FOR THE EASTERN DISTRICT OF PENNSYLVANIA (continued 2)

Another technique that filtering companies use in order to deal with a structural feature of the Internet is blocking the root level URLs of so-called "loophole" Web sites. These are Web sites that provide access to a particular Web page, but display in the user's browser a URL that is different from the URL with which the particular page is usually associated. Because of this feature, they provide a "loophole" that can be used to get around filtering software, i.e., they display a URL that is different from the one that appears on the filtering company's control list. "Loophole" Web sites include caches of Web pages that have been removed from their original location, "anonymizer" sites, and translation sites. Caches are archived copies that some search engines, such as Google, keep of the Web pages they index. The cached copy stored by Google will have a URL that is different from the original URL. Because Web sites often change rapidly, caches are the only way to access pages that have been taken down, revised, or have changed their URLs for some reason. For example, a magazine might place its current stories under a given URL, and replace them monthly with new stories. If a user wanted to find an article published six months ago, he or she would be unable to access it if not for Google's cached version.

Some sites on the Web serve as a proxy or intermediary between a user and another Web page. When using a proxy server, a user does not access the page from its original URL, but rather from the URL of the proxy server. One type of proxy service is an "anonymizer." Users may access Web sites indirectly via an anonymizer when they do not want the Web site they are visiting to be able to determine the IP address from which they are accessing the site, or to leave "cookies" on their browser. Some proxy servers can be used to attempt to translate Web page content from one language to another. Rather than directly accessing the original Web page in its original language, users can instead indirectly access the page via a proxy server offering translation features. As noted above, filtering companies often block loophole sites, such as caches, anonymizers, and translation sites. The practice of blocking loophole sites necessarily results in a significant amount of overblocking, because the vast majority of the pages that are cached, for example, do not contain content that would match a filtering company's category definitions. Filters that do not block these loophole sites, however, may enable users to access any URL on the Web via the loophole site, thus resulting in substantial underblocking. 3. The Process for "Re-Reviewing" Web Pages After Their Initial Categorization Most filtering software companies do not engage in subsequent reviews of categorized sites or pages on a scheduled basis. Priority is placed on reviewing and categorizing new sites and pages, rather than on re-reviewing already categorized sites and pages. Typically, a filtering software vendor's previous categorization of a Web site is not re-reviewed for accuracy when new pages are added to the Web site. To the extent the Web site was previously categorized as a whole, the new pages added to the site usually share the categorization assigned by the blocking product vendor. This necessarily results in both over- and underblocking, because, as noted above, the content of Web pages and Web sites changes relatively rapidly.

In addition to the content on Web sites or pages changing rapidly, Web sites themselves may disappear and be replaced by sites with entirely different content. If an IP address associated with a particular Web site is blocked under a particular category and the Web site goes out of existence, then the IP address likely would be reassigned to a different Web site, either by an Internet service provider or by a registration organization, such as the American Registry for Internet Numbers, see http://www.arin.net. In that case, the site that received the reassigned IP address would likely be miscategorized. Because filtering companies do not engage in systematic re-review of their category lists, such a site would likely remain miscategorized unless someone submitted it to the filtering company for re-review, increasing the incidence of over- and underblocking. This failure to re-review Web pages primarily increases a filtering company's rate of overblocking. However, if a filtering company does not re-review Web pages after it determines that they do not fall into any of its blocking categories, then that would result in underblocking (because, for example, a page might add sexually explicit content). 3. The Inherent Tradeoff Between Overblocking and Underblocking

There is an inherent tradeoff between any filter's rate of overblocking (which information scientists also call "precision") and its rate of underblocking (which is also referred to as "recall"). The rate of overblocking or precision is measured by the proportion of the things a classification system assigns to a certain category that are appropriately classified. The plaintiffs' expert, Dr. Nunberg, provided the hypothetical example of a classification system that is asked to pick out pictures of dogs from a database consisting of 1000 pictures of animals, of which 80 were actually dogs. If it returned 100 hits, of which 80 were in fact pictures of dogs, and the remaining 20 were pictures of cats, horses, and deer, we would say that the system identified dog pictures with a precision of 80%. This would be analogous to a filter that overblocked at a rate of 20%. The recall measure involves determining what proportion of the actual members of a category the classification system has been able to identify. For example, if the hypothetical animal- picture database contained a total of 200 pictures of dogs, and the system identified 80 of them and failed to identify 120, it would have performed with a recall of 40%. This would be analogous to a filter that underblocked 60% of the material in a category. In automated classification systems, there is always a tradeoff between precision and recall. In the animal-picture example, the recall could be improved by using a looser set of criteria to identify the dog pictures in the set, such as any animal with four legs, and all the dogs would be identified, but cats and other animals would also be included, with a resulting loss of precision. The same tradeoff exists between rates of overblocking and underblocking in filtering systems that use automated classification systems. For example, an automated system that classifies any Web page that contains the word "sex" as sexually explicit will underblock much less, but overblock much more, than a system that classifies any Web page containing the phrase "free pictures of people having sex" as sexually explicit.

This tradeoff between overblocking and underblocking also applies not just to automated classification systems, but also to filters that use only human review. Given the approximately two billion pages that exist on the Web, the 1.5 million new pages that are added daily, and the rate at which content on existing pages changes, if a filtering company blocks only those Web pages that have been reviewed by humans, it will be impossible, as a practical matter, to avoid vast amounts of underblocking. Techniques used by human reviewers such as blocking at the IP address level, domain name level, or directory level reduce the rates of underblocking, but necessarily increase the rates of overblocking, as discussed above. To use a simple example, it would be easy to design a filter intended to block sexually explicit speech that completely avoids overblocking. Such a filter would have only a single sexually explicit Web site on its control list, which could be re-reviewed daily to ensure that its content does not change. While there would be no overblocking problem with such a filter, such a filter would have a severe underblocking problem, as it would fail to block all the sexually explicit speech on the Web other than the one site on its control list. Similarly, it would also be easy to design a filter intended to block sexually explicit speech that completely avoids underblocking. Such a filter would operate by permitting users to view only a single Web site, e.g., the Sesame Street Web site. While there would be no underblocking problem with such a filter, it would have a severe overblocking problem, as it would block access to millions of non-sexually explicit sites on the Web other than the Sesame Street site.

