previously granted the plaintiffs' motion for a temporary restraining order, this court granted the plaintiffs' motion for a preliminary injunction. ACLU v. Reno, 31 F. Supp.2d 473 (E.D. Pa. 1999). After an interim trip to the Supreme Court (see Ashcroft v. ACLU, 535 U.S. 564 (2002)), this court's decision granting the preliminary injunction was finally affirmed by the Supreme Court on June 29, 2004, and remanded to this court for a trial on the merits in order to, inter alia, update the factual record to reflect current technological developments, account for any changes in the legal landscape, and to determine whether Internet content filters are more effective than COPA or whether other possible alternatives are less restrictive and more effective than COPA. Ashcroft, 542 U.S. at 671-673. For a more detailed description of the history and background of this case, see the Supreme Court's opinion. Id. at 663-664. This court held a trial on the merits of the within action, beginning on October 23, 2006 and concluding on November 20, 2006.
The Court of Appeals for the Third Circuit twice reviewed the efficacy of the preliminary injunction, once on direct appeal(ACLU v. Reno, 217 F.3d 162 (3d Cir. 2000)) and once upon remand from the Supreme Court (ACLU v. Ashcroft, 322 F.3d 240 (3d Cir. 2003)), each time affirming the decision of this court.
II. THE RELEVANT LANGUAGE OF COPA AND THE CONSTITUTION
COPA provides that:
Whoever knowingly and with knowledge of the character of the material, in interstate or foreign commerce by means of the World Wide Web, makes any communication for commercial purposes that is available to any minor and that includes any material that is harmful to minors shall be fined not more than $50,000, imprisoned not more than 6 months, or both.47 U.S.C. § 231(a)(1). There is an additional monetary penalty for intentional violations of the above quoted language and a provision for additional civil penalties. 47 U.S.C. § 231(a)(2) (3).
The crux of the statute is found in the definition of "harmful to minors" which tracks the familiar Miller obscenity standard.See Miller v. California, 413 U.S. 15, 24 (1973). Specifically, "material that is harmful to minors", means:
any communication, picture, image, graphic image file, article, recording, writing, or other matter of any kind that is obscene or that —
(A) the average person, applying contemporary community standards, would find, taking the material as a whole and with respect to minors, is designed to appeal to, or is designed to pander to, the prurient interest;
(B) depicts, describes, or represents, in a manner patently offensive with respect to minors, an actual or simulated sexual act or sexual contact, an actual or simulated normal or perverted sexual act, or a lewd exhibition of the genitals or post-pubescent female breast; and
(C) taken as a whole, lacks serious literary, artistic, political, or scientific value for minors.47 U.S.C. § 231(e)(6). A minor is defined as "any person under 17 years of age." 47 U.S.C. § 231(e)(7).
"[B]y means of the World Wide Web" is defined as the "placement of material in a computer server-based file archive so that it is publicly accessible, over the Internet, using hypertext transfer protocol [("HTTP")] or any successor protocol." 47 U.S.C. § 231(e)(1). Under COPA, the Internet "means the combination of computer facilities and electromagnetic transmission media, and related equipment and software, comprising the interconnected worldwide network of computer networks that employ the Transmission Control Protocol/Internet Protocol or any successor protocol to transmit information." 47 U.S.C. § 231(e)(3).
Another important feature of COPA for the purposes of this action is that "[a] person shall be considered to make a communication for commercial purposes only if such person is engaged in the business of making such communication." 47 U.S.C. § 231(e)(2)(a). Moreover, "engaged in the business" means that:
the person who makes a communication, or offers to make a communication, by means of the World Wide Web, that includes any material that is harmful to minors, devotes time, attention, or labor to such activities, as a regular course of such person's trade or business, with the objective of earning a profit as a result of such activities (although it is not necessary that the person make a profit or that the making or offering to make such communications be the person's sole or principal business or source of income). A person may be considered to be engaged in the business of making, by means of the World Wide Web, communications for commercial purposes that include material that is harmful to minors, only if the person knowingly causes the material that is harmful to minors to be posted on the World Wide Web or knowingly solicits such material to be posted on the World Wide Web.47 U.S.C. § 231(e)(2)(b).
Although COPA brands all speech falling within its reach as criminal speech, it also provides an affirmative defense against liability if:
the defendant, in good faith, has restricted access by minors to material that is harmful to minors —
(A) by requiring use of a credit card, debit account, adult access code, or adult personal identification number;
(B) by accepting a digital certificate that verifies age; or
(C) by any other reasonable measures that are feasible under available technology.47 U.S.C. § 231(c)(1).
Moreover, those exempt from liability include telecommunications carriers, Internet access service providers, those engaged in the business of providing an Internet information location tool, or those:
similarly engaged in the transmission, storage, retrieval, hosting, formatting, or translation (or any combination thereof) of a communication made by another person, without selection or alteration of the content of the communication, except that such person's deletion of a particular communication or material made by another person in a manner consistent with subsection (c) of this section or section 230 of this title shall not constitute such selection or alteration of the content of the communication.47 U.S.C. § 231(b).
The First Amendment to the Constitution of the United States provides that "Congress shall make no law . . . abridging the freedom of speech, or of the press." U.S. Const. Amend. I.
The plaintiffs also rely upon the Fifth Amendment to the Constitution which is the due process vehicle by which this action arrives in federal court.
III. FINDINGS OF FACT
To the extent that the following Findings of Fact include Conclusions of Law or mixed Findings of Fact and Conclusions of Law, those Findings and Conclusions are hereby adopted by this court.
Having presided at the trial, having seen and heard the testimony of the parties' representatives and other witnesses, and having reviewed the other evidence received, I find that, unless otherwise noted, the facts set forth in the parties' Joint Exhibit 1, and the testimony of the witnesses as well as the evidence excerpted and referenced in these Findings of Fact are true, reliable, and credible and I accept those facts and that testimony as the foundation of the following Findings of Fact and Conclusions of Law.
A. The Internet
1. The Internet is an interactive medium based on a decentralized network of computers. One portion of the Internet is known as the World Wide Web ("Web"). The Internet may also be used to engage in other activities such as sending and receiving emails, trading files, exchanging instant messages, chatting online, streaming audio and video, and making voice calls. Joint Exhibit ("J. Ex.") 1 ¶¶ 78-79, 94.
2. The results of the U.S. Census Bureau's Current Population Survey show that in September 2001, approximately 54 percent of the U.S. population was using the Internet from any location. That figure rose to 59 percent in 2003. Id. ¶ 97.
3. On the Web, a client program called a Web browser retrieves information from the Internet, such as Web pages and other computer files using their network addresses and displays them, typically on a computer monitor, using a markup language that determines the details of the display. One can then follow hyperlinks in each Web page to other resources on the Web of information whose location is provided by these hyperlinks. The act of following hyperlinks is frequently called "browsing" or "surfing" the Web. Id. ¶ 79.
4. Web pages, which can contain, inter alia, text, still and moving picture files, sound files, and computer scripts, are often arranged in collections of related material called Web sites, which consist of one or more Web pages. Id. ¶ 80.
5. Modern search engines search for and index Web pages individually. Search engines are Web sites that provide links to relevant Web pages, in response to search terms (words or phrases) entered by a user. They are a popular way of finding information online. Id. ¶ 83.
6. It is estimated that there are between 25 and 64 billion Web pages on the surface portion of the Web ("Surface Web") — that is, the portion of the Web that is capable of being indexed by search engines. Mewett Testimony, 11/7 Tr. 100:23-101:1; Def. Ex. 82, at 13. These Web pages may be displayed on a monitor screen and, thus, the content may be seen by anyone operating a computer or other Internet capable device which is properly connected to the Internet. The court takes judicial notice of the fact that the computers relevant to this case are used throughout the modern world in, inter alia, homes, schools, hotels, businesses, public Internet cafes, and libraries and that portable computers and other Internet capable devices can be operated almost anywhere and have wide access to the Internet.
7. HTTP stands for hypertext transfer protocol which is widely used on the Internet. In fact, most Web site addresses ("URLs") use HTTP. J. Ex. 1 ¶¶ 111, 113.
8. FTP stands for file transfer protocol. It is used primarily to transfer files across the Internet. Id. ¶ 110.
B. The Parties
9. Defendant Alberto R. Gonzales is the Attorney General of the United States and is charged with enforcing the provisions of COPA challenged in this action. Id. ¶ 1. Attorney General Gonzales is sued here in his official capacity. Doc. No. 175, at 15.
10. The plaintiffs represent a range of individuals and entities including speakers, content providers, and ordinary users on the Web, as that term is defined in COPA. The plaintiffs post content on their Web sites including, inter alia, resources on sexual health, safer sex, and sexual education; visual art and poetry; resources for gays and lesbians; online magazines and articles; music; and books and information about books that are being offered for sale. J. Ex. 1 ¶ 2.
11. Some of the plaintiffs provide interactive fora on their Web sites, such as online discussion groups, bulletin boards and chat rooms, which enable users to create their own material on the plaintiffs' Web sites. Some of the verbal and visual exchanges that could potentially occur in these chatrooms or in the postings on their bulletin boards may include language or images that contain sexually explicit content. Id. ¶ 3.
12. Plaintiff American Civil Liberties Union ("ACLU") is a nationwide, non-partisan organization which states that it is dedicated to defending the principles of the Bill of Rights. ACLU members Patricia Nell Warren ("Warren") and Lawrence Ferlinghetti ("Ferlinghetti") engage in speech on the Internet. Id. ¶ 4.
13. Plaintiff ACLU sues in part on behalf of its member Ferlinghetti, who is a writer and San Francisco's poet laureate. Ferlinghetti is the co-founder of City Lights Bookstore, which maintains a website "that promotes books available from the bookstore" and "contains lists of literary events and a brief history of City Lights Bookstore and Publishing," has a section describing Ferlinghetti's 1956 obscenity trial for selling the Allen Ginsberg poem Howl, and also has Ferlinghetti's poetry. Id. ¶ 5.
14. Plaintiff ACLU also sues in part on behalf of Warren, who is an author of novels, poetry, numerous articles, and essays. Her novels are alleged to be the most popular novels among classic gay literature. Warren is a co-owner of Wildcat International and its publishing arm, Wildcat Press. The Web site for Wildcat Press contains excerpts of her work, including "sexually explicit details such as the description of a `foursome' [of people] erotically dancing and a description of two men passionately kissing."Id. ¶ 7.
15. Plaintiff Condomania is the nation's first condom store and a leading seller of condoms and distributor of safer sex materials. Condomania engages in speech on the Internet. Id. ¶ 8. Adam Glickman ("Glickman") is the CEO of Condomania. Glickman Testimony, 10/30 Tr. 91:18-20.
16. Plaintiff Heather Corinna ("Corinna") is a writer, artist, sex-educator, and activist whose primary presence on the Web consists of Scarletletters.com, Scarleteen.com, and Femmerotic.com, "each of which deals with issues of sex and sexuality with an explicit focus on challenging and combating the sexual oppression of traditionally marginalized groups." J. Ex. 1 ¶¶ 9-10.
17. Corinna operates the website Scarleteen.com. "Scarleteen is the Internet's largest independent, unaffiliated, free resource for young adult sex education, information, and discussion, serving nearly two million teens, young adults, parents, and educators each year." The Scarleteen Web site states that "[w]e offer Scarleteen as a far better resource for sex information for teens than adult sexuality sites, as well as a supplement to in-home and schoolbased sex education. Many parents we have heard from have used it as a tool to initiate discussion with their teens on some of the topics addressed. Homeschooling parents have used Scarleteen as curricula for sex education; colleges add our articles to their syllabi often." Id. ¶ 11.
18. "Femmerotic is Heather Corinna's personal Web site for showcasing her photographic and textual work and providing an `open and intimate look at her life as an artist and activist.'" On this Web site, Corinna states that "[g]enerally, I intend to examine sexuality, to document sexual relationship[s], to explore the human body and how I and viewers perceive it, to examine the female body and feelings about it, to explore my own identity and use all those aims to create work that creates questions." Id. ¶ 12.
19. Plaintiff Electronic Frontier Foundation ("EFF") sues in part on behalf of John W. "Bill" Boushka, who has work on the Web site www.doaskdotell.com. In the Amended Complaint, Mr. Boushka states that he fears prosecution for his book "Do Ask, Do Tell: A Gay Conservative Lashes Back," which he describes as "an exposé about gays in the military" that is a "politically-charged text" containing "subject-matter and language that might be deemed harmful to minors." Id. ¶ 13.
20. Plaintiff Free Speech Media, LLC in partnership with Public Communicators Inc., operates freespeech.org, which provides speech on the Internet and is "designed to encourage the democratic expression of progressive ideals through promoting, curating and hosting independent creators of audio and video content on the Web." Its video and audio files "cover a wide range of topics, including human rights, homelessness, labor issues, racism, prison conditions, sexuality, AIDS, feminism and environmentalism." Id. ¶¶ 15-16.
21. Plaintiff Nerve.com, Inc. ("Nerve") is an online magazine consisting of original fiction, personal essays, columns, photography, video, blogs, quizzes, polls, and crosswords. Griscom Testimony, 10/23 Tr. 61:15-62:5; Pl. Ex. 38. Nerve is run by Rufus Griscom ("Griscom"). J. Ex. 1 ¶ 17. According to Griscom, "Nerve is, in theory and hopefully in practice, a smart magazine about sex and culture." Griscom Testimony, 10/23 Tr. 52:13-14.
22. Plaintiff Aaron Peckham d/b/a Urban Dictionary operates an online dictionary of contemporary slang "whose terms and definitions are solely user-generated and user-rated." J. Ex. 1 ¶¶ 18-19.
23. Plaintiff Philadelphia Gay News ("PGN"), is the "oldest gay newspaper in Philadelphia" and publishes both in print and online. The online and print editions "share much of the same content, including national and local news stories written by PGN correspondents, arts and events sections, regular columns, a calendar of events, and editorials on a variety of social and political topics." The online edition also contains personal and classified advertisements. Id. ¶¶ 20-21.
24. Plaintiff American Booksellers Foundation For Free Expression ("ABFFE") is a non-profit organization founded by the American Booksellers Association. Plaintiff Powell's Bookstore is a member of ABFFE. Id. ¶ 22.
25. Plaintiff Powell's Bookstore operates seven bookstores in Portland, Oregon and states that it is the "world's largest independent new and used bookstore." Powell's Bookstore also operates a website that "allows users to browse and purchase new, used, rare, and out-of-print books." Id. ¶ 23-24.
26. Plaintiff Salon Media Group, Inc. ("Salon") publishes an online magazine featuring articles on current events, the arts, politics, the media, and relationships and states that it "is a well-known, popular on-line magazine" that contains "news articles; commentaries on and reviews of music, art, television, and film; and regular columns on politics, relationships, the media, business, and other areas of interest." Salon also has music and video downloads and user-generated content. Salon's "goal is to break news as well as produce the most compelling sort of social commentary . . . on the web," and it seeks to attract a "broad general interest audience" with its readership. Joan Walsh ("Walsh") is the Editor and Chief of Salon. Id. ¶¶ 25-26; Walsh Testimony, 10/23 Tr. 107:17-18, 112:8-12.
27. Plaintiff Sexual Health Network owns and operates Sexualhealth.com which is "dedicated to providing easy access to sexuality information, education, support and other sexuality resources for everyone, including those with disability, chronic illness or other health-related problems." The Web site is run by Dr. Mitchell Tepper ("Dr. Tepper"). Id. ¶¶ 27-28.
28. Plaintiff Electronic Privacy Information Center ("EPIC") "is a nonprofit educational organization established in 1994 to examine civil liberties and privacy issues arising on the Internet." EPIC alleges that it accesses information on the Internet, including sexually explicit pages, as part of its mission, which includes reporting on how well content filters work. Id. ¶ 29.
C. The Experts
1. Plaintiffs' Experts
29. Professor Lorrie Faith Cranor ("Dr. Cranor") is currently employed at Carnegie Mellon University as an associate research professor in the school of computer science. Cranor Testimony, 10/23 Tr. 201:25-202:4. Dr. Cranor was qualified in this case as an expert in the areas of Internet filtering products and other parental control tools used to control access to material on the Internet. Cranor Testimony, 10/23 Tr. 227:6-21.
30. Professor Edward William Felten ("Dr. Felten") is currently employed at Princeton University as a tenured professor of computer science and public affairs and director of the Center for Information Technology Policy. Felten Testimony, 10/24 Tr. 182:2-22. Dr. Felten was qualified in this case as an expert on the technology and use of Internet protocols, the technology and use of filtering products, and the technology and use of search engines. Felten Testimony, 10/24 Tr. 186:25-187:16.