While it is thus quite simple to design a filter that does not overblock, and equally simple to design a filter that does not underblock, it is currently impossible, given the Internet's size, rate of growth, rate of change, and architecture, and given the state of the art of automated classification systems, to develop a filter that neither underblocks nor overblocks a substantial amount of speech. The more effective a filter is at blocking Web sites in a given category, the more the filter will necessarily overblock. Any filter that is reasonably effective in preventing users from accessing sexually explicit content on the Web will necessarily block substantial amounts of non- sexually explicit speech. 4. Attempts to Quantify Filtering Programs' Rates of Over- and Underblocking The government presented three studies, two from expert witnesses, and one from a librarian fact witness who conducted a study using Internet use logs from his own library, that attempt to quantify the over- and underblocking rates of five different filtering programs. The plaintiffs presented one expert witness who attempted to quantify the rates of over- and underblocking for various programs. Each of these attempts to quantify rates of over- and underblocking suffers from various methodological flaws.

The fundamental problem with calculating over- and underblocking rates is selecting a universe of Web sites or Web pages to serve as the set to be tested. The studies that the parties submitted in this case took two different approaches to this problem. Two of the studies, one prepared by the plaintiffs' expert witness Chris Hunter, a graduate student at the University of Pennsylvania, and the other prepared by the defendants' expert, Chris Lemmons of eTesting Laboratories, in Research Triangle Park, North Carolina, approached this problem by compiling two separate lists of Web sites, one of URLs that they deemed should be blocked according to the filters' criteria, and another of URLs that they deemed should not be blocked according to the filters' criteria. They compiled these lists by choosing Web sites from the results of certain key word searches. The problem with this selection method is that it is neither random, nor does it necessarily approximate the universe of Web pages that library patrons visit.

The two other studies, one by David Biek, head librarian at the Tacoma Public Library's main branch, and one by Cory Finnell of Certus Consulting Group, of Seattle, Washington, chose actual logs of Web pages visited by library patrons during specific time periods as the universe of Web pages to analyze. This method, while surely not as accurate as a truly random sample of the indexed Web would be (assuming it would be possible to take such a sample), has the virtue of using the actual Web sites that library patrons visited during a specific period. Because library patrons selected the universe of Web sites that Biek and Finnell's studies analyzed, this removes the possibility of bias resulting from the study author's selection of the universe of sites to be reviewed. We find that the Lemmons and Hunter studies are of little probative value because of the methodology used to select the sample universe of Web sites to be tested. We will therefore focus on the studies conducted by Finnell and Biek in trying to ascertain estimates of the rates of over- and underblocking that takes place when filters are used in public libraries. The government hired expert witness Cory Finnell to study the Internet logs compiled by the public libraries systems in Tacoma, Washington; Westerville, Ohio; and Greenville, South Carolina. Each of these libraries uses filtering software that keeps a log of information about individual Web site requests made by library patrons. Finnell, whose consulting firm specializes in data analysis, has substantial experience evaluating Internet access logs generated on networked systems. He spent more than a year developing a reporting tool for N2H2, and, in the course of that work, acquired a familiarity with the design and operation of Internet filtering products.

The Tacoma library uses Cyber Patrol filtering software, and logs information only on sites that were blocked. Finnell worked from a list of all sites that were blocked in the Tacoma public library in the month of August 2001. The Westerville library uses the Websense filtering product, and logs information on both blocked sites and non-blocked sites. When the logs reach a certain size, they are overwritten by new usage logs. Because of this overwriting feature, logs were available to Finnell only for the relatively short period from October 1, 2001 to October 3, 2001. The Greenville library uses N2H2's filtering product and logs both blocked sites and sites that patrons accessed. The logs contain more than 500,000 records per day. Because of the volume of the records, Finnell restricted his analysis to the period from August 2, 2001 to August 15, 2001.

Finnell calculated an overblocking rate for each of the three libraries by examining the host Web site containing each of the blocked pages. He did not employ a sampling technique, but instead examined each blocked Web site. If the contents of a host Web site or the pages within the Web site were consistent with the filtering product's definition of the category under which the site was blocked, Finnell considered it to be an accurate block. Finnell and three others, two of whom were temporary employees, examined the Web sites to determine whether they were consistent with the filtering companies' category definitions. Their review was, of course, necessarily limited by: (1) the clarity of the filtering companies' category definitions; (2) Finnell's and his employees' interpretations of the definitions; and (3) human error. The study's reliability is also undercut by the fact that Finnell failed to archive the blocked Web pages as they existed either at the point that a patron in one of the three libraries was denied access or when Finnell and his team reviewed the pages. It is therefore impossible for anyone to check the accuracy and consistency of Finnell's review team, or to know whether the pages contained the same content when the block occurred as they did when Finnell's team reviewed them. This is a key flaw, because the results of the study depend on individual determinations as to overblocking and underblocking, in which Finnell and his team were required to compare what they saw on the Web pages that they reviewed with standard definitions provided by the filtering company.

Tacoma library's Cyber Patrol software blocked 836 unique Web sites during the month of August. Finnell determined that 783 of those blocks were accurate and that 53 were inaccurate. The error rate for Cyber Patrol was therefore estimated to be 6.34%, and the true error rate was estimated with 95% confidence to lie within the range of 4.69% to 7.99%. Finnell and his team reviewed 185 unique Web sites that were blocked by Westerville Library's Websense filter during the logged period and determined that 158 of them were accurate and that 27 of them were inaccurate. He therefore estimated the Websense filter's overblocking rate at 14.59% with a 95% confidence interval of 9.51% to 19.68%. Additionally, Finnell examined 1,674 unique Web sites that were blocked by the Greenville Library's N2H2 filter during the relevant period and determined that 1,520 were accurate and that 87 were inaccurate. This yields an estimated overblocking rate of 5.41% and a 95% confidence interval of 4.33% to 6.55%. Finnell's methodology was materially flawed in that it understates the rate of overblocking for the following reasons. First, patrons from the three libraries knew that the filters were operating, and may have been deterred from attempting to access Web sites that they perceived to be "borderline" sites, i.e., those that may or may not have been appropriately filtered according to the filtering companies' category definitions. Second, in their cross-examination of Finnell, the plaintiffs offered screen shots of a number of Web sites that, according to Finnell, had been appropriately blocked, but that Finnell admitted contained only benign materials. Finnell's explanation was that the Web sites must have changed between the time when he conducted the study and the time of the trial, but because he did not archive the images as they existed when his team reviewed them for the study, there is no way to verify this. Third, because of the way in which Finnell counted blocked Web sites i.e., if separate patrons attempted to reach the same Web site, or one or more patrons attempted to access more than one page on a single Web site, Finnell counted these attempts as a single block, see supra note 10 his results necessarily understate the number of times that patrons were erroneously denied access to information.

At all events, there is no doubt that Finnell's estimated rates of overblocking, which are based on the filtering companies' own category definitions, significantly understate the rate of overblocking with respect to CIPA's category definitions for filtering for adults. The filters used in the Tacoma, Westerville, and Greenville libraries were configured to block, among other things, images of full nudity and sexually explicit materials. There is no dispute, however, that these categories are far broader than CIPA's categories of visual depictions that are obscene, or child pornography, the two categories of material that libraries subject to CIPA must certify that they filter during adults' use of the Internet. Finnell's study also calculated underblocking rates with respect to the Westerville and Greenville Libraries (both of which logged not only their blocked sites, but all sites visited by their patrons), by taking random samples of URLs from the list of sites that were not blocked. The study used a sample of 159 sites that were accessed by Westerville patrons and determined that only one of them should have been blocked under the software's category definitions, yielding an underblocking rate of 0.6%. Given the size of the sample, the 95% confidence interval is 0% to 1.86%. The study examined a sample of 254 Web sites accessed by patrons in Greenville and found that three of them should have been blocked under the filtering software's category definitions. This results in an estimated underblocking rate of 1.2% with a 95% confidence interval ranging from 0% to 2.51%.

We do not credit Finnell's estimates of the rates of underblocking in the Westerville and Greenville public libraries for several reasons. First, Finnell's estimates likely understate the actual rate of underblocking because patrons, who knew that filtering programs were operating in the Greenville and Westerville Libraries, may have refrained from attempting to access sites with sexually explicit materials, or other contents that they knew would probably meet a filtering program's blocked categories. Second, and most importantly, we think that the formula that Finnell used to calculate the rate of underblocking in these two libraries is not as meaningful as the formula that information scientists typically use to calculate a rate of recall, which we describe above in Subsection II.E.3. As Dr. Nunberg explained, the standard method that information scientists use to calculate a rate of recall is to sort a set of items into two groups, those that fall into a particular category (e.g., those that should have been blocked by a filter) and those that do not. The rate of recall is then calculated by dividing the number of items that the system correctly identified as belonging to the category by the total number of items in the category.

In the example above, we discussed a database that contained 1000 photographs. Assume that 200 of these photographs were pictures of dogs. If, for example, a classification system designed to identify pictures of dogs identified 80 of the dog pictures and failed to identify 120, it would have performed with a recall rate of 40%. This would be analogous to a filter that underblocked at a rate of 60%. To calculate the recall rate of the filters in the Westerville and Greenville public libraries in accordance with the standard method described above, Finnell should have taken a sample of sites from the libraries' Internet use logs (including both sites that were blocked and sites that were not), and divided the number of sites in the sample that the filter incorrectly failed to block by the total number of sites in the sample that should have been blocked. What Finnell did instead was to take a sample of sites that were not blocked, and divide the total number of sites in this sample by the number of sites in the sample that should have been blocked. This made the denominator that Finnell used much larger than it would have been had he used the standard method for calculating recall, consequently making the underblocking rate that he calculated much lower than it would have been under the standard method.

Moreover, despite the relatively low rates of underblocking that Finnell's study found, librarians from several of the libraries proffered by defendants that use blocking products, including Greenville, Tacoma, and Westerville, testified that there are instances of underblocking in their libraries. No quantitative evidence was presented comparing the effectiveness of filters and other alternative methods used by libraries to prevent patrons from accessing visual depictions that are obscene, child pornography, or in the case of minors, harmful to minors. Biek undertook a similar study of the overblocking rates that result from the Tacoma Library's use of the Cyber Patrol software. He began with the 3,733 individual blocks that occurred in the Tacoma Library in October 2000 and drew from this data set a random sample of 786 URLs. He calculated two rates of overblocking, one with respect to the Tacoma Library's policy on Internet use that the pictorial content of the site may not include "graphic materials depicting full nudity and sexual acts which are portrayed obviously and exclusively for sensational or pornographic purposes" and the other with respect to Cyber Patrol's own category definitions. He estimated that Cyber Patrol overblocked 4% of all Web pages in October 2000 with respect to the definitions of the Tacoma Library's Internet Policy and 2% of all pages with respect to Cyber Patrol's own category definitions.