31. Professor Matthew Alan Zook ("Dr. Zook") is currently employed at the University of Kentucky as an assistant professor of geography. Zook Testimony, 10/26 Tr. 53:12-19. Dr. Zook was qualified in this case as an expert in the area of Internet geography, which entails finding the locations of people using the Internet. Zook Testimony, 10/26 Tr. 66:22-67:5, 70:23-71:12.
32. Michael Russo ("Russo") is currently employed as the president of the YNOT Network. Russo Testimony, 10/25 Tr. 67:14-16. Russo was qualified as an expert in this case on the effectiveness of various online verification schemes, including payment card screens and data verification services, the availability of adult material outside the United States, and the availability of adult materials on the Internet. Russo Testimony, 10/25 Tr. 106:20-107:19.
33. Professor Ronald Mann ("Mann") is currently employed at the University of Texas as a law professor of electronic commerce and payment systems. Mann Testimony, 11/6 Tr. 59:6-10. Mann was qualified in this case as an expert on payment systems, payment card companies, business models of payment card companies, and the use of payment cards in E-commerce. Mann Testimony, 11/6 Tr. 73:2-75:16.
34. Professor Henry Reichman ("Dr. Reichman") is currently employed at California State University, East Bay, and is the associate editor of the American Library Association's "Newsletter on Intellectual Freedom." Reichman Testimony, 10/30 Tr. 5:5-7, 7:1-8:5. Dr. Reichman was qualified in this case as an expert in the area of censorship and the suppression of speech. Reichman Testimony, 10/30 Tr. 18:2-20.
2. Defendant's Experts
35. Dr. Jeffrey Eisenach ("Dr. Eisenach") is currently employed as chairman of Criterion Economics. Eisenach Testimony, 11/13 Tr. 42:1-5. Dr. Eisenach was qualified as an expert in this case on the Internet and its impact on markets and public policy. Eisenach Testimony, 11/13 Tr. 71:25-72:10.
36. Professor Stephen Neale ("Dr. Neale") is currently employed at Rutgers University as a professor of philosophy. Neale Testimony, 11/8 Tr. 186:1-5. Dr. Neale was qualified in this case as an expert on information content on the theoretical bases of linguistic classification and the classification of text documents for content. Dr. Neale was also received as an expert regarding the theoretical or inherent limits of Internet filtering software as a mechanism to block access to types of content on the Web to the extent that such software relies on text-based classification. Neale Testimony, 11/8 Tr. 203:11-205:21.
37. Professor Philip Bradford Stark ("Dr. Stark") is currently employed at the University of California, Berkley as a tenured statistics professor. Stark Testimony, 11/8 Tr. 74:1-25. Dr. Stark was qualified as an expert in this case in the areas of statistics and computer-related statistics. Stark Testimony, 11/8 Tr. 83:3-12.
38. Paul Mewett ("Mewett") is currently employed at CRA International as the head of the Internet Intelligence Unit. Mewett Testimony, 11/7 Tr. 85:18-86:11. Mewett was qualified in this case as an expert in the areas of computer technology, the interaction of computers with the Internet and the Web, and the identification of content on the Internet. Mewett Testimony, 11/7 Tr. 93:4-18.
39. Arthur E. Clark, Jr. ("Clark") is currently employed as a managing partner of Business Insights. Clark Testimony, 11/14 Tr. 11:12-13. Clark was qualified as an expert in this case on the use and effectiveness of payment cards on the Internet and related aspects. Clark Testimony, 11/14 Tr. 34:18-37:13.
40. Professor Scott Morris Smith ("Dr. Smith") is currently employed at Brigham Young University as a professor of marketing and director of the institute for marketing. S. Smith Testimony, 11/15 Tr. 4:3-9. Dr. Smith was qualified as an expert in this case in the areas of Internet research and methodology, advanced computer applications for Internet survey research and analysis, Internet marketing in businesses, and buyer or consumer behavior. S. Smith Testimony, 11/15 Tr. 40:17-42:15.
D. Information Regarding Plaintiffs' Web Sites and the Content Thereon and Select Plaintiffs' Fear of Prosecution under COPA
41. There are numerous examples of material on the plaintiffs' Web pages that contain an actual or simulated sexual act or sexual contact, an actual or simulated normal or perverted sexual act, or a lewd exhibition of the genitals or post-pubescent female breast which might be considered harmful to minors. Walsh Testimony, 10/23 Tr. 141:10-17, 144:3-145:8, 146:6-13, 146:24-147:5, 147:9-18, 147:25-148:12, 152:8-15; Pl. Ex. 39, at 1-5, 61-78, 94-135; Findings of Fact 55, 58; Griscom Testimony, 10/23 Tr. 68:16-69:1, 70:4-9, 73:25-74:6, 74:16-23, 77:10-14, 79:7-13; Pl. Ex. 38 at 13, 17-18; Findings of Fact 51-52; Glickman Testimony, 10/30 Tr. 130:23-131:10 (stating that Condomania has frank and honest discussions on the "website about anal sex, lubricants and condoms for anal sex, lubricants for such acts as fisting, dental dams for both vaginal oral sex and anal oral sex"); Peckham Testimony, 10/31 Tr. 26:13-27:4; Pl. Ex. 41 (defining sexual slang words and giving sexually graphic examples of their use); Corinna Testimony, 11/2 Tr. 81:10-15, 82:7-12, 87:25-88:6, 89:6-11, 95:10-15; Pl. Ex. 42, at 1-2, 12, 13-15, 16-17, 26-28; Findings of Fact 46-48.
42. The plaintiffs speak in support of their businesses on the Web. The speech on the plaintiffs' Web sites is designed to assist in making a profit. Walsh Testimony, 10/23 Tr. 112:20-21; Tepper Testimony, 10/30 Tr. 176:11-13; Glickman Testimony, 10/30 Tr. 92:12-13; Griscom Testimony, 10/23 Tr. 52:25-53:1; Peckham Testimony, 10/31 Tr. 23:6-10; Corinna Testimony, 11/2 Tr. 100:1-5.
43. Nonetheless, the majority of information on the plaintiffs' Web sites is provided to users for free. Walsh Testimony, 10/23 Tr. 161:4-6 (all of Salon's content is available today for free); Tepper Testimony, 10/30 Tr. 194:13-15 (all of Sexual Health Network's content is available for free); Glickman Testimony, 10/30 Tr. 108:11-20; Griscom Testimony, 10/23 Tr. 64:5-13 (most of Nerve's content is available for free); Peckham Testimony, 10/31 Tr. 55:17-56:6; Corinna Testimony, 11/2 Tr. 74:10-11 (Scarleteen.com's content is available for free), 95:25-96:1 (some of the content on Femmerotic.com is available for free), 102:1-24, 126:2-127:9 (all of the content of Scarletletters.com posted in the last two years is available for free).
44. Most of the information on the plaintiffs' Web sites can be accessed without requiring users to register, provide a password or log-in, or otherwise provide any personal, identifying information in order to access the material. Walsh Testimony, 10/23 Tr. 158:16-22; Tepper Testimony, 10/30 Tr. 190:11-18; Peckham Testimony, 10/31 Tr. 30:12-18, 42:19-23.
45. A significant number of the Internet users who access the material on the plaintiffs' Web sites are individuals who do not live in the United States. Walsh Testimony, 10/23 Tr. 113:23-114:19 ("we get roughly 20 percent of our traffic now from international readers"); Tepper Testimony, 10/30 Tr. 181:10-18 ("approximately 15 percent" of users come from overseas); Griscom Testimony, 10/23 Tr. 56:14-17 (15 percent of Nerve's visitors are from overseas); Peckham Testimony, 10/31 Tr. 25:14-17 (37 percent of Urban Dictionary's users are from overseas); Glickman Testimony, 10/30 Tr. 99:16-22.
46. Scarletletters.com is a Web site "intended to deliver sexuality information as well as entertainment by women for female users" and includes content that is sexually explicit. Corinna Testimony, 11/2 Tr. 73:2-5, 88:1-89:25; J. Ex. 1 ¶¶ 9-10, 35; Pl. Ex. 42, at 12 (showing sketches of male and female genitalia and intercourse), 13-15 (displaying an erotic story describing, inter alia, interludes involving masturbation, bondage, intercourse, and oral sex), 16 (depicting, inter alia, erotic photographs of genitalia and sexual situations).
47. Scarleteen.com is a "sex education and information clearing house that's aimed at teenagers and young adults" which includes content that is sexual explicit. Corinna Testimony, 11/2 Tr. 74:1-2, 77:24-85:23; J. Ex. 1 ¶¶ 11, 34; Pl. Ex. 42.
48. Femmerotic.com provides sexuality information "by women pertaining to women" and includes content that is sexually explicit. Corinna Testimony, 11/2 Tr. 75:1-5, 94:21-95:15, 127:13-16; J. Ex. 1 ¶¶ 12, 36; Pl. Ex. 42, at 1-2, 17, 26-28 (displaying erotic pictures of naked breasts, genitalia and buttocks and sexual situations).
49. Scarleteen.com, Scarletletters.com, and Femmerotic.com are operated to make a profit. Corinna Testimony, 11/2 Tr. 100:2-5.
50. Corinna does not understand the terms in COPA or what speech the statute prohibits. Corinna Testimony, 11/2 Tr. 75:20-22. Corinna also fears prosecution under COPA because she believes that some of her content is pornographic and would be prohibited by COPA. Corinna Testimony, 11/2 Tr. 76:7-17, 91:13-22, 97:13-19.
51. Nerve has speech that "frequently" includes nudity and descriptions of sexual acts. Griscom Testimony, 10/23 Tr. 58:7-13, 66:18-79:13 (inter alia, discussing the Henry Miller awards which include excerpts of the best sex scenes in American novels); Pl. Ex. 38, at 13 (depicting, inter alia, a photograph of a naked woman with stars obscuring her nipples and genitals apparently masturbating), 17-18 (describing, inter alia, vaginal and oral sex in graphic language), 22-24 (describing, in erotic detail, Tantric sex); J. Ex. 1 ¶¶ 39-40.
52. Nerve has video and blog sections that include nudity and depictions of sexual acts and sexual contact that are available for free and accessible to anyone. Griscom Testimony, 10/23 Tr. 73:25-74:6, 77:2-78:7.
53. Nerve is a for-profit venture with advertising being its largest revenue stream. Griscom Testimony, 10/23 Tr. 81:17-82:23.
54. Griscom, on behalf of Nerve, does not understand the terms in COPA or what speech the statute prohibits. Griscom Testimony, 10/23 Tr. 56:18-57:21. Griscom also fears prosecution under COPA and believes that others could find that some content on Nerve is harmful to minors. Griscom Testimony, 10/23 Tr. 58:3-59:7, 80:7-17.
55. Salon's Web site, Salon.com, contains content that describes and depicts sexual acts and sexual contact and exhibitions of the genitals or post-pubescent female breast. Walsh Testimony, 10/23 Tr. 141:10-17, 144:3-145:8 (discussing article entitled "My Date With A Virtual Sex Machine"), 146:6-13, 146:24-147:5 (discussing sexually explicit Japanese wood cuts), 147:9-18 (discussing sex gallery photographs from Kinsey Institute), 147:25-148:12 (discussing Abu Ghraib prison photographs including one in which a prisoner is "appearing to sodomize himself"), 152:8-15 (discussing explicit photographs on a blog entitled "My So-Called Lesbian Life"); Pl. Ex. 39, at 1-5 (depicting, inter alia, a photograph of two topless women in an erotic position, and two other photographs of nude women posing erotically), 24, 29 (describing sexual encounters involving oral and vaginal sex), 57-59 (discussing anal sex with a strap-on phallus), 64-74 (describing a memoir of anal sex), 75-78, 94-100 (depicting Japanese wood cuts involving, inter alia, oral sex with sheet-like demons, hot wax, children engaging in digital penetration, intercourse with a half-woman, half-octopus creature, vaginal penetration with a demon's nose, and clitoral stimulation with chop-sticks), 101-111 (depicting Kinsey Institute images involving bondage, naked breasts, buttocks, erect penises, vaginas, and a variety of sexual situations including a woman straddling a seated man whose penis is penetrating the woman's vagina), 112-118 (depicting erotic photographs of naked breasts buttocks and genitalia), 119-135.
56. Salon is a for-profit company which primarily generates revenue through online advertising. Walsh Testimony, 10/23 Tr. 112:20-21, 157:17-158:4.
57. Walsh, on behalf of Salon, does not understand the terms in COPA or what speech the statute prohibits. Walsh Testimony, 10/23 Tr. 135:22-137:5. Walsh also believes that some of Salon's content might be considered harmful to minors. Walsh Testimony, 10/23 Tr. 137:15-138:10.
58. Walsh fears prosecution under COPA, in part, because Salon has received complaints about the sexual nature of some of its material and has lost some advertising due to articles incorporating unpopular views on sexuality. Walsh Testimony, 10/23 Tr. 154:19-156:7, 157:4-15.
59. Glickman, on behalf of Condomania, does not understand the terms in COPA or what speech the statute prohibits. Glickman Testimony, 10/30 Tr. 130:3-22.
60. Dr. Tepper, on behalf of Sexual Health Network, does not understand the terms in COPA or what speech the statute prohibits. Tepper Testimony, 10/30 Tr. 195:14-197:2.
61. If COPA is enforced, Griscom, Walsh, Glickman, and Corinna would have various reactions such as considering moving overseas and risking prosecution. Griscom Testimony, 10/23 Tr. 91:7-17; Walsh Testimony, 10/23 Tr. 173:21-24; Glickman Testimony, 10/30 Tr. 136:13-18; Corinna Testimony, 11/2 Tr. 104:17-105:5. Dr. Tepper does not know what he would do if COPA went into effect because he does not think it would be financially feasible for his company to use an age verification system and he does not know if he would be willing to risk violating COPA. Tepper Testimony, 10/30 Tr. 243:7-17.
E. Sexually Explicit Materials Available on the Web
1. In General
62. A little more than 1 percent of all Web pages on the Surface Web (amounting to approximately 275 million to 700 million Web pages) are sexually explicit. Zook Testimony, 10/26 Tr. at 88:22-89:17; Mewett Testimony, 11/8 Tr. 48:19-49:9; Pl. Ex. 29, at 6-7; Pl. Ex. 54, at 101; Def. Ex. 65.
2. The Amount of Foreign Sexually Explicit Material on the Web
63. Although the parties disagree on how to determine whether a Web site is foreign or domestic in origin, their experts' views regarding the amount of foreign sexually explicit materials available on the Web are not dissimilar. The evidence submitted by the plaintiffs' expert Dr. Zook shows that 32 percent of adult membership Web sites and 58 percent of free adult Web sites originate from outside the United States. Zook Testimony, 10/26 Tr. 111:2-112:1; Pl. Ex. 29 at 18-19. The evidence submitted by defendant's expert Dr. Stark shows that 55.8 percent of the Web pages randomly sampled from the Google search engine index were hosted outside of the United States and 44.4 percent of the Web pages randomly sampled from the MSN search engine index were hosted outside of the United States. Def. Ex. 62 ¶ 10; Def. Ex. 65. From the weight of the evidence excerpted here, I find that a substantial number (approximately 50 percent) of sexually explicit websites are foreign in origin.
64. Dr. Stark's data also indicated that, of the sexually explicit Web pages returned in response to a random sample of search terms entered into the AOL, MSN and Yahoo! search engines, 11.6 percent of those Web pages were from foreign Web sites. Dr. Stark's data further indicated that, of the sexually explicit Web pages returned in response to the most popular search terms according to Wordtracker, which markets lists of the most popular search terms, 12.6 percent of those Web pages were from foreign Web sites. Def. Ex. 62, at 10-11; Def. Ex. 65. However, I find that this data is not relevant because I find that Dr. Stark's samples from search engine index data (detailed in Finding of Fact 63), which consists of all of the Web sites indexed by the search engines, are a more accurate indication of how many Web sites are sexually explicit and foreign than samples of search term results which show only the frequency with which searches return sexually explicit Web pages.
65. The National Research Council ("NRC") report, commissioned by Congress, specifically noted that some estimates place as much as 75 percent of adult membership Web sites overseas. Pl. Ex. 54, at 101.
66. The percentage of adult Web sites registered overseas is increasing while, in the past five years, there has been a corresponding decrease in the percentage of adult Web sites located in the United States. Zook Testimony, 10/26 Tr. 107:24-112:9; Pl. Ex. 29, at 13-17, Tables 4, 6, 8, 9. Free adult Web sites are migrating at the highest rates. From 2001 to 2006, the United States' share of free adult Web sites dropped from 60 percent to 42 percent. Zook Testimony, 10/26 Tr. 109:25-111:1; Pl. Ex. 29, Table 9. F. Internet Content Filtering Technology and its Effectiveness
1. In General
67. Internet content filters ("filters") are computer applications which, inter alia, attempt to block certain categories of material from view that a Web browser or other Internet application is capable of displaying or downloading, including sexually explicit material. Filters categorize and block Web sites or pages based on their content. By classifying a site or page, and refusing to display it on the user's computer screen, filters can be used to prevent children from seeing material that might be considered unsuitable. In addition, businesses often use filters to prevent employees from accessing on employer controlled computers Internet resources that are either not work related or otherwise deemed inappropriate. J. Ex. 1 ¶ 85.