It is difficult to determine how reliable Biek's conclusions are, because he did not keep records of the raw data that he used in his study; nor did he archive images of the Web pages as they looked when he made the determination whether they were properly classified by the Cyber Patrol program. Without this information, it is impossible to verify his conclusions (or to undermine them). And Biek's study certainly understates Cyber Patrol's overblocking rate for some of the same reasons that Finnell's study likely understates the true rates of overblocking used in the libraries that he studied. We also note that Finnell's study, which analyzed a set of Internet logs from the Tacoma Library during which the same filtering program was operating with the same set of blocking categories enabled, found a significantly higher rate of overblocking than the Biek study did. Biek found a rate of overblocking of approximately 2% while the Finnell study estimated a 6.34% rate of overblocking. At all events, the category definitions employed by CIPA, at least with respect to adult use visual depictions that are obscene or child pornography are narrower than the materials prohibited by the Tacoma Library policy, and therefore Biek's study understates the rate of overblocking with respect to CIPA's definitions for adults. In sum, we think that Finnell's study, while we do not credit its estimates of underblocking, is useful because it states lower bounds with respect to the rates of overblocking that occurred when the Cyber Patrol, Websense, and N2H2 filters were operating in public libraries. While these rates are substantial between nearly 6% and 15% we think, for the reasons stated above, that they greatly understate the actual rates of overblocking that occurs, and therefore cannot be considered as anything more than minimum estimates of the rates of overblocking that happens in all filtering programs. 5. Methods of Obtaining Examples of Erroneously Blocked Web Sites

The plaintiffs assembled a list of several thousand Web sites that they contend were, at the time of the study, likely to have been erroneously blocked by one or more of four major commercial filtering programs: SurfControl Cyber Patrol 6.0.1.47, N2H2 Internet Filtering 2.0, Secure Computing SmartFilter 3.0.0.01, and Websense Enterprise 4.3.0. They compiled this list using a two-step process. First, Benjamin Edelman, an expert witness who testified before us, compiled a list of more than 500,000 URLs and devised a program to feed them through all four filtering programs in order to compile a list of URLs that might have been erroneously blocked by one or more of the programs. Second, Edelman forwarded subsets of the list that he compiled to librarians and professors of library science whom the plaintiffs had hired to review the blocked sites for suitability in the public library context. Edelman assembled the list of URLs by compiling Web pages that were blocked by the following categories in the four programs: Cyber Patrol: Adult/Sexually Explicit; N2H2: Adults Only, Nudity, Pornography, and Sex, with "exceptions" engaged in the categories of Education, For Kids, History, Medical, Moderated, and Text/Spoken Only; SmartFilter: Sex, Nudity, Mature, and Extreme; Websense: Adult Content, Nudity, and Sex.

Edelman then assembled a database of Web sites for possible testing. He derived this list by automatically compiling URLs from the Yahoo index of Web sites, taking them from categories from the Yahoo index that differed significantly from the classifications that he had enabled in each of the blocking programs (taking, for example, Web sites from Yahoo's "Government" category). He then expanded this list by entering URLs taken from the Yahoo index into the Google search engine's "related" search function, which provides the user with a list of similar sites. Edelman also included and excluded specific Web sites at the request of the plaintiffs' counsel.

Taking the list of more than 500,000 URLs that he had compiled, Edelman used an automated system that he had developed to test whether particular URLs were blocked by each of the four filtering programs. This testing took place between February and October 2001. He recorded the specific dates on which particular sites were blocked by particular programs, and, using commercial archiving software, archived the contents of the home page of the blocked Web sites (and in some instances the pages linked to from the home page) as it existed when it was blocked. Through this process, Edelman, whose testimony we credit, compiled a list of 6,777 URLs that were blocked by one or more of the four programs. Because these sites were chosen from categories from the Yahoo directory that were unrelated to the filtering categories that were enabled during the test (i.e., "Government" vs. "Nudity"), he reasoned that they were likely erroneously blocked. As explained in the margin, Edelman repeated his testing and discovered that Cyber Patrol had unblocked most of the pages on the list of 6,777 after he had published the list on his Web site. His records indicate that an employee of SurfControl (the company that produces Cyber Patrol software) accessed his site and presumably checked out the URLs on the list, thus confirming Edelman's judgment that the majority of URLs on the list were erroneously blocked. Edelman forwarded the list of blocked sites to Dr. Joseph Janes, an Assistant Professor in the Information School of the University of Washington who also testified at trial as an expert witness. Janes reviewed the sites that Edelman compiled to determine whether they are consistent with library collection development, i.e., whether they are sites to which a reference librarian would, consistent with professional standards, direct a patron as a source of information.

Edelman forwarded Janes a list of 6,775 Web sites, almost the entire list of blocked sites that he collected, from which Janes took a random sample of 859 using the SPSS statistical software package. Janes indicated that he chose a sample size of 859 because it would yield a 95% confidence interval of plus or minus 2.5%. Janes recruited a group of 16 reviewers, most of whom were current or former students at the University of Washington's Information School, to help him identify which sites were appropriate for library use. We describe the process that he used in the margin. Due to the inability of a member of Janes's review team to complete the reviewing process, Janes had to cut 157 Web sites out of the sample, but because the Web sites were randomly assigned to reviewers, it is unlikely that these sites differed significantly from the rest of the sample. That left the sample size at 699, which widened the 95% confidence interval to plus or minus 2.8%.

Of the total 699 sites reviewed, Janes's team concluded that 165 of them, or 23.6% percent of the sample, were not of any value in the library context (i.e., no librarian would, consistent with professional standards, refer a patron to these sites as a source of information). They were unable to find 60 of the Web sites, or 8.6% of the sample. Therefore, they concluded that the remaining 474 Web sites, or 67.8% of the sample, were examples of overblocking with respect to materials that are appropriate sources of information in public libraries. Applying a 95% confidence interval of plus or minus 2.8%, the study concluded that we can be 95% confident that the actual percentage of sites in the list of 6,775 sites that are appropriate for use in public libraries is somewhere between 65.0% and 70.6%. In other words, we can be 95% certain that the actual number of sites out of the 6,775 that Edelman forwarded to Janes that are appropriate for use in public libraries (under Janes's standard) is somewhere between 4,403 and 4,783.