68. Filters can be programmed or configured in a variety of different ways according to, inter alia, the values of the parents using them and the age and maturity of their children. As discussed more fully below, filters can be set up to restrict materials available on Web pages and other Internet applications based on numerous factors including the type of content they contain, the presence of particular words, the address of the Web site, the Internet protocol used, or computer application used. Some filters can also restrict Internet access based on time of day, day of week, how long the computer has been connected to the Internet, or which user is logged onto a computer. Cranor Testimony, 10/23 Tr. 233:1-234:13, 250:1-251:16; Cranor Testimony, Tr. 10/24 Tr. 5:9-6:8, 7:1-8:9; Allan Testimony, 11/2 Tr. 204:22-207:16, 236:22-237:7, 238:16-20, 240:18-241:8, 246:19-247:21; Allan Testimony, 11/6 Tr. 5:22-6:2; Whittle Testimony, 10/31 Tr. 200:2-16, 201:18-212:24, 212:25-213:24, 220:5-221:14; Murphy Testimony, 11/1 Tr. 210:7-213:25, 217:21-221:7; Pl. Ex. 6; Pl. Ex. 11; Pl. Ex. 54; Pl. Ex. 86.
69. Some filters can be purchased on a Compact Disc ("CD") or downloaded from the Internet and installed on a personal computer. Some filters are designed to be run on a server in a business, library, or school environment. Other filters are built into the services provided by Internet Service Providers ("ISP"). J. Ex. 1 ¶ 86.
70. Filters use different mechanisms to attempt to block access to material on the Internet including: black lists, white lists, and dynamic filtering. Id. ¶¶ 87, 90, 92.
71. Black lists are lists of URLs or Internet Protocol ("IP") addresses that a filtering company has determined lead to content that contains the type of materials its filter is designed to block. Id. ¶ 87.
72. White lists are lists of URLs or IP addresses that a filtering company has determined do not lead to any content its filter is designed to block, and, thus, should never be blocked. A very restrictive filter, like a "walled garden" filter, might block all URLs except those included on a white list. Id. ¶ 90.
73. In addition to its own black and white lists, filters often give parents or administrators the option of creating customized black or white lists. Id. ¶ 91.
74. Dynamic filtering products use artificial intelligence to analyze Web site content in real-time as it is being requested and determine whether it should be blocked by evaluating a number of different parts of the content, both what the user can actually see on the Web page, and the various hidden pieces of information contained with the content that are part of its software code or script, known as the "metadata." Among other things, dynamic filters analyze the words on the page, the metadata, the file names for images, the URLs, the links on a page, the size of images, the formatting of the page, and other statistical pattern recognition features, such as the spatial patterns between certain words and images, which can often help filters categorize content even if the actual words are not recognized. Id., ¶ 92; Cranor Testimony, 10/23 Tr. 239:23-244:18.
75. In addition to analyzing the content of Web pages, dynamic filters also take the context of the page into consideration, to ensure that the determinations are as accurate as possible. For example, many companies will develop templates that provide additional context to teach the software how to recognize certain contexts — for example, to block the word "breast" when used in combination with the word "sexy," but not when used in combination with the words "chicken" or "cancer." The software analyzes context, in part, by utilizing statistical pattern recognition techniques to identify common features of acceptable and unacceptable Web pages, depending on the context in which the content appears. Cranor Testimony, 10/23 Tr. 243:5-244:6; Whittle Testimony, 10/31 Tr. 201:4-17, 204:17-205:25.
76. Filters can be used by parents to block material that is distributed on the Web and on the other widely used parts of the Internet through protocols other than HTTP and through other Internet applications. For example, filters can be used to block any Internet application, including email, chat, instant messaging, peer-to-peer file sharing, newsgroups, streaming video and audio, Internet television and voice over Internet protocol ("VoIP"), and other Internet protocols such as FTP. J. Ex. 1 ¶ 95; Cranor Testimony, 10/24 Tr. 47:25-48:19; Pl. Ex. 6; Pl. Ex. 8; Pl. Ex. 54; Pl. Ex. 86; Pl. Ex. 88; Whittle Testimony, 10/31 Tr. 207:17-209:24, 212:25-213:15, 220:14-20; Murphy Testimony, 11/1 Tr. 203:21-204:3, 217:19-218:22; Allan Testimony, 11/2 Tr. 236:22-239:5, 246:19-248:10; Allan Testimony, 11/6 Tr. 5:22-8:23.
77. In addition to blocking access to these Internet applications completely, some products provide parents with the option of providing limited access to these applications. For example, instant messaging and email may be permitted, but some of the filtering products will only permit the sending and receiving of messages from certain authorized individuals, and will block e-mails or instant messages containing inappropriate words or any images. Filtering programs can also completely prevent children from entering or using chat rooms, or some can merely filter out any inappropriate words that come up during a chat session. Cranor Testimony, 10/24 Tr. 7:1-17; Allan Testimony, 11/2 Tr. 236:24-237:3, 238:11-20; Allan Testimony, 11/6 Tr. 5:22-7:8; Whittle Testimony, 10/31 Tr. 202:10-209:24, 212:25-213:15, 220:14-20; Murphy Testimony, 11/1 Tr. 210:7-213:25, 217:19-221:7, 235:21-238:13; Pl. Ex. 86, at 6-7.
78. Some filtering programs offer only a small number of settings, while others are highly customizable, allowing a parent to make detailed decisions about what to allow and what to block. Filtering products do this by, among other things, enabling parents to choose which categories of speech they want to be blocked (such as sexually explicit material, illicit drug information, information on violence and weapons, and hate speech) and which age setting they want the product to apply. For example, AOL's filtering product enables parents to choose from four different age settings: general (unrestricted); mature teen; young teen; and kids only. Surfcontrol's product has 13 different categories of speech that can be blocked if a parent so desires. Cranor Testimony, 10/23 Tr. 233:2-21; Pl. Ex. 86; Allan Testimony, 11/2 Tr. 205:16-207:16, 240:18-243:2; Whittle Testimony, 10/31 Tr. 200:2-16, 202:10-203:8, 206:23-212:13, 220:5-25; Murphy Testimony, 11/1 Tr. 210:7-213:25, 217:25-221:7.
79. Filtering products can be used by parents even if they have more than one child. For example, if a family has four children, many filtering products will enable the parent to set up different accounts for each child, to ensure that each child is able to access only the content that the parents want that particular child to access. Cranor Testimony, 10/23 Tr. 239:11-22; Cranor Testimony, 10/24 Tr. 36:16-38:11; Pl. Ex. 86, at 15-22, 33-39.
80. Filtering products block both Web pages originating from within the United States and Web pages originating from outside the United States. The geographic origin of a Web page is not a factor in how a filter works because the filter analyzes the content of the Web page, not the location from which it came. Cranor Testimony, 10/24 Tr. 46:20-47:8; Pl. Ex. 6; Pl. Ex. 54; Allan Testimony, 11/2 Tr. 185:6-12; Whittle Testimony, 10/31 Tr. 202:1-3; Murphy Testimony, 11/1 Tr. 224:6-14, 226:6-228:21; Pl. Ex. 133.
81. Filtering products block both non-commercial and commercial Web pages. It does not make a difference to filtering products' effectiveness if a page is from a non-commercial or a commercial entity. Cranor Testimony, 10/24 Tr. 47:9-24; Pl. Ex. 6; Pl. Ex. 54; Whittle Testimony, 10/31 Tr. 202:7-9; Allan Testimony, 11/2 Tr. 246:6-11.
82. In addition to their content filtering features, filtering products have a number of additional tools to help parents control their children's Internet activities. Other tools available to parents include monitoring and reporting features that allow supervising adults to know which sites a minor has visited and what other types of activities a minor has engaged in online. AOL, for example, offers a feature called AOL Guardian, which provides a parent with a report indicating which Web sites a child visited, which sites were blocked, the number of emails and instant messages a child sent, and to whom a child sent email or instant messages. Surfcontrol similarly provides parents with reports of the Web sites a child has visited, as well as those that were blocked. Surfcontrol's product also has the ability simply to monitor a child's activity without actually blocking anything, if a parent prefers that option. Some of the products, such as Contentwatch's filter, have features that permit parents to monitor their child's Internet activities remotely, for example, while they are at work, and some products even send email alerts to parents when inappropriate material is accessed by a child so that, if a parent so desires, it can supervise their child's Internet activities even when they are not physically with the child. Cranor Testimony, 10/23 Tr. 234:2-13, 249:21-251:15; Cranor Testimony, 10/24 Tr. 28:5-29:13; Pl. Ex. 2; Pl. Ex. 86, at 10-13, 32; Whittle Testimony, 10/31 Tr. 210:2-212:13; Murphy Testimony, 11/1 Tr. 218:23-220:23.
83. Some Internet content is now capable of being viewed on devices other than traditional personal computers. Examples include mobile devices such as cellular phones, personal digital assistants ("PDAs") such as the Blackberry, portable audio/video players such as the iPod, and game consoles such as the XBox or PlayStation. J. Ex. 1 ¶ 96.
84. Several vendors, including large, experienced software companies, currently offer content filtering products for alternative devices. Examples include products offered by Ace*comm, Bytemobile, Blue Coat, Cisco, and RuleSpace. Felten Testimony, 10/25 Tr. 25:4-20; Sena Testimony, 11/2 Tr. 33:4-6, 60:7-14; Pl. Ex. 13, at 22-23; Pl. Ex. 70; Allan Testimony, 11/2 Tr. 223:2-23.
85. At this time, however, there are no U.S. mobile telecommunications carriers that use filters for their cellular phones other than walled garden filters and certain other parental control features which can prevent children from using chat rooms, instant messaging, text messaging, email, purchasing any file downloads or having any access to the Internet at all. Felten Testimony, 10/25 Tr. 25:22-26:17; Ryan Testimony, 11/6 Tr. 30:20-36:19; Allan Testimony, 11/2 Tr. 223:2-23.
86. Nonetheless, mobile carriers are actively soliciting bids for the provision of mobile content filtering services. The top five mobile carriers in the United States, Cingular, Verizon Wireless, T-Mobile, Sprint, and Alltel, are all soliciting bids. Sena Testimony, 11/2 Tr. 56:19-57:09.
2. The Availability and Cost of Filters
87. Filters are widely available and easy to obtain. Numerous filtering products are sold directly to consumers, either in stores or over the Internet. Filters are also readily available through ISPs. Because most ISPs offer filtering products, a parent does not have to do anything to obtain a filter other than to activate it through the ISP's Web site or to call the ISP. Cranor Testimony, 10/24 Tr. 8:8-9:9.
88. Many of the ISPs offer filters to their customers for free. AOL's filter is now even available for free to anyone who wants to use it, even non-AOL subscribers. Cranor Testimony, 10/24 Tr. 9:10-24.
89. Non-ISP filtering products vary in cost, ranging from approximately $20 to $60. Cranor Testimony, 10/24 Tr. 9:10-17.
90. Most of the filtering products offer money-back guarantees or free trial periods, so that parents can simply download a filtering product for free over the Internet and then use it for a set time period to see if it is something that they want to continue using. Cranor Testimony, 10/24 Tr. 12:12-22; Eisenach Testimony, 11/13 Tr. 177:8-25.
91. Microsoft's new operating system for personal computers, Vista, also includes parental controls and filters which are available at no additional cost to users of computers with the Vista operating system. Vista's content filter provides features similar to what are found in most current filtering products, including the ability to select which categories of speech should be filtered. Vista's filter also provides parents with other access control tools, such as time management, the ability to filter non-Web Internet applications like email, and the ability to block or restrict access to online games. Cranor Testimony, 10/24 Tr. 12:23-13:7, 16:14-17:8; Pl. Ex. 2.
3. Filter Ease of Use and User Satisfaction
92. Filtering programs are fairly easy to install, configure, and use and require only minimal effort by the end user to configure and update. Cranor Testimony, 10/24 Tr. 21:3-39:7; Pl. Ex. 3, at 4-5; Pl. Ex. 6; Pl. Ex. 54, at 317-320; Pl. Ex. 85, at 4; Pl. Ex. 86.
93. The plaintiffs' expert Dr. Cranor has confirmed this finding in various tests performed over the past decade in connection with her work for the Internet Online Summit, her testimony before the COPA Commission, and her expert testimony in the five previous lawsuits challenging state versions of COPA. For example, Dr. Cranor recently tested four filters and found that three were very, very easy to use and one was somewhat easy to use. Dr. Cranor also found that the current versions of the filter products had improved and were easier to use than the older versions. Cranor Testimony, 10/24 Tr. 18:13-19:1, 19:2-13, 168:11-18.
94. Dr. Cranor's opinion is consistent with the findings of filtering studies conducted over the years. Those studies have found that many filtering products require little effort for parents to install and use. For example, a study conducted for NetAlert and the Australia Broadcast Authority concluded that certain products, such as AOL's filter, were quite easy to use and install. Cranor Testimony, 10/24 Tr. 57:9-18, 68:7-21; Pl. Ex. 5, at 32; Pl. Ex. 6, at 21; Pl. Ex. 85, at 4.
95. Almost all parents will be able to install filtering products and use them by selecting from one of their standard settings. Many filters have user interfaces that are quite easy to use and that make it easy for users to create customized settings, especially if all they are concerned about blocking is adult material. Cranor Testimony, 10/24 Tr. 19:19-20:7, 20:19-21:2, 27:1-24; Whittle Testimony, 10/31 Tr. 200:2-16, 206:23-212:13; Murphy Testimony, 11/1 Tr. 221:8-224:5; Pl. Ex. 2, at 17; Pl. Ex. 6; Pl. Ex. 85, at 4; Pl. Ex. 86, at 8-9.
96. Installing and setting up a filter will usually take a typical computer user no more than ten or fifteen minutes. The installation and set-up process is not technically complex and does not require any special training or knowledge. Cranor Testimony, 10/24 Tr. 21:3-22:8; Pl. Ex. 86.
97. Configuring a filtering product for more than one child is straightforward and easy with many products. For example, it takes about two minutes to set up an account for an additional child using AOL's filter product. Cranor Testimony, 10/24 Tr. 36:16-38:15; Pl. Ex. 86, at 33-39.
98. Most filtering products do not pose any compatibility issues for computers, meaning that using filters will not affect the typical user's ability to use other computer software. Cranor Testimony, 10/24 Tr. 40:9-41:6.
99. A study done for AOL found that 85 percent of parents are highly satisfied with their AOL Parental Controls products, and that 87 percent of the parents find them easy to use. Surfcontrol has also found that customer response is positive and that 70 to 80 percent of their customers renew their subscriptions to Surfcontrol's filter. Cranor Testimony, 10/24 Tr. 83:7-11, 129:9-130:13; Murphy Testimony, 11/1 Tr. 222:25-223:20; Pl. Ex. 85, at 4.
4. The Effectiveness of Filters
a. In General
100. There are two main concerns regarding the effectiveness of filters: underblocking and overblocking. Underblocking occurs when the filter fails to block content that the filter is configured to block. Overblocking occurs when the filter prevents access to material that it is not configured to block. Cranor Testimony, 10/24 Tr. 52:16-21; Stark Testimony, 11/8 Tr. 95:16-96:13, 105:25-106:12.
101. The plaintiffs contend that in determining whether filters are effective, the filter's underblocking rate is more important than its overblocking rate. Cranor Testimony, 10/24 Tr. 52:22-53:6. Defendant claims that overblocking is a significant concern as well. Stark Testimony, 11/8 Tr. 95:22-96:18. While both aspects are important, I agree with the plaintiffs that underblocking is the more important concern since the underlying issue in this case is the prevention of children from accessing sexually explicit material deemed harmful to them. Moreover, when a filter overblocks, a parent may add the Web sites that were erroneously overblocked to the filter's white list so that those Web sites are not blocked again. J. Ex. 1 ¶ 91.
102. Even though the Web is very large, only a small fraction of it is actually viewed frequently. To ensure that those parts that are actually being viewed by users have been located, filtering companies review lists of the most popular Web sites because the pages on those sites are the most likely ones that a child will be able to find and access. Cranor Testimony, 10/23 Tr. 236:22-237:7; Murphy Testimony, 11/1 Tr. 194:6-196:6.
103. Filtering products have improved over time and are now more effective than ever before. This is because, as with all software, the filtering companies have addressed problems with the earlier versions of the products in an attempt to make their products better. Cranor Testimony, 10/24 Tr. 81:18-82:10; Murphy Testimony, 11/1 Tr. 194:6-196:6, 221:8-222:24.
104. Another reason the effectiveness of filtering products has improved is that many products now provide multiple layers of filtering. Whereas many filters once only relied on black lists or white lists, many of today's products utilize black lists, white lists, and real-time, dynamic filtering to catch any inappropriate sites that have not previously been classified by the product. Cranor Testimony, 10/23 Tr. 246:20-247:9; Cranor Testimony, 10/24 Tr. 81:18-82:4.
105. There is a high level of competition in the field of Internet content filtering. That factor, along with the development of new technologies, has also caused the products to improve over time. Murphy Testimony, 11/1 Tr. 223:21-224:5; Pl. Ex. 2, at 16-17.
106. One of the features of filtering programs that adds to their effectiveness is that they have built-in mechanisms to prevent children from bypassing or circumventing the filters, including password protection and other devices to prevent children from uninstalling the product or changing the settings. Some products even have a tamper detection feature, by which they can detect when someone is trying to uninstall or disable the product, and then cut off Internet access altogether until it has been properly reconfigured. Cranor Testimony, 10/24 Tr. 86:19-87:21; Felten Testimony, 10/25 Tr. 37:5-38:7; Murphy Testimony, 11/1 Tr. 216:12-217:18; Whittle Testimony, 10/31 Tr. 215:7-14; Pl. Ex. 2; Pl. Ex. 86.
107. Filtering companies actively take steps to make sure that children are not able to come up with ways to circumvent their filters. Filtering companies monitor the Web to identify any methods for circumventing filters, and when such methods are found, the filtering companies respond by putting in extra protections in an attempt to make sure that those methods do not succeed with their products. Cranor Testimony, 10/24 Tr. 86:19-87:21; Felten Testimony, 10/25 Tr. 38:8-39:1.
108. It is difficult for children to circumvent filters because of the technical ability and expertise necessary to do so by disabling the product on the actual computer or by accessing the Web through a proxy or intermediary computer and successfully avoiding a filter on the minor's computer. Cranor Testimony, 10/24 Tr. 86:19-87:21; Felten Testimony, 10/25 Tr. 36:6-40:4; Murphy Testimony, 11/1 Tr. 216:12-217:18; Whittle Testimony, 10/31 Tr. 215:7-14.
109. Accessing the Web through a proxy or intermediary computer will not enable a minor to avoid a filtering product that analyzes the content of the Web page requested, in addition to where the page is coming from. Any product that contains a real-time, dynamic filtering component cannot be avoided by use of a proxy, whether the filter is located on the network or on the user's computer. Felten Testimony, 10/25 Tr. 38:08-39:24.
b. Study Results
110. Based upon the testimony of Dr. Cranor, which I accept, I find that filters generally block about 95% of sexually explicit material. Cranor Testimony, 10/24 Tr. 55:8-23.
111. One study, conducted for NetAlert and the Australia Broadcast Authority, measured the effectiveness of various filtering products at blocking a variety of different categories of content that parents might want to block, including pornography and erotica. The study found that some products, such as AOL's filter, blocked close to 100 percent of all pornography or erotica when the most restrictive setting (for children under the age of 12) was chosen. When a less restrictive setting (for 13 to 15 year-olds) was selected, the study found that about 90 percent of the pornography and erotica was blocked. Pl. Ex. 5, at 35-36.
112. Another study, conducted by Corey Finnel ("Finnel") for the government in another case, analyzed the overblocking rates of three filtering products. Finnel found that the overblocking rates for those three products respectively were between 4.69 percent and 7.99 percent, between 5.25 percent and 11.03 percent, and between 6.92 and 9.36 percent, using a 95 percent confidence interval. Cranor Testimony, 10/24 Tr. 60:4-61:25; Pl. Ex. 4.
113. Consumer Reports has also conducted reviews of the various filtering products available to parents. Their most recent study concluded that filters are very good or excellent at blocking pornography, and that they block most, but not all, of that content. More specifically, Consumer Reports found that three products, from AOL, KidsNet and MSN, blocked practically every pornographic site that they tested, and that the least effective product they tested still blocked 88 percent of pornography. Cranor Testimony, 10/24 Tr. 70:9-22; Pl. Ex. 8, at 3. Although the methodology for this study may well be less rigorous than that of other more academic studies, the study is still informative because Consumer Reports focuses its evaluations on the criteria that are important to potential consumers and helps to shed light on whether the filters tested will be usable by a parent. Cranor Testimony, 10/24 Tr. 69:14-70:4.
114. Two separate reports commissioned by Congress, from the Commission on Child Online Protection ("COPA Commission") and the NRC, have confirmed that content filters can be effective at preventing minors from accessing harmful materials online. Cranor Testimony, 10/24 Tr. 71:2-76:5; Pl. Ex. 6; Pl. Ex. 54.
115. The COPA Commission was established by Congress as part of the COPA legislation. The COPA Commission report concluded that although filters are not perfect, server-side filters (meaning filters provided by an ISP) using only black lists and not utilizing other technologies such as dynamic filtering "can be highly effective" and client-side filters (meaning filters installed on a home computer) using only black lists "can be effective" in "directly blocking access to global harmful to minors content on the Web and also on newsgroups, email, and chat rooms." Cranor Testimony, 10/24 Tr. 71:22-72:9; Pl. Ex. 6, at 19, 21.
116. The NRC issued a lengthy report in 2005. The NRC report concluded that although not perfect because filters overblock and underblock, and children can gain access to computers without filters, "[f]ilters have some significant utility in denying access to content that may be regarded as inappropriate" and, "filters can be highly effective in reducing the exposure of minors to inappropriate content if the inability to access large amounts of appropriate material is acceptable." Cranor Testimony, 10/24 Tr. 75:15-76:3; Pl. Ex. 54, at 40, 331.
117. Defendant's expert Mewett found that, in the filter study conducted by Dr. Stark and himself, with regard to the Web pages that were returned in response to the most popular search terms, the AOL filter performed the best and blocked 98.7 percent of sexually explicit Web pages. However, Mewett found that the AOL filter overblocked 19.6 percent of non-sexually explicit Web pages. Mewett also found that the other filters he tested accurately blocked between 98.6 and 87.4 percent of the sexually explicit Web pages. In fact, of the filters tested, only two failed to block at least 90 percent of the sexually explicit Web pages and the vast majority blocked at least 95 percent of such pages. Mewett further found that these filters overblocked between 2.9 and 32.8 percent of non-sexually explicit Web pages. Def. Ex. 78.
118. Mewett also found that, with regard to the Web pages drawn randomly from the search engine indexes, the AOL filter again performed the best and blocked between 91.1 and 91.4 percent of sexually explicit Web pages. However, Mewett found that the AOL filter overblocked 22.3 to 23.6 percent of non-sexually explicit Web pages. Mewett further found that the other filters tested accurately blocked between 87.6 and 39.8 percent of the sexually explicit Web pages and overblocked between .4 and 21.9 percent of non-sexually explicit Web pages. Def. Ex. 68.
119. Mewett also found that, with regard to the Web pages that were returned in response to a random sample of search terms, the AOL filter again performed the best and blocked 93.8 percent of sexually explicit web pages. However, Mewett found that the AOL filter overblocked 12.5 percent of non-sexually explicit Web pages. Mewett further found that the other filters he tested accurately blocked between 90 and 56.6 percent of the sexually explicit Web pages and overblocked between 0 and 20.7 percent of the non-sexually explicit Web pages. Def. Ex. 74.
120. I do not find Mewett's overblocking rates to be reliable because he sometimes concluded that a filter had overblocked even when the filter was performing exactly as intended. This occurred because Mewett was not always able to limit the filter to screening only sexually explicit material and sometimes the filter was configured to block other types of material as well. Mewett Testimony, 11/7 Tr. 210:13-212:20; see e.g. Def. Ex. 82, at 17 (stating that the "AOL [filter] does not have a default setting, nor does it allow for customization beyond choosing an appropriate age range for the surfer. Thus, AOL was tested on the mature teen setting . . . the mature teen setting should allow the surfer to visit all Web sites except those known to contain violent or adult content") (emphasis added).
121. Many of the findings in defendant's Mewett/Stark study are consistent with and similar to the findings of other filtering studies which have been conducted over the years in that the Mewett/Stark data shows that there are several filtering products that are quite effective and accurate at blocking sexually explicit material, especially the most popular Web content, and that many of the products have less than a 10 percent underblocking rate regarding such content. Cranor Testimony, 10/24 Tr. 78:3-12, 81:1-17; Def. Ex. 68; Def. Ex. 74; Def. Ex. 78.
G. Select Legislative History of COPA and the Limitations of COPA
122. According to House Report 105-775, "[t]he purpose of [COPA] is to amend the Communications Act of 1934 [ 47 U.S.C. § 151, et seq.] by prohibiting the sale of pornographic materials on the World Wide Web (or the Web) to minors." H.R. Rep. 105-775, at *5.
123. The intended "effect of [COPA] is simply to reorder the process in such a way as to require age verification before pornography is made available, essentially requiring the commercial pornographer to put sexually explicit images `behind the counter.'" Id. at * 15.
124. The House Report also lists the following Congressional findings:
(1) while custody, care, and nurture of the child resides first with the parent, the widespread availability of the Internet presents opportunities for minors to access materials through the World Wide Web in a manner that can frustrate parental supervision or control;
(2) the protection of the physical and psychological well-being of minors by shielding them from materials that are harmful to them is a compelling governmental interest;
(3) to date, while the industry has developed innovative ways to help parents and educators restrict material that is harmful to minors through parental control protections and self-regulation, such efforts have not provided a national solution to the problem of minors accessing harmful material on the World Wide Web;
(4) a prohibition on the distribution of material harmful to minors, combined with legitimate defenses, is currently the most effective means by which to satisfy the compelling government interest; and
(5) notwithstanding the existence of protections that limit the distribution over the World Wide Web of material that is harmful to minors, parents, educators, and industry must continue efforts to protect children from dangers posed by the Internet.Id. at *2.
125. COPA was drafted in direct response to the Supreme Court's decision in Reno, 521 U.S. 844 regarding the CDA. Id. at *5.
126. COPA's reach is specifically limited only to files publically accessible over the Web via HTTP or a successor protocol and does not reach other forms of communication and data transfer over the Internet including email, newsgroups, message boards, peer-to-peer and other file sharing networks, chat, instant messaging, VoIP, and FTP. 47 U.S.C. § 231(e)(1); H.R. Rep. 105-775, at *12.
127. Congress added this limitation in an attempt to not burden more speech than was necessary and to attempt to ensure that COPA was narrowly tailored, unlike the CDA. H.R. Rep. 105-775, at *12.
128. COPA does not apply to Web sites which are completely free and which do not fit within the definitions of "commercial purposes" and "engaged in the business." 47 U.S.C. § 231(e)(2).
129. This limitation was specifically added by Congress to address the Supreme Court's concern that the CDA was too broad in that it covered both commercial and non-commercial speech. H.R. Rep. 105-775, at *8, *12.
130. In discussing COPA, the Commerce Committee dismissed their opponent's concerns that a "domestic legislative solution will not stop material from being sent into the United States" and noted that the amount of foreign harmful to minors material was undocumented while "the fact remains that much of the harmful material is produced and posted in the United States." Id. at *20. The Committee further noted that the United States had an eight million dollar adult entertainment industry and concluded that "[c]learly domestic restrictions in the United States will help reduce a child's access to pornography. . . . To the extent that an international problem exists, the Committee has requested that the Commission on Online Child Protection study the matter and report back to Congress." Id.
131. In a letter dated October 5, 1998, from the Acting Assistant Attorney General to the Chairman for the Commerce Committee, explaining the views of the Department of Justice on COPA, the Assistant Attorney General explained why the Department felt that COPA would be problematic. Specifically, the Assistant Attorney General stated, inter alia, that:
The Department's enforcement of [COPA] could require an undesirable diversion of critical investigative and prosecutorial resources that the Department currently invests in combating traffickers in hard-core child pornography, in thwarting child predators, and in prosecuting large-scale and multi district commercial distributors of obscene materials.
And that:
We do not believe that it would be wise to divert the resources that are used for important initiatives . . . to prosecutions of the kind contemplated under the COPA. Such a diversion would be particularly ill-advised in light of the uncertainty concerning whether the COPA would have a material effect in limiting minors' access to harmful materials. There are thousands of newsgroups and Internet relay chat channels on which anyone can access pornography, and children would still be able to obtain ready access to pornography from a myriad of overseas web sites. The COPA apparently would not attempt to address those sources of Internet pornography, and admittedly it would be difficult to do so because restrictions on newsgroups and chat channels could pose constitutional questions, and because any attempt to regulate overseas web sites would raise difficult questions regarding extraterritorial enforcement.
Pl. Ex. 55, at 2-3. The Assistant Attorney General also stated the opinion that COPA contains "troubling ambiguities" concerning the scope of its coverage including the difference between "knowing" violations of COPA and "intentional" violations of COPA. Id. at 3.
H. Statistical Information on Obscenity Prosecutions
132. Existing laws make it illegal to distribute material over the Internet that constitutes obscenity, 18 U.S.C. ch. 71, or child pornography, 18 U.S.C. ch 110. From 2000 to 2005, defendant initiated fewer than 20 prosecutions for obscenity which did not also accompany charges of child pornography, travel in interstate commence to engage in sex with a minor, or attempting to transfer obscene material to a minor. J. Ex. 1 ¶ 121.
133. There have been fewer than 10 prosecutions for obscenity which did not also accompany charges of child pornography, travel in interstate commerce to engage in sex with a minor, or attempting to transfer obscene material to a minor since 2005.Id. ¶ 122.
I. The Affirmative Defenses in COPA and Their Availability and Effectiveness
134. It is an affirmative defense to liability under COPA when a Web site owner restricts access by minors to material that is harmful to them by requiring the use of a "credit card, debit account, adult access code, or adult personal identification number", "by accepting a digital certificate that verifies age", or by "any other reasonable measures that are feasible under available technology" in order to attempt to verify that the consumer is not a minor. 47 U.S.C. § 231(c)(1).
135. Congress included these affirmative defenses in COPA, in part, because such defenses "requiring either payment by credit card or authorization by access or identification code" had been included in "the FCC's dial-a-porn regulations [for the Communications Act of 1934, 47 U.S.C. § 223(b)], which were upheld in Dial Information Services Corp. v. Thornburgh, 938 F.2d 1535 (2d Cir. 1991)." H.R. Rep. 105-775, at *14.
136. In discussing this issue, the Commerce Committee also mistakenly asserted that these defenses had been "cited with approval in Sable [Communications of California, Inc. v. F.C.C.], [492] U.S. 115 (1989)" and stated that:
In Sable, the Court found that such commercial restrictions would be effective in excluding most juveniles, stating: "the FCC's technological approach to restricting dial-a-porn messages to adults who seek them would be extremely effective, and only a few of the most enterprising and disobedient young people would manage to secure access to such messages." [492] U.S. at 130, 109 S.Ct. 2829.Id. However, the Supreme Court in Sable actually stated:
For all we know from this record, the FCC's technological approach to restricting dial-a-porn messages to adults who seek them would be extremely effective, and only a few of the most enterprising and disobedient young people would manage to secure access to such messages. If this is the case . . . "Sable, 492 U.S. at 130 (emphasis added). The Supreme Court did note that the FCC, after "lengthy proceedings, determined that its credit card, access code, and scrambling rules were a satisfactory solution to the problem of keeping indecent dial-a-porn messages out of the reach of minors" and that the Court of Appeals had agreed with this assessment. Id. at 128. However, the Court ultimately did not rule on this issue but instead found only that Congress had not generated any "legislative findings that would justify [the Court] in concluding that there is no constitutionally acceptable less restrictive means, short of a total ban [on indecent and obscene telephone communications], to achieve the Government's interest in protecting minors." Id. at 128-129. The Supreme Court concluded only that the congressional record "contain[ed] no evidence as to how effective or ineffective the FCC's [credit card, access code, and scrambling rules] were or might prove to be." Id. at 130 (emphasis original).
137. In the physical world, assessing the validity of an assertion about a person's age is relatively straightforward because of face-to-face interactions. In a retail establishment, for example, someone seeking to purchase harmful to minors material can be asked to show identification indicating the purchaser's age. The provider can view the buyer face-to-face and compare this identification to the person presenting it. That level of assurance is not available during an Internet purchase because of the absence of face-to-face interactions over the Web. Pl. Ex. 25, at 30-31; Pl. Ex. 54, at 88, 91-92.