The government raised some valid criticisms of Janes's methodology, attacking in particular the fact that, while sites that received two "yes" votes in the first round of voting were determined to be of sufficient interest in a library context to be removed from further analysis, sites receiving one or two "no" votes were sent to the next round. The government also correctly points out that results of Janes's study can be generalized only to the population of 6,775 sites that Edelman forwarded to Janes. Even taking these criticisms into account, and discounting Janes's numbers appropriately, we credit Janes's study as confirming that Edelman's set of 6,775 Web sites contains at least a few thousand URLs that were erroneously blocked by one or more of the four filtering programs that he used, whether judged against CIPA's definitions, the filters' own category criteria, or against the standard that the Janes study used. Edelman tested only 500,000 unique URLs out of the 4000 times that many, or two billion, that are estimated to exist in the indexable Web. Even assuming that Edelman chose the URLs that were most likely to be erroneously blocked by commercial filtering programs, we conclude that many times the number of pages that Edelman identified are erroneously blocked by one or more of the filtering programs that he tested. Edelman's and Janes's studies provide numerous specific examples of Web pages that were erroneously blocked by one or more filtering programs. The Web pages that were erroneously blocked by one or more of the filtering programs do not fall into any neat patterns; they range widely in subject matter, and it is difficult to tell why they may have been overblocked. The list that Edelman compiled, for example, contains Web pages relating to religion, politics and government, health, careers, education, travel, sports, and many other topics. In the next section, we provide examples from each of these categories. 6. Examples of Erroneously Blocked Web Sites

Several of the erroneously blocked Web sites had content relating to churches, religious orders, religious charities, and religious fellowship organizations. These included the following Web sites: the Knights of Columbus Council 4828, a Catholic men's group associated with St. Patrick's Church in Fallon, Nevada, http://msnhomepages.talkcity.com/SpiritSt/kofc4828, which was blocked by Cyber Patrol in the "Adult/Sexually Explicit" category; the Agape Church of Searcy, Arkansas, http://www.agapechurch.com, which was blocked by Websense as "Adult Content"; the home page of the Lesbian and Gay Havurah of the Long Beach, California Jewish Community Center, http://www.compupix.com/gay/havurah.htm, which was blocked by N2H2 as "Adults Only, Pornography," by Smartfilter as "Sex," and by Websense as "Sex"; Orphanage Emmanuel, a Christian orphanage in Honduras that houses 225 children, http://home8.inet.tele.dk/rfb_viva, which was blocked by Cyber Patrol in the "Adult/Sexually Explicit" category; Vision Art Online, which sells wooden wall hangings for the home that contain prayers, passages from the Bible, and images of the Star of David, http://www.visionartonline.com, which was blocked in Websense's "Sex" category; and the home page of Tenzin Palmo, a Buddhist nun, which contained a description of her project to build a Buddhist nunnery and international retreat center for women, http://www.tenzinpalmo.com, which was categorized as "Nudity" by N2H2.

Several blocked sites also contained information about governmental entities or specific political candidates, or contained political commentary. These included: the Web site for Kelley Ross, a Libertarian candidate for the California State Assembly, http://www.friesian.com/ross/ca40, which N2H2 blocked as "Nudity"; the Web site for Bob Coughlin, a town selectman in Dedham, Massachusetts, http://www.bobcoughlin.org, which was blocked under N2H2's "Nudity" category; a list of Web sites containing information about government and politics in Adams County, Pennsylvania, http://www.geocities.com/adamscopa, which was blocked by Websense as "Sex"; the Web site for Wisconsin Right to Life, http://www.wrtl.org, which N2H2 blocked as "Nudity"; a Web site that promotes federalism in Uganda, http://federo.com, which N2H2 blocked as "Adults Only, Pornography"; "Fight the Death Penalty in the USA," a Danish Web site dedicated to criticizing the American system of capital punishment, http://www.fdp.dk, which N2H2 blocked as "Pornography"; and "Dumb Laws," a humor Web site that makes fun of outmoded laws, http://www.dumblaws.com, which N2H2 blocked under its "Sex" category. Erroneously blocked Web sites relating to health issues included the following: a guide to allergies, http://www.x- sitez.com/allergy, which was categorized as "Adults Only, Pornography" by N2H2; a health question and answer site sponsored by Columbia University, http://www.goaskalice.com.columbia.edu, which was blocked as "Sex" by N2H2, and as "Mature" by Smartfilter; the Western Amputee Support Alliance Home Page, http://www.usinter.net/wasa, which was blocked by N2H2 as "Pornography"; the Web site of the Willis-Knighton Cancer Center, a Shreveport, Louisiana cancer treatment facility, http://cancerftr.wkmc.com, which was blocked by Websense under the "Sex" category; and a site dealing with halitosis, http://www.dreamcastle.com/tungs, which was blocked by N2H2 as "Adults, Pornography," by Smartfilter as "Sex," by Cyber Patrol as "Adult/Sexually Explicit," and by Websense as "Adult Content."