1. The General Availability of Age Verification Technologies
138. From the weight of the evidence, I find that there is no evidence of age verification services or products available on the market to owners of Web sites that actually reliably establish or verify the age of Internet users. Nor is there evidence of such services or products that can effectively prevent access to Web pages by a minor. Russo Testimony, 10/25 Tr. 124:1-6, 143:17-20, 157:3-158:8, 164:15-165:7, 166:14-167:12; Tepper Testimony, 10/30 Tr. 234:13-15; Peckham Testimony, 10/31 Tr. 48:7-9; Cadwell Testimony, 10/31 Tr. 162:20-164:1, 174:17-175:16; Meiser Testimony, 10/31 Tr. 120:7-13, 122:15-123:20, 123:21-125:3, 127:23-128:15, 133:20-134:8, 135:13-135:25, 136:1-8, 138:3-139:21, 139:22-141:6, 143:5-144:5; Pl. Ex. 54, at 88, 91, 376-77; Pl. Ex. 25, at 3.
2. The Effectiveness of Payment Cards as a Defense and Minors Access Thereto
139. "Traditional payment cards" consist of credit, debit, and reloadable prepaid cards. Utilization of a credit card essentially creates a loan from a card issuing company to the card holder. Clark Testimony, 11/14 Tr. 17:4-5. A debit card is a payment card that is used to make a purchase and is tied to a deposit account, primarily a checking account. Purchases of goods and services on a debit card are deducted from the individual's checking account. Clark Testimony, 11/14 Tr. 17:5-8. A prepaid card is a payment card that an individual electronically loads with cash and then proceeds to purchase goods and services. Clark Testimony, 11/14 16:24-25. Reloadable prepaid cards can be purchased for cash and after that amount is used up more money can be added; nonreloadable prepaid cards are one-time use cards such as gift cards. Clark Testimony, 11/14 Tr. 48:10-24.
140. The rules of payment card associations in this country prohibit Web sites from claiming that use of a payment card is an effective method of verifying age, and prohibit Web site owners from using credit or debit cards to verify age. Russo Testimony, 10/25 Tr. 72:23-73:10; Cadwell Testimony, 10/31 Tr. 185:15-186:11, 187:3-187:23; Thaler Testimony, 11/1 Tr. 111:20-113:20; Bergman Testimony, 11/7 Tr. 16:4-17:17, 18:21-19:25; Pl. Ex. 106, at 4; Pl. Ex. 139, at 2-4; Pl. Ex. 141, at 1.
141. Payment card associations in this country advise consumers not to offer payment cards to merchants as a proxy for age. Russo Testimony, 10/25 Tr. 72:16-73:3; Bergman Testimony, 11/7 Tr. 16:4-17:17; Pl. Ex. 139, at 2-4.
142. Defendant's expert Clark, concedes that payment cards cannot be used to verify age because minors under 17 have access to credit cards, debit cards, and reloadable prepaid cards. Clark Testimony, 11/14 Tr. 179:20-180:6.
143. Payment card issuers usually will not issue credit and debit cards directly to minors without their parent's consent because of the financial risks associated with minors. Clark Testimony, 11/14 Tr. 118:7-120:3; Rinchiuso Testimony, 11/6 Tr. 241:16-242:6; Def. Ex. 93, at 17. Nonetheless, as described below, there are many other ways in which a minor may obtain and use payment cards.
144. The plaintiffs contend that about one-half of all minors have access to credit cards, debit cards, or prepaid cards and that the percentage of 16 year-olds with access to payment cards is significantly higher than the percentage of 12 year-olds with access to such cards. Mann Testimony, 11/6 Tr. 86:12-17, 96:18-97:14, 114:17-24; Pl. Ex. 17, at 2, 7; Pl. Ex. 34, at 3-8; Pl. Ex. 93, at 10. Defendant claims that about 22 percent of minors age 12 to 17 have access to traditional payment cards either through cards in their own name (with a co-signing parent) or through a borrowed card. Clark Testimony, 11/14 Tr. 83:17-25, 84:6-85:9, 91:8-22, 96:2-96:12, 114:11-15. Clark did admit, however, that this number would be higher if he were to include minors who have access to prepaid cards which were not included in his data source. Clark Testimony, 11/14 Tr. 190:6-24, 191:5-21, 192:11-18, 193:8-13. While the parties do not agree on the exact number of minors who have access to traditional payment cards, I find from the weight of the evidence that even the lowest figures show that a significant number of minors have access to them.
145. In addition, payment card issuers are increasingly marketing credit cards, debit cards, and prepaid cards to minors as young as 13 because the "coveted youth segment" presents "dramatic revenue generating opportunities." Visa's "Visa Buxx" card is one example of a payment card that is specifically designed to be used by minors. Mann Testimony, 11/6 Tr. 75:18-77:2, 85:14-86:11; Bergman Testimony, 11/7 Tr. 4:3-6:2, 21:16-23:19, 46:20-47:25; Pl. Ex. 25, at 23-24; Pl. Ex. 34, at 6-7; Pl. Ex. 54, at 371, 377; Pl. Ex. 148, at 2, 4.
146. It is also possible for minors to have access to credit or debit cards without the knowledge or consent of their parents. Bergman Testimony, 11/7 Tr. 8:8-9:22.
147. Even if parents review periodic payment card statements, either their own or those of cards issued with their permission to their children, they may not be able to identify transactions on sexually explicit Web sites because the adult nature of such transactions is often not readily identifiable from information provided on the statement. Clark Testimony, 11/14 Tr. 214:7-10; Pl. Ex. 54, at 93. Moreover, delay in issuing a payment card statement to parents means that unauthorized access to harmful to minors materials can occur. Pl. Ex. 6, at 25.
3. The Effectiveness of Data Verification Services
148. Some companies offer non-payment card-based services that attempt to verify the age or identity of an individual Internet user. These companies are referred to as data verification services ("DVS"). They seek to accomplish in cyberspace what a clerk checking an ID card or driver's license accomplishes in an adult bookstore, only they do not verify the age or identity of an individual; instead, they merely verify the data entered by an Internet user. Russo Testimony, 10/25 Tr. 182:20-183:13, 196:18-197:1; Meiser Testimony, 10/31 Tr. 127:25-128:15; Pl. Ex. 25, at 25-26; Pl. Ex. 54, at 367, 376. IDology is one such company. Dancu Testimony, 11/9 Tr. 143:25-144:15, 161:12-25.
149. DVS companies cannot determine whether the person entering information into the Web site is the person to whom the information pertains. Nor is there any way for the person to whom the information pertains to know that his or her information has been used because the DVS companies do not notify people when their information has been verified. Dancu Testimony, 11/9 Tr. 260:12-261:12; Russo Testimony, 10/25 Tr. 97:24-98:8, 182:20-183:18; Meiser Testimony, 10/31 Tr. 127:23-128:15, 138:3-139:21, 143:23-145:3; Pl. Ex. 79, at 5; Pl. Ex. 25, at 31.
150. Internet users attempting to access content on a Web page that is using a DVS system will be required to provide specified personal information, such as the person's name, last four digits of the social security number, home address, home telephone number, or driver's license number. The DVS company will check this information against commercially available databases that aggregate public records, and then provide a response to the Web page operator who will have the ability to permit or decline access to that user. Russo Testimony, 10/25 Tr. 95:22-96:10, 181:24-182:19; Pl. Ex. 25, at 26-27; Pl. Ex. 54, at 92; Dancu Testimony, 11/9 Tr. 160:15-161:3, 164:10-166:5; Def. Ex. 109.
151. DVS companies rely on public records such as property, voting, and vehicle registration records, records from state Departments of Motor Vehicles, and some privately acquired information in an attempt to verify information. Russo Testimony, 10/25 Tr. 175:24-176:17; Dancu Testimony, 11/9 Tr. 164:10-18; Pl. Ex. 54, at 92.
152. The minimum information required by a DVS company to attempt a verification is a first name, last name, street address, and zip code. Russo Testimony, 10/25 Tr. 172:24-173:1; Dancu Testimony, 11/9 Tr. 160:15-22; Pl. Ex. 76.
153. This minimum information requirement can easily be circumvented by children who generally know the first and last name, street address and zip codes of their parents or another adult. Russo Testimony, 10/25 Tr. 97:24-98:8; Dancu Testimony, 11/9 Tr. 244:11-245:3. However, in order to heighten reliability and for an additional cost, it is possible for DVS companies to generate user-specific questions based on personal historical information, such as the color of a given car or the address of a previous home. The more questions that are asked, the more apparently effective the verification process will be. Dancu Testimony, 11/9 Tr. 177:11-181:3; Russo Testimony, 10/25 Tr. 192:10-19; Def. Ex. 109, at 5. Nonetheless, I find from the testimony that without a physical delivery of goods and an accompanying visual age verification, neither the DVS nor the Web page operator can know whether an adult or a child provided the information. Attempting to verify age with this information in a consumer-not-present transaction is therefore unreliable. Russo Testimony, 10/25 Tr. 98:6-8, 183:2-13; Dancu Testimony, 11/9 Tr. 245:18-246:5; Pl. Ex. 25, at 31; Pl. Ex. 54, at 92-93.
154. DVS companies do not have access to every state's Department of Motor vehicle records, vehicle registration records, property records or voting records and not every adult in the United States has a driver's licence, owns property, or is registered to vote. As a result, it is less likely that the information from people in such states and in such situations will be properly verified. Dancu Testimony, 11/9 Tr. 252:7-20; Meiser Testimony, 10/31 Tr. 134:11-135:25; Pl. Ex. 54, at 370.
155. DVS companies also have difficulty verifying the information of people recently married, divorced, or who otherwise have legally changed their name. Russo Testimony, 10/25 Tr. 179:4-12; Pl. Ex. 79, at 4.
156. DVS databases do not contain foreign records and cannot verify information for individuals residing outside of the United States who are not United States citizens. This limits the audience of Web sites using DVS products exclusively to Americans whose data can be verified and would be problematic for the plaintiffs who have an international audience. Finding of Fact 45; Dancu Testimony, 11/9 Tr. 253:23-254:16; Meiser Testimony, 10/31 Tr. 142:9-19; Russo Testimony, 10/25 Tr. 177:22-178:5, 180:14-181:1; Pl. Ex. 25, at 32; Pl. Ex. 79, at 1.
157. It is especially difficult for DVS companies to verify young adults (between the ages of 17 and 21) or minors, because there is little data available on younger adults, and very little, if any, data available on minors. Therefore, if Web page operators such as the plaintiffs utilize DVS products to comply with COPA, there will be young adults who will not be able to access those Web pages. Dancu Testimony, 11/9 Tr. 257:22-259:20; Meiser Testimony, 10/31 Tr. 140:1-18, 155:24-156:10; Russo Testimony, 10/25 Tr. 178:16-179:3; Pl. Ex. 79, at 4.
158. DVS companies also have difficulty verifying information for recent immigrants, visa holders, or anyone who is not a United States citizen. Dancu Testimony, 11/9 Tr. 259:21-260:5; Meiser Testimony, 10/31 Tr. 140:1-18; Pl. Ex. 79, at 4; Pl. Ex. 76.
4. The Effectiveness of Digital Certificates and Other Reasonable Measures that Are Feasible under Available Technology
159. Defendant is unaware of any digital certificates that can be used to comply with COPA. 47 U.S.C. § 231(c)(1)(B); Pl. Ex. 163, at 1-2.
160. Defendant is also unaware of any other available or feasible options which would allow Web site owners to restrict access to certain material to minors, while continuing to provide it to adults. 47 U.S.C. § 231(c)(1)(C); Pl. Ex. 163, at 1-2.
5. The Economic Burdens and Loss of Web Viewership Associated with the Affirmative Defenses
161. Based upon the below referenced evidence, I find that there are fees associated with all of the affirmative defenses and verification services identified in COPA, as well as all other services that claim to provide age verification. These fees apply any time a user attempts to access material on a Web site, even if there is no purchase. The fees must either be paid by the Web site or passed on to the users. As a result, Web sites such as the plaintiffs' sites, which desire to provide free distribution of their information, will be prevented from doing so. Russo Testimony, 10/25 Tr. 162:17-163:7, 166:1-13; Pl. Ex. 25, at 24, 33; Pl. Ex. 106, at 5, 15; Thaler Testimony, 11/1 Tr. 115:25-116:15, 116:16-117:8, 117:9-118:2; Cadwell Testimony, 10/31 Tr. 183:2-184:22; Meiser Testimony, 10/31 Tr. 146:23-147:17; Pl. Ex. 6, at 25; Pl. Ex. 54, at 93; Peckham Testimony, 10/31 Tr. 55:22-25 (Urban Dictionary's mission is to provide content for free); Corinna Testimony, 11/2 Tr. 104:13-16 (Corinna wishes to provide content for free); Griscom Testimony, 10/23 Tr. 84:25-85:14 (Nerve wants to reach widest audience possible).
162. Requiring the use of a payment card to enter a Web site would impose a significant economic cost on Web site owners. In addition to set-up fees and administrative fees, Web site owners would also need to pay fees for processing payment card information for each transaction. Russo Testimony, 10/25 Tr. 162:17-163:7; Lewis Testimony, 10/31 Tr. 109:6-24; Cadwell Testimony, 10/31 Tr. 183:2-184:22; Thaler Testimony, 11/1 Tr. 116:13-117:8, 117:9-118:2, 118:3-119:7; Corinna Testimony, 11/2 Tr. 104:17-105:7; Clark Testimony, 11/14 Tr. 239:12-240:15; Pl. Ex. 106, at 5, 15.
163. Financial institutions will not process or verify a payment card in the absence of a financial transaction. Express policies of the payment card associations prohibit online merchants who sell content from processing transactions in the amount of zero dollars. Verification by payment card will therefore be practically unfeasible for all of the plaintiffs and most other Web site operators and content providers covered by COPA who distribute their content for free. Thaler Testimony, 11/1 Tr. 119:10-16; Rinchiuso Testimony, 11/6 Tr. 242:16-24; Clark Testimony, 11/14 Tr. 214:11-218:16; Pl. Ex. 6, at 25; Pl. Ex. 34, at 10; Pl. Ex. 54, at 373.
164. Defendant contends that a product called Bitpass enables consumers to utilize credit cards for near zero-dollar transactions. Defendant's Representative, 11/9 Tr. 89:22-90:6. The Bitpass product is used to facilitate online purchases of digital content and is not currently designed for Web sites which do not sell digital content. Knopper Testimony, 11/9 Tr. 93:7-14, 111:14-112:15, 125:2-4. Moreover, Bitpass charges Web sites that use its services fees for each transaction. The fees range from 5 to 15 percent of each transaction. In addition, Bitpass may charge reporting fees and set-up fees. Knopper Testimony, 11/9 Tr. 106:17-22, 124:3-18.
165. Credit and verification charges must either be absorbed by the content provider or passed on to users. This cost will increase according to the number of visitors to a Web site. Some plaintiffs, like Nerve and Salon, have one million to three million unique visitors per month. Russo Testimony, 10/25 Tr. 162:17-163:7, 166:1-13; Griscom Testimony, 10/23 Tr. 54:25-55:4; Walsh Testimony, 10/23 Tr. 116:1-117:1; Pl. Ex. 6, at 25; Pl. Ex. 106, at 5, 15; Peckham Testimony, 10/31 Tr. 24:16-23 (Urban Dictionary has 330,000 visitors per day).
166. DVS companies like IDology charge a minimum of 37 cents for each verification transaction. The cost can rise to 97 cents per transaction, for a more complete verification service. Every time an Internet user comes to a Web page that requires its visitors to pass through a DVS screen, the DVS will charge that Web page operator something in the range of 37 to 97 cents to perform the verification service. Dancu Testimony, 11/9 Tr. 221:25-222:4; Russo Testimony, 10/25 Tr. 186:2-6, 192:10-19; Meiser Testimony, 10/31 Tr. 146:23-147:17; Pl. Ex. 25, at 26-28.
167. In addition to the per-transaction fees, DVS companies also charge other fees such as application and set-up fees and optional integration fees which range from $195 to $495. Dancu Testimony, 11/9 Tr. 222:5-18; Russo Testimony, 10/25 Tr. 192:20-193:4, 195:8-14.
168. It is not economically feasible for a Web page operator, especially one that provides free content, to verify the information of every customer that visits the Web page with a DVS. Russo Testimony, 10/25 Tr. 96:25-97:14; Pl. Ex. 25, at 30, 33. As one example, Urban Dictionary had approximately 40 million visitors between January and October of 2006. Peckham Testimony, 10/31 Tr. 24:20-23. If Urban Dictionary had been forced to have its users pass through a DVS such as IDology, and was not given a bulk discount, Urban Dictionary would have incurred costs from between $14,800,000 to $38,800,000. Dancu Testimony, 11/9 Tr. 221:25-222:4.