The filtering programs also erroneously blocked several Web sites having to do with education and careers. The filtering programs blocked two sites that provide information on home schooling. "HomEduStation the Internet Source for Home Education," http://www.perigee.net/~mcmullen/homedustation/, was categorized by Cyber Patrol as "Adult/Sexually Explicit." Smartfilter blocked "Apricot: A Web site made by and for home schoolers," http://apricotpie.com, as "Sex." The programs also miscategorized several career-related sites. "Social Work Search," http://www.socialworksearch.com/, is a directory for social workers that Cyber Patrol placed in its "Adult/Sexually Explicit" category. The "Gay and Lesbian Chamber of Southern Nevada," http://www.lambdalv.com, "a forum for the business community to develop relationships within the Las Vegas lesbian, gay, transsexual, and bisexual community" was blocked by N2H2 as "Adults Only, Pornography." A site for aspiring dentists, http://www.vvm.com/~bond/home.htm, was blocked by Cyber Patrol in its "Adult/Sexually Explicit" category. The filtering programs erroneously blocked many travel Web sites, including: the Web site for the Allen Farmhouse Bed & Breakfast of Alleghany County, North Carolina, http://planet- nc.com/Beth/index.html, which Websense blocked as "Adult Content"; Odysseus Gay Travel, a travel company serving gay men, http://www.odyusa.com, which N2H2 categorized as "Adults Only, Pornography"; Southern Alberta Fly Fishing Outfitters, http://albertaflyfish.com, which N2H2 blocked as "Pornography"; and "Nature and Culture Conscious Travel," a tour operator in Namibia, http://www.trans-namibia-tours.com, which was categorized as "Pornography" by N2H2.

The filtering programs also miscategorized a large number of sports Web sites. These included: a site devoted to Willie O'Ree, the first African-American player in the National Hockey League, http://www.missioncreep.com/mw/oree.html, which Websense blocked under its "Nudity" category; the home page of the Sydney University Australian Football Club, http://www.tek.com.au/suafc, which N2H2 blocked as "Adults Only, Pornography," Smartfilter blocked as "Sex," Cyber Patrol blocked as "Adult/Sexually Explicit" and Websense blocked as "Sex"; and a fan's page devoted to the Toronto Maple Leafs hockey team, http://www.torontomapleleafs.atmypage.com, which N2H2 blocked under the "Pornography" category. 7. Conclusion: The Effectiveness of Filtering Programs Public libraries have adopted a variety of means of dealing with problems created by the provision of Internet access. The large amount of sexually explicit speech that is freely available on the Internet has, to varying degrees, led to patron complaints about such matters as unsought exposure to offensive material, incidents of staff and patron harassment by individuals viewing sexually explicit content on the Internet, and the use of library computers to access illegal material, such as child pornography. In some libraries, youthful library patrons have persistently attempted to use the Internet to access hardcore pornography.

Those public libraries that have responded to these problems by using software filters have found such filters to provide a relatively effective means of preventing patrons from accessing sexually explicit material on the Internet. Nonetheless, out of the entire universe of speech on the Internet falling within the filtering products' category definitions, the filters will incorrectly fail to block a substantial amount of speech. Thus, software filters have not completely eliminated the problems that public libraries have sought to address by using the filters, as evidenced by frequent instances of underblocking. Nor is there any quantitative evidence of the relative effectiveness of filters and the alternatives to filters that are also intended to prevent patrons from accessing illegal content on the Internet. Even more importantly (for this case), although software filters provide a relatively cheap and effective, albeit imperfect, means for public libraries to prevent patrons from accessing speech that falls within the filters' category definitions, we find that commercially available filtering programs erroneously block a huge amount of speech that is protected by the First Amendment. Any currently available filtering product that is reasonably effective in preventing users from accessing content within the filter's category definitions will necessarily block countless thousands of Web pages, the content of which does not match the filtering company's category definitions, much less the legal definitions of obscenity, child pornography, or harmful to minors. Even Finnell, an expert witness for the defendants, found that between 6% and 15% of the blocked Web sites in the public libraries that he analyzed did not contain content that meets even the filtering products' own definitions of sexually explicit content, let alone CIPA's definitions.

This phenomenon occurs for a number of reasons explicated in the more detailed findings of fact supra. These include limitations on filtering companies' ability to: (1) harvest Web pages for review; (2) review and categorize the Web pages that they have harvested; and (3) engage in regular re-review of the Web pages that they have previously reviewed. The primary limitations on filtering companies' ability to harvest Web pages for review is that a substantial majority of pages on the Web are not indexable using the spidering technology that Web search engines use, and that together, search engines have indexed only around half of the Web pages that are theoretically indexable. The fast rate of growth in the number of Web pages also limits filtering companies' ability to harvest pages for review. These shortcomings necessarily result in significant underblocking. Several limitations on filtering companies' ability to review and categorize the Web pages that they have harvested also contribute to over- and underblocking. First, automated review processes, even those based on "artificial intelligence," are unable with any consistency to distinguish accurately material that falls within a category definition from material that does not. Moreover, human review of URLs is hampered by filtering companies' limited staff sizes, and by human error or misjudgment. In order to deal with the vast size of the Web and its rapid rates of growth and change, filtering companies engage in several practices that are necessary to reduce underblocking, but inevitably result in overblocking. These include: (1) blocking whole Web sites even when only a small minority of their pages contain material that would fit under one of the filtering company's categories (e.g., blocking the Salon.com site because it contains a sex column); (2) blocking by IP address (because a single IP address may contain many different Web sites and many thousands of pages of heterogenous content); and (3) blocking loophole sites such as translator sites and cache sites, which archive Web pages that have been removed from the Web by their original publisher.

Finally, filtering companies' failure to engage in regular re-review of Web pages that they have already categorized (or that they have determined do not fall into any category) results in a substantial amount of over- and underblocking. For example, Web publishers change the contents of Web pages frequently. The problem also arises when a Web site goes out of existence and its domain name or IP address is reassigned to a new Web site publisher. In that case, a filtering company's previous categorization of the IP address or domain name would likely be incorrect, potentially resulting in the over- or underblocking of many thousands of pages. The inaccuracies that result from these limitations of filtering technology are quite substantial. At least tens of thousands of pages of the indexable Web are overblocked by each of the filtering programs evaluated by experts in this case, even when considered against the filtering companies' own category definitions. Many erroneously blocked pages contain content that is completely innocuous for both adults and minors, and that no rational person could conclude matches the filtering companies' category definitions, such as "pornography" or "sex."