169. Because of the sexually explicit nature of the content on their Web sites, many Web site owners, including some of the plaintiffs, would be forced to place a credit card or age verification screen on the initial home page of their Web sites and/or on each individual Web page in order to ensure that no user could access any of the content on the site until passing through such a screen. Tepper Testimony, 10/30 Tr. 241:18-242:7; Peckham Testimony, 10/31 Tr. 48:1-6.
170. Requiring users to provide a credit card or personal information before they can browse a Web page to determine what it offers will deter most users from ever accessing those pages, causing the traffic to Web sites such as the plaintiffs' to fall precipitously. Walsh Testimony, 10/23 Tr. 161:25-162:11, 163:4-164:2, 170:7-11, 171:15-172:9, 173:10-20; Glickman Testimony, 10/30 Tr. 134:6-136:4; Tepper Testimony, 10/30 Tr. 238:9-239:17; A. Smith Testimony, 10/26 Tr. 188:24-189:10; Snellen Testimony, 11/2 Tr. 136:2-12; Clark Testimony, 11/14 Tr. 242:13-243:16.
171. For a plethora of reasons including privacy and financial concerns (discussed further below) and the fact that so much Web content is available for free, many Web users already refuse to register, provide credit card information, or provide real personal information to Web sites if they have any alternative. Because requiring age verification would lead to a significant loss of users, content providers would have to either self-censor, risk prosecution, or shoulder the large financial burden of age verification. Griscom Testimony, 10/23 Tr. 88:8-20; Walsh Testimony, 10/23 Tr. 161:25-162:11, 163:4-11, 164:20-165:16, 170:7-11, 171:5-172:9, 173:10-20; Russo Testimony, 10/25 Tr. 146:14-148:16; A. Smith Testimony, 10/26 Tr. 188:24-189:10; Glickman Testimony, 10/30 Tr. 135:14-136:4; Tepper Testimony, 10/30 Tr. 238:9-239:17; Peckham Testimony, 10/31 Tr. 51:15-53:13; Lewis Testimony, 10/31 Tr. 110:4-13; Snellen Testimony, 11/2 Tr. 136:2-12; S. Smith Testimony, 11/15 Tr. 172:13-16.
6. Web Users' Privacy Concerns and Reluctance to Provide Personal Information
a. Web Users' Privacy Concerns
172. Requiring users to go through an age verification process would lead to a distinct loss of personal privacy. Many people wish to browse and access material privately and anonymously, especially if it is sexually explicit. Web users are especially unlikely to provide a credit card or personal information to gain access to sensitive, personal, controversial, or stigmatized content on the Web. As a result of this desire to remain anonymous, many users who are not willing to access information non-anonymously will be deterred from accessing the desired information. Web site owners such as the plaintiffs will be deprived of the ability to provide this information to those users. Walsh Testimony, 10/23 Tr. 170:17-171:4; Tepper Testimony, 10/30 Tr. 178:7-23, 190:14-191:5, 212:1-10, 236:22-237:5, 238:9-239:17 (stating that anonymity is important for users with embarrassing medical and sexual questions which they would not even discuss with their mates or personal physicians); Corinna Testimony, 11/2 Tr. 103:23-104:12 (anonymity is particularly important to women seeking information about sexuality); A. Smith Testimony, 10/26 Tr. 188:24-189:10; Snellen Testimony, 11/2 Tr. 139:18-141:8, 156:8-17; Griscom Testimony, 10/23 Tr. 89:25-90:2; Pl. Ex. 54, at 372.
173. COPA's requirement that Web sites maintain the confidentiality of information submitted for purposes of age verification would not alleviate the deterrent effect of age verification on users, because users must still disclose the personal information to a Web site to pass through the screen, and then rely on these entities, many of whom are unknown and have no actual person identified with them, to comply with the confidentiality requirement. Tepper Testimony, 10/30 Tr. 238:9-239:17; Pl. Ex. 6, at 25 (noting that the "[c]ollection of individually-identifiable information at central points via this system poses privacy risks"); 47 U.S.C. § 231(d).
b. Web Users' Security Concerns
174. Requiring Internet users to provide payment card information or other personally identifiable information to access a Web site would significantly deter many users from entering the site, because Internet users are concerned about security on the Internet and because Internet users are afraid of fraud and identity theft on the Internet. S. Smith Testimony, 11/15 Tr. 116:4-6, 117:18-24; Walsh Testimony, 10/23 Tr. 171:15-172:9.
175. Although the Internet has become much more familiar to most people, numerous studies show that Internet users are concerned about identity theft and fraud on the Internet and that their fears have grown, not decreased, over time. S. Smith Testimony, 11/15 Tr. 119:11-15.
176. Consumer Reports Web Watch issued a report on Internet users' security concerns on October 26, 2005 which was based on a survey conducted by Princeton Survey Research Associates International, a reliable sources of information. S. Smith Testimony, 11/15 Tr. 119:20-120:12. The study found that nine out of ten U.S. Internet users over 18 years old have made changes to their behavior due to fear of identity theft including reducing Internet use and stopping or cutting back on making Internet purchases. S. Smith Testimony, 11/15 Tr. 120:13-121:16.
177. The Consumer Reports study also found that 88 percent of the people surveyed said that keeping personal information safe and secure is very important for a Web site they visit, and that for all online users, concern about identity theft is substantial and a worry that has changed their behavior in sweeping ways. More specifically, because of their concerns, a majority of Internet users (53 percent) have stopped giving out personal information on the Internet. S. Smith Testimony, 11/15 Tr. 121:25-122:17, 124:14-18.
178. The Javelin company, a reliable source of information, issued a report on identity fraud in 2005. The Javelin study similarly found that there are growing fears today about identity theft on the Internet. S. Smith Testimony, 11/15 Tr. 124:19-125:5.
179. In April 2006, Forrester Research, a reliable source of information, issued a report which found that overall satisfaction with e-commerce shopping experiences and credit card security trust is declining, that credit card security concerns have intensified, and that whether founded on reality or not, all online shoppers, even experienced buyers, worry about credit card security. S. Smith Testimony, 11/15 Tr. 125:6-128:13.
180. In a study published in July 2003 and in an updated study published in Summer 2006 (after he submitted his expert report), defendant's expert, Dr. Smith, found that consumers have serious concerns about using credit cards on the Internet, in part, because of fear of theft or fraud. S. Smith Testimony, 11/15 Tr. 151:1-154:25, 158:24-160:13.
181. These two studies, along with many other studies discussed above, contradict Dr. Smith's opinion in this case that Internet users would not be significantly affected by having to provide credit card information to access free content on Web sites were COPA to go into effect. Moreover, Dr. Smith did not cite any evidence as to why his opinion had changed. S. Smith Testimony, 11/15 Tr. 118:7-20, 166:11-167:3. Therefore, I find Dr. Smith's testimony on this topic at trial to be unreliable.
J. Geolocation
182. Quova, Inc. provides IP intelligence services that allow online businesses to attempt to determine the location of users of their Web sites. Alexander Testimony, 11/13 Tr. 5:5-9.
183. A product that Quova markets can determine, within a 20 to 30 mile radius, the location from which a user is accessing a Web site through a proxy server, satellite connection, or large corporate proxy. Alexander Testimony, 11/13 Tr. 19:10-19, 28:2-7. The fact that Quova can only narrow down a user's location to a 20 to 30 mile radius results in Quova being unable to determine with 100 percent accuracy which side of a city or state border a user lives on if the user lives close to city or state borders. Alexander Testimony, 11/13 Tr. 28:8-16.
184. If a visitor is accessing a Web site through AOL, Quova can only determine whether the person is on the East or West coast of the United States. Alexander Testimony, 11/13 Tr. 20:20-21:3.
185. Quova has been used by Web site operators to direct traffic so that only users in the United States can view products that can only be distributed in the United States and to customize content for users in the United States as opposed to users in a another country. Alexander Testimony, 11/13 Tr. 22:11-24:11.
186. The services Quova offers can cost anywhere from $6,000 to $500,000 a year. Alexander Testimony, 11/13 Tr. 24:12-20.
IV. CONCLUSIONS OF LAW
To the extent that the following Conclusions of Law include Findings of Fact or mixed Findings of Fact and Conclusions of Law, those Findings and Conclusions are hereby adopted by this court.
A. Standing
1. In order to maintain standing, the plaintiffs must show, inter alia, that they have sustained or are immediately in danger of sustaining some direct injury that is not abstract, conjectural or hypothetical. City of Los Angeles v. Lyons, 461 U.S. 95, 102, 103 S.Ct. 1660, 75 L.Ed.2d 675 (1983). In a pre-enforcement challenge to a statute carrying criminal penalties, standing exists when "the plaintiff has alleged an intention to engage in a course of conduct arguably affected with a constitutional interest, but proscribed by a statute, and there exists a credible threat of prosecution." Babbitt v. United Farm Workers Nat. Union, 442 U.S. 289, 298 (1979); Virginia v. Am. Booksellers Ass'n, Inc., 484 U.S. 383, 392-93 (1988).
2. Because all of the plaintiffs in this case seek the same injunctive relief, if the court concludes that one of the plaintiffs has standing, it is unnecessary to determine the standing of the remaining plaintiffs. Dep't of Commerce v. U.S. House of Representatives, 525 U.S. 316, 330 (1999) (finding that the presence of one party with standing assures that controversy before the Court is justiciable); see Sec. of the Interior v. California, 464 U.S. 312, 319 n. 3 (1984); Babbitt, 442 U.S. at 299, n. 11.
3. As this court has found previously and today, and as the Third Circuit observed, there is nothing in the language of COPA which limits its reach to commercial pornographers. ACLU, 322 F.3d at 256 (citing 31 F. Supp.2d at 480); 47 U.S.C. § 231.
Like Congress and both parties, I will not attempt to define "commercial pornographer." Such a definition is not necessary to this adjudication, however, as: (1) COPA clearly covers far more speakers on the Web than those who might be defined as commercial pornographers; and (2) defendant contends and, thus, admits that the plaintiffs are not commercial pornographers. see Conclusions of Law 3, 4, 9, 39, 41, 52; Doc. No. 429, Def.'s Proposed Finding of Fact 11.
4. I conclude that under COPA, the terms "commercial purposes" and "engaged in the business" are defined very broadly and include within their reach Web sites which only receive revenue from advertising or which generate profit for their owners only indirectly. 47 U.S.C. § 231(e)(2)(A) (B). Again this conclusion is in accord with an insight by the Third Circuit in ruling on the efficacy of the 1999 preliminary injunction. ACLU, 322 F.3d at 256 (observing that these definitions expand "COPA's reach beyond those enterprises that sell services or goods to consumers" to include "those persons who sell advertising space on their otherwise noncommercial Web sites").
For example, Barbara DeGenevieve ("DeGenevieve"), a tenured professor at the School of the Art Institute in Chicago, uses her Web site to generate profit even though all of her work on her Web site is free to view, she does not sell anything directly on her Web site, and she does not provide advertising space. Instead, DeGenevieve's Web site gives others the opportunity to contact her in order to purchase copies of her work or engage her to give lectures or workshops, which is a common way to do business in her academic field. DeGenevieve Testimony, 11/1 Tr. 25:18-27:9, 31:14-24, 42:22-43:3, 61:22-66:7.
5. Based upon Findings of Fact 16 through 18, 21, 26, and 41 though 58, Corinna, Nerve, and Salon engage in communications on the Web for commercial purposes and have content which an average person utilizing contemporary community standards could find was designed to appeal to or pander to the prurient interest of minors and depicts, describes, or represents, in a manner patently offensive with respect to minors, an actual or simulated sexual act or sexual contact, an actual or simulated normal or perverted sexual act, or a lewd exhibition of the genitals or post-pubescent female breast. See 47 U.S.C. § 231(e)(6). The same content of these three plaintiffs could also be found by a reasonable person to include content that lacks serious literary, artistic, political, or scientific value for minors. See 47 U.S.C. § 231(e)(6). Therefore, this case is justiciable as a credible threat of prosecution exists and at least these three plaintiffs have standing.See Babbitt, 442 U.S. at 298. B. Strict Scrutiny Applies to this Action
As stated previously, under COPA, the relevant inquiry relies on whether an average person, utilizing community standards would find that the material was: (1) as a whole, designed to appeal to the prurient interest of minors; (2) depicts sexual acts or contact or certain proscribed nudity that is patently offensive to minors; and (3) whether a reasonable person would find that the content as a whole lacked certain types of value for minors. 47 U.S.C. § 231(e)(6); Pope v. Illinois, 481 U.S. 497, 500-501 (1987) (stating that under the Miller test for obscenity, in determining whether a given work has value, the proper inquiry is "whether a reasonable person would find such value in the material, taken as a whole"). As a result, because the plaintiffs' subjective beliefs are unimportant to the statutory analysis, it does not matter whether the plaintiffs themselves believe that their content is harmful to minors as such beliefs are irrelevant.
6. There is no doubt that COPA restricts speech based upon its content. See 47 U.S.C. § 231; Ashcroft, 542 U.S. at 660. Therefore strict scrutiny applies to the claims in this action. Turner Broadcasting Sys., Inc. v. F.C.C., 512 U.S. 622, 642 (1994). Such regulations are presumptively invalid. R.A.V. v. City of St. Paul, 505 U.S. 377, 382 (1992).
Defendant's contention that COPA regulates only commercial speech and, thus, should be analyzed under the less exacting standard for such speech is utterly meritless. If accepting advertising or selling subscriptions transformed speech into commercial speech, the First Amendment protections afforded to many modes of communication, including print, would be completely destroyed. See e.g. Central Hudson Gas Elec. Corp. v. Public Serv. Comm'n of New York, 447 U.S. 557, 561 (1980) (defining commercial speech as "expression related solely to the economic interests of the speaker and its audience"); Friedman v. Rogers, 440 U.S. 1, 10 (1979) (stating that commercial speech "relates to a particular product or service"); Virginia State Bd. of Pharmacy v. Virginia Citizens Consumer Council, Inc., 425 U.S. 748, 761-762 (1976) (recognizing that speech receives full First Amendment protection "even though it is carried in a form that is sold for profit" or "even though it may involve a solicitation to purchase or otherwise pay or contribute money" and defining commercial speech as speech that does "no more than propose a commercial transaction") (internal quotation marks omitted).
7. Because COPA suppresses a large amount of speech that adults have a constitutional right to receive, under the strict scrutiny standard, COPA may only be upheld as constitutional if defendant meets his burden of proving that COPA is narrowly tailored to the compelling interest that COPA was enacted to serve and there are no less restrictive alternatives that would be at least as effective in achieving that interest. Ashcroft, 542 U.S. at 665 (citing Reno, 521 U.S. at 874).
8. There has never been any question that the interest espoused by Congress, as related to this court by defendant of "protecting minors from exposure to sexually explicit material on the World Wide Web" is a compelling interest. Reno, 521 U.S. at 869-870 (stating that "`there is a compelling interest in protecting the physical and psychological well-being of minors' which extended to shielding them from indecent messages that are not obscene by adult standards") (quoting Sable, 492 US. at 126, 109 S.Ct. 2829). C. Defendant Has Failed to Meet His Burden of Proof
1. Defendant Has Failed to Show that COPA Is Narrowly Tailored to Congress' Compelling Interest
a. COPA Is Overinclusive
9. COPA is overinclusive. Simon Schuster, Inc. v. Members of New York State Crime Victims Bd., 502 U.S. 105, 121 (1991) (finding that laws which are "significantly overinclusive" are not narrowly tailored). Due to the broad definitions and provisions of COPA, COPA prohibits much more speech than is necessary to further Congress' compelling interest. 47 U.S.C. § 231. For example, as discussed above in Conclusions of Law 3 and 4, the definitions of "commercial purposes" and "engaged in the business" apply to an inordinate amount of Internet speech and certainly cover more than just commercial pornographers, contrary to the claim of defendant. Moreover, as discussed below in Conclusions of Law 42 and 49, the fact that COPA applies to speech that is obscene as to all minors from newborns to age sixteen, and not just to speech that is obscene as to older minors, also renders COPA over-inclusive. See ACLU, 322 F.3d at 253-254 (noting that the term "minor" in COPA applies to "an infant, a five-year old, or a person just shy of age seventeen").
b. COPA Is Underinclusive
10. COPA is also underinclusive. See Arkansas Writers' Project, Inc. v. Ragland, 481 U.S. 221, 232 (1987) (finding the statutory provision at issue unconstitutional because it was "both overinclusive and underinclusive"). For example, as shown in Findings of Fact 63, 65, and 66, there is a significant amount of sexually explicit material on the Internet which originates from outside of the United States. As discussed below, unlike Internet content filters which are able to block from view unsuitable material regardless of its origin (see Finding of Fact 80), COPA has no extra-territorial application. See also Finding of Fact 131 and Pl. Ex. 55, at 3 (the Department of Justice concluding that COPA "apparently would not attempt to address", inter alia, sexually explicit overseas Web sites and that "any attempt to regulate overseas web sites would raise difficult questions regarding extraterritorial enforcement"); Pl. Ex. 54, at 235 (The NRC concluding that because a substantial percentage of sexually explicit Web sites exist outside the United States, even strict enforcement of COPA will likely have only a marginal effect on the availability of such material on the Internet in the United States). As a result, and as is more fully discussed below, COPA is not applicable to a large amount of material that is unsuitable for children which originates overseas but is nevertheless available to children in the United States.