The number of overblocked sites is of course much higher with respect to the definitions of obscenity and child pornography that CIPA employs for adults, since the filtering products' category definitions, such as "sex" and "nudity," encompass vast amounts of Web pages that are neither child pornography nor obscene. Thus, the number of pages of constitutionally protected speech blocked by filtering products far exceeds the many thousands of pages that are overblocked by reference to the filtering products' category definitions.

No presently conceivable technology can make the judgments necessary to determine whether a visual depiction fits the legal definitions of obscenity, child pornography, or harmful to minors. Given the state of the art in filtering and image recognition technology, and the rapidly changing and expanding nature of the Web, we find that filtering products' shortcomings will not be solved through a technical solution in the foreseeable future. In sum, filtering products are currently unable to block only visual depictions that are obscene, child pornography, or harmful to minors (or, only content matching a filtering product's category definitions) while simultaneously allowing access to all protected speech (or, all content not matching the blocking product's category definitions). Any software filter that is reasonably effective in blocking access to Web pages that fall within its category definitions will necessarily erroneously block a substantial number of Web pages that do not fall within its category definitions. 2. Analytic Framework for the Opinion: The Centrality of Dole and the Role of the Facial Challenge

Both the plaintiffs and the government agree that, because this case involves a challenge to the constitutionality of the conditions that Congress has set on state actors' receipt of federal funds, the Supreme Court's decision in South Dakota v. Dole, 483 U.S. 203 (1987), supplies the proper threshold analytic framework. The constitutional source of Congress's spending power is Article I, Sec. 8, cl. 1, which provides that "Congress shall have Power . . . to pay the Debts and provide for the common Defence and general Welfare of the United States." In Dole, the Court upheld the constitutionality of a federal statute requiring the withholding of federal highway funds from any state with a drinking age below 21. Id. at 211-12. In sustaining the provision's constitutionality, Dole articulated four general constitutional limitations on Congress's exercise of the spending power.

First, "the exercise of the spending power must be in pursuit of 'the general welfare.'" Id. at 207. Second, any conditions that Congress sets on states' receipt of federal funds must be sufficiently clear to enable recipients "to exercise their choice knowingly, cognizant of the consequences of their participation." Id. (internal quotation marks and citation omitted). Third, the conditions on the receipt of federal funds must bear some relation to the purpose of the funding program. Id. And finally, "other constitutional provisions may provide an independent bar to the conditional grant of federal funds." Id. at 208. In particular, the spending power "may not be used to induce the States to engage in activities that would themselves be unconstitutional. Thus, for example, a grant of federal funds conditioned on invidiously discriminatory state action or the infliction of cruel and unusual punishment would be an illegitimate exercise of the Congress' broad spending power." Id. at 210.

Plaintiffs do not contend that CIPA runs afoul of the first three limitations. However, they do allege that CIPA is unconstitutional under the fourth prong of Dole because it will induce public libraries to violate the First Amendment. Plaintiffs therefore submit that the First Amendment "provide[s] an independent bar to the conditional grant of federal funds" created by CIPA. Id. at 208. More specifically, they argue that by conditioning public libraries' receipt of federal funds on the use of software filters, CIPA will induce public libraries to violate the First Amendment rights of Internet content-providers to disseminate constitutionally protected speech to library patrons via the Internet, and the correlative First Amendment rights of public library patrons to receive constitutionally protected speech on the Internet. The government concedes that under the Dole framework, CIPA is facially invalid if its conditions will induce public libraries to violate the First Amendment. The government and the plaintiffs disagree, however, on the meaning of Dole's "inducement" requirement in the context of a First Amendment facial challenge to the conditions that Congress places on state actors' receipt of federal funds. The government contends that because plaintiffs are bringing a facial challenge, they must show that under no circumstances is it possible for a public library to comply with CIPA's conditions without violating the First Amendment. The plaintiffs respond that even if it is possible for some public libraries to comply with CIPA without violating the First Amendment, CIPA is facially invalid if it "will result in the impermissible suppression of a substantial amount of protected speech."

Because it was clear in Dole that the states could comply with the challenged conditions that Congress attached to the receipt of federal funds without violating the Constitution, the Dole Court did not have occasion to explain fully what it means for Congress to use the spending power to "induce [recipients] to engage in activities that would themselves be unconstitutional." Dole, 483 U.S. at 210; see id. at 211 ("Were South Dakota to succumb to the blandishments offered by Congress and raise its drinking age to 21, the State's action in so doing would not violate the constitutional rights of anyone."). Although the proposition that Congress may not pay state actors to violate citizens' First Amendment rights is unexceptionable when stated in the abstract, it is unclear what exactly a litigant must establish to facially invalidate an exercise of Congress's spending power on this ground. In general, it is well-established that a court may sustain a facial challenge to a statute only if the plaintiff demonstrates that the statute admits of no constitutional application. See United States v. Salerno, 481 U.S. 739, 745 (1987) ("A facial challenge to a legislative Act is, of course, the most difficult challenge to mount successfully, since the challenger must establish that no set of circumstances exists under which the Act would be valid."); see also Bowen v. Kendrick, 487 U.S. 589, 612 (1988) ("It has not been the Court's practice, in considering facial challenges to statutes of this kind, to strike them down in anticipation that particular applications may result in unconstitutional use of funds.") (internal quotation marks and citation omitted).