11. Statutes are presumed to only have domestic effect unless a contrary intent appears. Foley Bros., Inc. v. Filardo, 336 U.S. 281, 285 (1949). Although, COPA claims to prohibit the transmission of material that is harmful to minors "in interstate or foreign commerce", 47 U.S.C. § 231(a)(1), such language is legally insufficient to demonstrate Congress' clear intent that COPA applies to Web sites hosted or registered outside of the United States. See E.E.O.C. v. Arabian American Oil Co., 499 U.S. 244, 251 (1991) (superceded by statute on the issue of retroactive application) (finding that the Supreme Court has "repeatedly held that even statutes that contain broad language in their definitions of `commerce' that expressly refer to `foreign commerce' do not apply abroad") (citing New York Cent. R. Co. v. Chisholm, 268 U.S. 29, 45, 69 L.Ed. 828 (1925) and McCulloch v. Sociedad Nacional de Marineros de Honduras, 372 U.S. 10, 83 S.Ct. 671, 9 L.Ed.2d 547 (1963));U.S. v. Reeves, 62 M.J. 88 (C.A.A.F. 2005) (applying the presumption against extraterritoriality to a statute that prohibited conduct "in interstate or foreign commerce").
12. Defendant contends that there is precedent for reading into a criminal statute extraterritorial application where a statute does not expressly provide for such. See U.S. v. Harvey, 2 F.3d 1318 (3d Cir. 1993). However, because COPA has both civil and criminal penalties, and because of the lack of Congressional intent, the court will not read into COPA implied extraterritorial application. Smith v. U.S., 507 U.S. 197, 203-204 (1993) (stating that "[i]t is a longstanding principle of American law that legislation of Congress, unless a contrary intent appears, is meant to apply only within the territorial jurisdiction of the United States") (internal quotation marks omitted).
13. As shown in Finding of Fact 130, the legislative history of COPA does not support a finding that Congress intended for COPA to apply to Web sites that are hosted or registered outside of the United States and instead shows that Congress intended for the statute to have only domestic application. See also H.R. Rep. 105-775, at *20.
14. If Congress had intended for COPA to have extraterritorial application, it could have inserted appropriate language in the statute. Argentine Republic v. Amerada Hess Shipping Corp., 488 U.S. 428, 440 (1989) (remarking that "Congress knows how to place the high seas within the jurisdictional reach of a statute").
15. Enforcement of COPA against overseas Web site owners would also be burdensome and impractical due to the knotty questions of jurisdiction which arise in the Internet context. See e.g. Zippo Mfg. Co. v. Zippo Dot Com, Inc., 952 F. Supp. 1119 (W.D. Pa. 1997). Furthermore, even if a specific foreign Web site had sufficient contacts with the forum to allow personal jurisdiction, it could be quite difficult or impossible to ensure that the offender would obey or could be forced to obey the judgment of the U.S. court. As a result, COPA's lack of extraterritorial application renders it underinclusive.
16. As demonstrated in Finding of Fact 126, it is also accurate that COPA only concerns itself with material that is accessible over the Internet using HTTP or a successor protocol. See also 47 U.S.C. § 231(e)(1). As advocated by the plaintiffs, it appears that this is yet another reason why COPA is underinclusive. However, as discussed below in Conclusions of Law 35 and 36, the compelling interest of Congress, as submitted by defendant, is quite narrow in that it seeks only to protect children form harmful materials on the Web. Therefore, I will restrict my analysis to materials available on the Web and will not consider the ramifications of COPA's failure to reach other harmful materials on the Internet accessible by means other than the Web.
c. The Affirmative Defenses in COPA Do Not Aid in Narrowly Tailoring It to Congress' Compelling Interest
17. The affirmative defenses cannot cure COPA's failure to be narrowly tailored because they are effectively unavailable. Credit cards, debit accounts, adult access codes, and adult personal identification numbers do not in fact verify age. As a result, their use does not, "in good faith," "restrict access" by minors. 47 U.S.C. § 231(c)(1)(A); See Findings of Fact 137, 138, 140-147, 148-158, 159-160.
18. As shown by Findings of Fact 138 and 140 through 147, because payment card issuers prohibit the use of their credit or debit cards to verify age and because a significant number of minors have access to payment cards, such cards are not an effective age verification device.
19. The Commerce Committee's reliance on Sable was misplaced as the Court in Sable neither approved nor disapproved of credit card age verification and the other defenses listed in the FCC dial-a-porn regulations for 47 U.S.C. §§ 223(b)-(c). Findings of Fact 135, 136. Moreover, the evidence presented in this case is contrary to the findings of the Second Circuit in Dial Information Services, 938 F.2d 1535, and, thus, I do not find that case persuasive. See Conclusion of Law 18. Further distinguishing this case from Dial Infomation Services is that 47 U.S.C. § 223(b) was designed to specifically combat commercial dial-a-porn, see Sable 492 U.S. at 120, while the range of speech restricted in COPA is much wider and effects even those non-commercial Web sites which garner profit for their owners only through advertising or other indirect means. See Conclusions of Law 3, 4. Instead, I agree with the Fourth Circuit in PSINet, Inc. v. Chapman, 362 F.3d 227 (4th Cir. 2004). In PSInet, the Fourth Circuit found the Virginia Internet harmful-to-minors law overly broad and not narrowly tailored. 362 F.3d at 239. In its analysis the court noted, in regards to the government's suggestion that the use of PIN numbers for access to Web pages would provide an adequate affirmative defense, that "the Commonwealth would certainly not agree that a liquor or tobacco store that sold to anyone with a valid credit card number, without some additional step to ascertain the age of the customer, was taking reasonable steps to exclude juveniles from the purchase of age prohibitive products." Id. at 236. The Fourth Circuit also found that since some adults do not have credit cards and some would be unwilling to provide a credit card number online, requiring a credit card to access a site would "unduly burden protected speech in violation of the First Amendment" and that "requiring adult Web sites to utilize PIN numbers would unconstitutionally chill free speech." Id. at 236-37. I find the reasoning of the Fourth Circuit persuasive.
20. As demonstrated by Findings of Fact 138 and 148 through 158, because there are no DVS products that actually verify age but merely verify data, sometimes with as little as a name and address, DVS products are not effective age verification services.
21. The affirmative defenses also raise their own First Amendment concerns. For example, the utilization of those devices to trigger COPA's affirmative defenses will deter listeners, many of whom will be unwilling to reveal personal and financial information in order to access content and, thus, will chill speech. See Denver Area Educ. Telecomms. Consortium, Inc. v FCC, 518 U.S. 727, 754 (1996) (striking down an identification requirement because it would "further restrict viewing by subscribers who fear for their reputations should the operator, advertently or inadvertently, disclose the list of those who wish to watch the `patently offensive' channel"); Findings of Fact 171, 172-181.
22. Similarly, the affirmative defenses also impermissibly burden Web site operators with demonstrating that their speech is lawful. See Ashcroft v. Free Speech Coalition, 535 U.S. 234, 255 (2002) (stating that "The Government raises serious constitutional difficulties by seeking to impose on the defendant the burden of proving his speech is not unlawful"); ACLU, 322 F.3d at 260 (noting that "the affirmative defenses [in COPA] do not provide the Web publishers with assurances of freedom from prosecution"). Under the COPA regime, Web site operators are unable to defend themselves until after they are prosecuted. See 47 U.S.C. § 231(c).
23. Moreover, the affirmative defenses place substantial economic burdens on the exercise of protected speech because all of them involve significant cost and the loss of Web site visitors, especially to those plaintiffs who provide their content for free. See Simon Schuster, Inc., 502 U.S. at 115 (stating that "A statute is presumptively inconsistent with the First Amendment if it imposes a financial burden on speakers because of the content of their speech"); Findings of Fact 161-163, 165-171. Defendant's response to this proposition, that the defenses are not burdensome to commercial pornographers because they already accept credit cards to sell their content, shows his fundamental misunderstanding of the reach of COPA: COPA does not apply merely to commercial pornographers but to a wide range of speakers on the Web. See Conclusions of Law 3, 4.
24. Based upon Finding of Fact 164, Bitpass is not a reasonable solution to the problem faced by Web site owners who provide their content for free but who cannot accept zero-dollar transactions on credit cards.
25. As shown above, the affirmative defenses in COPA raise unique First Amendment issues and, in any event, do not aid in narrowly tailoring COPA to Congress' compelling interest.
2. Defendant Has Failed to Show that COPA Is the Least Restrictive Alternative for Advancing Congress' Compelling Interest
26. Defendant has failed to successfully defend against the plaintiffs' assertion that filter software and the Government's promotion and support thereof is a less restrictive alternative to COPA. The Supreme Court recognized, upon the evidence before it at the time of the issuance of the preliminary injunction in 1999, that:
Filters are less restrictive than COPA. They impose selective restrictions on speech at the receiving end, not universal restrictions at the source. Under a filtering regime, adults without children may gain access to speech they have a right to see without having to identify themselves or provide their credit card information. Even adults with children may obtain access to the same speech on the same terms simply by turning off the filter on their home computers. Above all, promoting the use of filters does not condemn as criminal any category of speech, and so the potential chilling effect is eliminated, or at least much diminished. All of these things are true, moreover, regardless of how broadly or narrowly the definitions in COPA are construed.Ashcroft, 542 U.S. at 667. The evidence at trial shows that this is still the case today. It remains true, for example, that unlike COPA there are no fines or prison sentences associated with filters which would chill speech. 47 U.S.C. § 231(a). Also unlike COPA, as shown by Findings of Fact 68, 78, and 79, filters are fully customizable and may be set for different ages and for different categorizes of speech or may be disabled altogether for adult use. As a result, filters are less restrictive than COPA.
27. Moreover, defendant contends that: (1) filters currently exist and, thus, cannot be considered a less restrictive alternative to COPA; and that (2) the private use of filters cannot be deemed a less restrictive alternative to COPA because it is not an alternative which the government can implement. These contentions have been squarely rejected by the Supreme Court in ruling upon the efficacy of the 1999 preliminary injunction by this court. The Supreme Court wrote:
Congress undoubtedly may act to encourage the use of filters. We have held that Congress can give strong incentives to schools and libraries to use them. It could also take steps to promote their development by industry, and their use by parents. It is incorrect, for that reason, to say that filters are part of the current regulatory status quo. The need for parental cooperation does not automatically disqualify a proposed less restrictive alternative. In enacting COPA, Congress said its goal was to prevent the "widespread availability of the Internet" from providing "opportunities for minors to access materials through the World Wide Web in a manner that can frustrate parental supervision or control." COPA presumes that parents lack the ability, not the will, to monitor what their children see. By enacting programs to promote use of filtering software, Congress could give parents that ability without subjecting protected speech to severe penalties.Ashcroft, 542 U.S. at 669-670 (internal citation omitted). I agree with the Supreme Court and conclude today that the mere fact that filters currently exist does not indicate that they cannot be a less restrictive or effective alternative to COPA nor does it make them part of the regulatory status quo. I also agree and conclude that in conjunction with the private use of filters, the government may promote and support their use by, for example, providing further education and training programs to parents and caregivers, giving incentives or mandates to ISP's to provide filters to their subscribers, directing the developers of computer operating systems to provide filters and parental controls as a part of their products (Microsoft's new operating system, Vista, now provides such features, see Finding of Fact 91), subsidizing the purchase of filters for those who cannot afford them, and by performing further studies and recommendations regarding filters.
3. Defendant Has Failed to Show that Other Alternatives Are Not at Least as Effective as COPA
28. Defendant has also failed to show that filters are not at least as effective as COPA at protecting minors from harmful material on the Web. The first hurdle in this analysis is that there is no showing of how effective COPA will be. However, the evidence shows that at a minimum, COPA will not reach a substantial amount of foreign source sexually explicit materials on the Web, which filters will reach. Findings of Fact 63, 65, 66 and Conclusions of Law 10-14.
29. COPA will also not be effective because its affirmative defenses including the age verification schemes are not effective. See Conclusions of Law 17-25.
30. Moreover, based on the recent sparse enforcement history of the obscenity laws detailed in Findings of Fact 132 and 133 and the concern of the Department of Justice that COPA "could require an undesirable diversion of critical investigative and prosecutorial resources that the Department currently invests in . . . prosecuting . . . large-scale and multi district commercial distributors of obscene materials", Pl. Ex. 55; Finding of Fact 131, it is unlikely that COPA will be widely enforced, thus further limiting its effectiveness.
31. The Supreme Court recognized that, on the record before them at the time, filters were "likely more effective as a means of restricting children's access to materials harmful to them."Ashcroft, 542 U.S. at 667. I conclude that the evidence in this case now confirms the Supreme Court's prediction.
32. Although filters are not perfect and are prone to some over and under blocking, the evidence shows that they are at least as effective, and in fact, are more effective than COPA in furthering Congress' stated goal for a variety of reasons. For example, as shown by Findings of Fact 68, 78 through 80, 87 through 91, and 92 through 99, filters block sexually explicit foreign material on the Web, parents can customize filter settings depending on the ages of their children and what type of content they find objectionable, and filters are fairly easy to install and use. See also Findings of Fact 102-109.
33. Reliable studies also show that filters are very effective at blocking potentially harmful sexually explicit materials. Findings of Fact 110-116.
34. Even defendant's own study shows that all but the worst performing filters are far more effective than COPA would be at protecting children from sexually explicit material on the Web, garnering percentages as high as nearly 99 percent in successfully blocking such material. Findings of Fact 117-121. As a result of Conclusions of law 28 through 34, it is clear that defendant has failed to establish that COPA is the least restrictive means of protecting children from harmful sexually explicit materials on the Web.
35. As discussed briefly above in Conclusion of Law 16, there are also several other factors which technically make filters more effective than COPA at protecting children from harmful materials on the Internet. For example, filters cover many formats other than HTTP and filters can block content from a host of Internet applications. Findings of Fact 76, 77, 82. COPA specifically does not cover any of these items as it is limited only to materials found on the Web via HTTP or a successor protocol. Findings of Fact 126-127. The plaintiffs allege that as a result, COPA ignores a substantial amount of Internet speech which could be deemed harmful to minors. This contention is technically correct, however for reasons stated below, it was not considered in my analysis. Defendant claims that COPA's limitations were added to the statute as a direct result of the Supreme Court's prior holding that the CDA was overbroad in part because it was not limited to commercial speech and it covered a wide array of Internet speech other than simply that found on the Web. Defendant's contention is also correct. Findings of Fact 125-129. This creates an interesting conundrum for defendant where broadening COPA's reach would likely make it overbroad, but by narrowing its reach to HTTP or successor protocols, Congress has made COPA underinclusive and less effective than filters.
The plaintiffs likewise contend that COPA is underinclusive and less effective than filters because it is limited to commercial speech. However, as shown above in Conclusions of Law 3 and 4, COPA is not actually limited to commercial speech as its reach is sufficiently broad to cover virtually every Web site which has sexually explicit materials. I conclude that the commercial restriction in COPA is nearly a nullity and, thus, no restriction at all. As a result, I will not consider the plaintiffs' argument regarding COPA's commercial limitations.
36. However, the compelling interest of Congress, as described to this court by defendant, is very narrow: to protect children from sexually explicit material on the Web. Although Congress' stated interest is listed more generally in the Congressional Findings as "the protection of the physical and psychological well-being of minors by shielding them from materials that are harmful to them", it is clear from the rest of the Findings that harmful material on the Web was Congress' true concern. Finding of Fact 124; H.R. Rep. 105-775, at *2 (discussing minors access to materials on the Web that can frustrate parental supervision or control and stating that past efforts have not provided a national solution to the problem of minors accessing harmful material on the Web). Therefore, I accept defendant's narrow interpretation of the compelling interest of Congress. I also recognize that Congress may legislate in stages and that "reform may take one step at a time, addressing itself to the phase of the problem that seem most acute to the legislative mind."McConnell v. Fed. Election Comm'n, 540 U.S. 93, 207-08 (2003). For this reason, and because it is clear that COPA is less effective than filters even without this additional contention, I have restricted my analysis to harmful materials available on the Web (and not beyond), which makes these other factors irrelevant to my decision.
D. Vagueness and Overbreadth
37. Facial challenges to legislation are "employed by the Court sparingly and only as a last resort." Nat'l Endowment for the Arts v. Finley, 524 U.S. 569, 580 (1998) (quoting Broadrick v. Oklahoma, 413 U.S. 601, 613 (1974)). To prevail on their facial challenge to COPA, the plaintiffs must "demonstrate a substantial risk that application of the provision will lead to suppression of speech." Id.
Although facial challenges should be avoided and the holding here is merely supplemental, addressing vagueness and overbreadth in regards to COPA is very important to the facilitation of any subsequent legislation in this area, since Congress clearly relied on the Supreme Court's vagueness and overbreadth analyses of the CDA in drafting COPA. See 535 U.S. at 569-70; ACLU, 322 F.3d at 244-45.
1. COPA Is Vague
38. The vagueness doctrine was created to ensure fair notice and nondiscriminatory application of the laws. U.S. v. Tykarsky, 446 F.3d 458, 472 n. 9 (3d Cir. 2006). A statute or regulation fails for vagueness if men of ordinary intelligence must speculate as to the meaning of what the statute or regulation requires or prohibits. See Gibson v. Mayor Council of Wilmington, 355 F.3d 215, 225 (3d Cir. 2004).
39. A party cannot bring a facial vagueness claim if the challenged regulation clearly applies to that party's speech.Gibson, 355 F.3d at 225; Rode v. Dellarciprete, 845 F.2d 1195, 1200 (3d Cir. 1988). Here, Congress intended COPA to apply only to commercial pornographers as demonstrated by Findings of Fact 122, 123, 128, and 129 and Conclusions of Law 3 and 4. However, the plaintiffs in this action are not commercial pornographers, a fact which has not escaped defendant's notice in his challenges to their standing. Therefore, because COPA does not clearly apply to the plaintiffs' speech, the plaintiffs may bring a facial vagueness claim.
40. "[W]here Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposefully in the disparate inclusion and exclusion."Russollo v. U.S., 464 U.S. 16, 23 (1983) (quoting U.S. v. Wong Kim Bo, 472 F.2d 720, 722 (5th Cir. 1972)). In contrast to the scienter requirement of "knowingly and with knowledge of the character of the material" found in 47 U.S.C. § 231(a)(1), the term "intentionally" is used in 47 U.S.C. § 231(a)(2). This court must assume that Congress intended the disparate use of "knowingly and with knowledge of the character of the material" and "intentionally" in 47 U.S.C. § 231(a)(1)-(2). However, since neither term is defined, the difference in scienter standards creates uncertainty in COPA's application and renders the terms vague. The chilling effect created by this uncertainty is exacerbated by the fact that violations of both of the standards could result in criminal proceedings and violations of the intentional standard could result in a fine of up to $50,000 a day. See 47 U.S.C. § 231(a)(1)-(2).
The court notes that the Department of Justice contended in a 1998 letter that the lack of clarity created by the use of these two terms was one of ten "confusing or troubling ambiguities" in COPA. Pl. Ex. 55; Finding of Fact 131.
41. Although Findings of Fact 122, 123, 128, and 129 demonstrate that Congress intended COPA to apply solely to commercial pornographers, as discussed previously in Conclusions of Law 3 and 4, the phrase "communication for commercial purposes", as it is modified by the phrase "engaged in the business", does not limit COPA's application to commercial pornographers. See ACLU, 322 F.3d at 256 (citing 31 F. Supp.2d at 480). The lack of clarity in these phrases results in Web sites, which only receive revenue from advertising or which generate profit for their owners only indirectly, from being included in COPA's reach. Conclusion of Law 4. Since the enforcement of COPA could result in prosecution of the plaintiffs in this case, it is reasonable for the plaintiffs to fear prosecution under COPA. See e.g. Findings of Fact 14, 41, 46-61; Conclusion of Law 5. The uncertainty resultant from the vagueness of "communication for commercial purposes" would cause the plaintiffs' speech to be chilled or self censored, as demonstrated by Findings of Fact 50, 54, and 57 through 61.
42. As noted above, COPA defines a minor as "any person under 17 years of age." 47 U.S.C. § 231(e)(7). Although the government argues that the term minor should be interpreted to mean an older minor, as the Third Circuit noted, such an interpretation would be "in complete disregard of the text" of COPA. ACLU, 322 F.3d at 253. As discussed by the Third Circuit, defining minors as "any person under 17 years of age," creates a serious issue with interpretation of COPA since no one could argue that materials that have "serious literary, artistic, political, or scientific value" for a sixteen-year-old would necessarily have the same value for a three-year old. Id., at 253-54. Likewise, what would be "patently offensive" to an eight-year-old would logically encompass a broader spectrum of what is available on the Web than what would be considered "patently offensive" for a sixteen-year-old. Presumably no material covered by COPA would be "designed to appeal to, or [be] designed to pander to, the prurient interest" of a two or four-year old. Id., at 254. As the Third Circuit noted, "[i]n abiding by this definition, Web publishers who seek to determine whether their Web sites will run afoul of COPA cannot tell which of these `minors' should be considered in deciding the particular content of their Internet postings."Id. Thus, the application of the definition of minors to COPA creates vagueness in the statute.
As the Third Circuit found, the cases cited by defendant in support of his argument are inapposite. ACLU, 322 F.3d at 254 n. 16. I agree. The Eleventh Circuit's interpretation in American Booksellers v. Webb of the Supreme Court's use of the "reasonable person" standard in Pope, 481 U.S. at 500-01 does not logically follow the holding in Pope, since nothing in Pope implies that where a reasonable minor standard is employed, that the reasonable minor should be considered an older reasonable minor.See ACLU, 322 F.3d at 254 n. 16; Am. Booksellers v. Webb, 919 F.2d 1493, 1508-09 (11th Cir. 1990). Additionally, the decisions of the Tennessee and Virginia Supreme Courts to apply an older minor standard to state statutes for minors are not binding on this Court. See ACLU, 322 F.3d at 254 n. 16; Davis-Kidd Booksellers, Inc. v. McWhereter, 866 S.W.2d 520, 527 (Tenn. 1993); Am. Booksellers Ass'n v. Virginia, 882 F.2d 125, 127 (4th Cir. 1989).
43. I recognize that state laws prohibiting obscene-as-to-minors material have been upheld by the Supreme Court with no discussion of whether they apply to minors of every age or only to older minors. See e.g. Ginsberg v. State of N. Y., 390 U.S. 629 (1968). In a store, the cashier can easily discern if a patron is 10 or 11 years old instead of a claimed age of 17. The same cashier, however, may have trouble discerning whether a patron is 16 years old or 17 years old without some form of photographic identification. Thus, with laws that are concerned with face-to-face transactions, in reality it is only those borderline cases in which a minor is close to the age of majority that are at issue. See e.g. Id. at 631 (wherein the patron at issue was 16 years old). As a result, in the pre-Internet age, it was not completely necessary to be more specific in delineating what was obscene as to minors of various age groups. However, on the Internet, everyone is faceless and fairly anonymous and, thus, the context is radically changed. The Internet merchant has no viable method of determining whether an individual is 6, 12, 17 or 51 years old. Consequently, we are presented with this novel problem where a general prohibition on materials obscene as to minors creates a vagueness in the context of Internet transactions that is lacking in other situations.
44. COPA does not define the term "as a whole" and the plain language of the statute does not lend itself to a obvious definition of "as a whole" as it might be applied to the Internet. 47 U.S.C. § 231. The Third Circuit concluded in a dictum that the language of COPA clearly demonstrated that each individual "communication, picture, image, graphic image file, article, recording, writing or other matter of any kind" should be considered without context. ACLU, 322 F.3d at 252. But, as Justice Breyer noted in his dissent, "as a whole" has been traditionally interpreted in obscenity cases to require an examination of the challenged material within the context of the book or magazine in which it is contained. Ashcroft, 542 U.S. at 681 (citing Roth v. U.S., 354 U.S. 476, 490 (1957)). As Justice Kennedy noted in his concurring opinion, "The notion of judging work as a whole is familiar in other media, but more difficult to define on the World Wide Web. It is unclear whether what is to be judged as a whole is a single image on a Web page, a whole Web page, an entire multipage Web site, or an interlocking set of Web sites." 535 U.S. at 592-93. Thus, with the disparate views noted above, and as discussed below, in the context of the Web, I conclude that the use in COPA of the phrase "as a whole" without any further definition, is vague.
45. There is no question that a printed book or magazine is finite, and, as a result, it is very easy to discern what needs to be examined in order to make an "as a whole" evaluation. The same is not true for a Web page or Web site since Web pages and sites are hyperlinked to other Web pages and sites. As demonstrated by online magazines such as Nerve and Salon, even the Web sites for online magazines, without considering the hyperlinks to off-site materials, have greater depth and breadth than their counterpart in print. See Findings of Fact 3, 21, 26. Instead of having a two-hundred page book or an issue of a magazine to look to for context, COPA invokes some undefined portion of the vast expanse of the Web to provide context for material allegedly violating the statute. As a result, a Web publisher cannot determine what could be considered context by a fact finder, prosecutor, or court, and therein lies the source of the vagueness.
46. The vagueness of COPA, like the vagueness of CDA, is especially concerning since they are both content-based regulations. See Reno, 521 U.S. at 871-72. "An impermissible chill is created when one is deterred from engaging in protected activity by the existence of a governmental regulation or the threat of prosecution thereunder." Aiello v. City of Wilmington, 623 F.2d 845, 857 (3d Cir. 1980). The plaintiffs have demonstrated, as recounted in Factual Findings 50, 54, and 57 through 61, that COPA has a chilling effect on free speech. The fact that Web publishers are faced with criminal prosecution for an alleged violation of COPA only serves to exacerbate the chilling effect resultant from the vagueness of the terms employed in COPA. See Reno, 521 U.S. at 872. Thus, I conclude that COPA is clearly unconstitutionally vague.
2. COPA Is Overbroad
47. "The overbreadth doctrine prohibits the Government from banning unprotected speech if a substantial amount of protected speech is prohibited or chilled in the process." Free Speech Coalition, 535 U.S. at 255. The Supreme Court has noted that, "the possible harm to society in permitting some unprotected speech to go unpunished is outweighed by the possibility that protected speech of others may be muted." Broadrick, 413 U.S. at 612.
48. Since the vagueness of "communication for commercial purposes" and "engaged in business" would allow prosecutors to use COPA against not only Web publishers with commercial Web sites who seek profit as their primary objective but also those Web publishers who receive revenue through advertising or indirectly in some other manner, the array of Web sites to which COPA could be applied is quite extensive. Such a widespread application of COPA would prohibit and undoubtably chill a substantial amount of constitutionally protected speech for adults.
49. As discussed above, because a story that might have "serious literary value" for a sixteen-year-old could be considered to appeal to the "prurient interest" of an eight-year-old and be "patently offensive" and without "serious value" to that child, Web publishers do not have fair notice regarding what they can place on the Web that will not be considered harmful to "any person under 17 years of age." As the Third Circuit stated in ruling on the efficacy of the 1999 preliminary injunction by this court, "[b]ecause COPA's definition of `minor' therefore broadens the reach of `material that is harmful to minors' under the statute to encompass a vast array of speech that is clearly protected for adults . . . the definition renders COPA significantly overinclusive." ACLU, 322 F.3d at 268.
50. As demonstrated by Conclusions of Law 17 through 25 and as noted by the Third Circuit, the affirmative defenses in COPA do not prevent it from sweeping too broadly since they do not verify age, impose additional burdens, and add to the statute's chilling effect. See ACLU, 322 F.3d at 268.
51. "When a federal court is dealing with a federal statute challenged as being overly broad, it should, of course, construe the statute to avoid constitutional problems, if the statute is subject to such a limiting construction." New York v. Ferber, 458 U.S. 747, 769 n. 24, 102 S.Ct. 3348, 73 L.Ed.2d 1113 (1982). If the statute is not subject to a limiting construction but can be severed so that a constitutional part remains, the statute should be severed accordingly. Id. However, a law should not be rewritten by a court so that it can pass constitutional muster.Am. Booksellers Ass'n, 484 U.S. at 397.
52. Nothing in the statute references commercial pornographers, for whom the statute was apparently intended, as demonstrated by Findings of Fact 122, 123, 128, and 129 and Conclusions of Law 3 and 4. To read such a limitation into the statute would result in an impermissible rewriting of the statute and assumption of the role of the legislature by this court. The term "minor" is clearly not subject to a narrowing construction, because, as noted by the Third Circuit, acting as if COPA only applied to older minors would be "in complete disregard of the text" of COPA. ACLU, 322 F.3d at 253. There is no portion of the statute that could be severed to satisfy the First Amendment since the terms "commercial purposes" and "minor" cannot be removed and leave a viable statute. Thus, I conclude that COPA is unconstitutional as a result of its overbreadth.
Findings of Fact 182 through 186 show that technology related to limiting access to Web sites based on the geographic location of the user has progressed since the preliminary injunction was issued by this court in 1999. But, these same Findings also establish that this technology is far from perfect and cost prohibitive for many Web site operators that could be subject to COPA. See 31 F. Supp.2d at 484. Such technology is relevant in addressing how "community standards" in the definition of harmful to minors applies to the Internet which does not lend itself easily to geographic constraints. 47 U.S.C. § 231(e)(6). However, since this court's overbreadth holding is only supplemental and the additional analysis of the thorny "community standards" question would merely serve to unnecessarily complicate this adjudication, I will not reach the issue.
V. CONCLUSIONS
I agree with Congress that its goal of protecting children from sexually explicit materials on the Web deemed harmful to them is especially crucial. This court, along with a broad spectrum of the population across the country yearn for a solution which would protect children from such material with 100 percent effectiveness. However, I am acutely aware of my charge under the law to uphold the principles found in our nation's Constitution and their enforcement throughout the years by the Supreme Court. I may not turn a blind eye to the law in order to attempt to satisfy my urge to protect this nation's youth by upholding a flawed statute, especially when a more effective and less restrictive alternative is readily available (although I do recognize that filters are neither a panacea nor necessarily found to be the ultimate solution to the problem at hand). My feelings resonate with the words of Justice Kennedy, who faced a similar dilemma when the Supreme Court struck down a statute that criminalized the burning of the American flag:
The hard fact is that sometimes we must make decisions we do not like. We make them because they are right, right in the sense that the law and the Constitution, as we see them, compel the result. And so great is our commitment to the process that, except in the rare case, we do not pause to express distaste for the result, perhaps for fear of undermining a valued principle that dictates the decision. This is one of those rare cases.Texas v. Johnson, 491 U.S. 397, 420-421 (1989) (Kennedy, J. concurring). Despite my personal regret at having to set aside yet another attempt to protect our children from harmful material, I restate today, as I stated when granting the preliminary injunction in this case, that "I without hesitation acknowledge the duty imposed on the Court [as Justice Kennedy observed] and the greater good such duty serves. Indeed, perhaps we do the minors of this country harm if First Amendment protections, which they will with age inherit fully, are chipped away in the name of their protection." 31 F. Supp.2d at 498.
For the forgoing reasons, I conclude that COPA facially violates the First and Fifth Amendment rights of the plaintiffs because: (1) COPA is not narrowly tailored to the compelling interest of Congress; (2) defendant has failed to meet his burden of showing that COPA is the least restrictive and most effective alternative in achieving the compelling interest; and (3) COPA is impermissibly vague and overbroad. Therefore, I will enter a permanent injunction against the enforcement of COPA.
As a result, it is unnecessary to address the plaintiffs' other claims that COPA interferes with the First and Fifth Amendment rights of older minors to access and view sexually explicit material that is not harmful to them and that COPA violates the First and Fifth Amendment right to communicate and access information anonymously. Superintendent, Massachusetts. Corr. Inst., Walpole v. Hill, 472 U.S. 445, 453 (1985) (noting "the rule of judicial restraint requiring [courts] to avoid unnecessary resolution of constitutional issues").
An appropriate Order follows.
ORDER
AND NOW, this 22nd day of March, 2007, upon consideration of the evidence, testimony of the witnesses and experts, and the arguments of counsel presented during the trial of this matter and the pre and post-trial submissions by the parties (see Doc. Nos. 319, 342, 343, 430, 429), it is hereby ORDERED, that based upon the Findings of Fact and Conclusions of Law detailed above:(1) The Child Online Protection Act, 47 U.S.C. § 231, is facially violative of the First and Fifth Amendments of the United States Constitution; and
(2) Defendant Alberto R. Gonzales, in his official capacity as Attorney General of the United States, and his officers, agents, employees, and attorneys, and those persons in active concert or participation with defendant who receive actual notice of this Order are PERMANENTLY ENJOINED from enforcing or prosecuting matters premised upon 47 U.S.C. § 231 at any time for any conduct.
This is a FINAL ORDER and this case is CONCLUDED.