First Amendment overbreadth doctrine creates a limited exception to this rule by permitting facial invalidation of a statute that burdens a substantial amount of protected speech, even if the statute may be constitutionally applied in particular circumstances. "The Constitution gives significant protection from overbroad laws that chill speech within the First Amendment's vast and privileged sphere. Under this principle, [a law] is unconstitutional on its face if it prohibits a substantial amount of protected expression." Ashcroft v. Free Speech Coalition, 122 S. Ct. 1389, 1399 (2002); see also Broadrick v. Oklahoma, 413 U.S. 601, 612 (1973). This more liberal test of a statute's facial validity under the First Amendment stems from the recognition that where a statute's reach contemplates a number of both constitutional and unconstitutional applications, the law's sanctions may deter individuals from challenging the law's validity by engaging in constitutionally protected speech that may nonetheless be proscribed by the law. Without an overbreadth doctrine, "the contours of regulation would have to be hammered out case by case and tested only by those hardy enough to risk criminal prosecution to determine the proper scope of regulation." Dombrowski v. Pfister, 380 U.S. 479, 487 (1965); see also Brockett v. Spokane Arcades, Inc., 472 U.S. 491, 503 (1985) ("[A]n individual whose own speech or expressive conduct may validly be prohibited or sanctioned is permitted to challenge a statute on its face because it also threatens others not before the court those who desire to engage in legally protected expression but who may refrain from doing so rather than risk prosecution or undertake to have the law declared partially invalid.").

Plaintiffs argue that the overbreadth doctrine is applicable here, since CIPA "threatens to chill free speech because it will censor a substantial amount of protected speech, because it is vague, and because the law creates a prior restraint . . . ." Unlike the statutes typically challenged as facially overbroad, however, CIPA does not impose criminal penalties on those who violate its conditions. Cf. Freedom of Speech Coalition, 122 S. Ct. at 1398 ("With these severe penalties in force, few legitimate movie producers or book publishers, or few other speakers in any capacity, would risk distributing images in or near the uncertain reach of this law."). Thus, the rationale for permitting facial challenges to laws that may be constitutionally applied in some instances is less compelling in cases such as this, which involve challenges to Congress's exercise of the spending power, than in challenges to criminal statutes. Nonetheless, "even minor punishments can chill protected speech," id., and absent the ability to challenge CIPA on its face, public libraries that depend on federal funds may decide to comply with CIPA's terms, thereby denying patrons access to substantial amounts of constitutionally protected speech, rather than refusing to comply with CIPA's terms and consequently losing the benefits of federal funds. See 47 C.F.R. Sec. 54.520(e)(1) ("A school or library that knowingly fails to ensure the use of computers in accordance with the certifications required by this section, must reimburse any funds and discounts received under the federal universal support service support mechanism for schools and libraries for the period in which there was noncompliance."). Even in cases where the only penalty for failure to comply with a statute is the withholding of federal funds, the Court has sustained facial challenges to Congress's exercise of the spending power. See, e.g., Legal Servs. Corp. v. Velazquez, 531 U.S. 533 (2001) (declaring unconstitutional on its face a federal statute restricting the ability of legal services providers who receive federal funds to engage in activity protected by the First Amendment).

The Court's unconstitutional conditions cases, such as Velazquez, are not strictly controlling, since they do not require a showing that recipients who comply with the conditions attached to federal funding will, as state actors, violate others' constitutional rights, as is the case under the fourth prong of Dole. However, they are highly instructive. The Supreme Court's pronouncements in the unconstitutional conditions cases on what is necessary for a plaintiff to mount a successful First Amendment facial challenge to an exercise of Congress's spending power have not produced a seamless web. For example, in Rust v. Sullivan, 500 U.S. 173 (1991), the Court rejected a First Amendment facial challenge to federal regulations prohibiting federally funded healthcare clinics from providing counseling concerning the use of abortion as a method of family planning, explaining that: Petitioners are challenging the facial validity of the regulations. Thus, we are concerned only with the question whether, on their face, the regulations are both authorized by the Act and can be construed in such a manner that they can be applied to a set of individuals without infringing upon constitutionally protected rights. Petitioners face a heavy burden in seeking to have the regulations invalidated as facially unconstitutional. . . . The fact that the regulations might operate unconstitutionally under some conceivable set of circumstances is insufficient to render them wholly invalid.

Id. at 183 (internal quotation marks, alterations, and citation omitted). In contrast, NEA v. Finley, 524 U.S. 569 (1998), which also involved a facial First Amendment challenge to an exercise of Congress's spending power, articulated a somewhat more liberal test of facial validity than Rust, explaining that "[t]o prevail, respondents must demonstrate a substantial risk that application of the provision will lead to the suppression of speech." Id. at 580. Against this background, it is unclear to us whether, to succeed in facially invalidating CIPA on the grounds that it will "induce the States to engage in activities that would themselves be unconstitutional," Dole, 483 U.S. at 210, plaintiffs must show that it is impossible for public libraries to comply with CIPA's conditions without violating the First Amendment, or rather simply that CIPA will effectively restrict library patrons' access to substantial amounts of constitutionally protected speech, therefore causing many libraries to violate the First Amendment. However, we need not resolve this issue. Rather, we may assume without deciding, for purposes of this case, that a facial challenge to CIPA requires plaintiffs to show that any public library that complies with CIPA's conditions will necessarily violate the First Amendment and, as explained in detail below, we believe that CIPA's constitutionality fails even under this more restrictive test of facial validity urged on us by the government. Because of the inherent limitations in filtering technology, public libraries can never comply with CIPA without blocking access to a substantial amount of speech that is both constitutionally protected and fails to meet even the filtering companies' own blocking criteria. We turn first to the governing legal principles to be applied to the facts in order to determine whether the First Amendment permits a library to use the filtering technology mandated by CIPA. 3. Level of Scrutiny Applicable to Content-based Restrictions on Internet Access in Public Libraries

In analyzing the constitutionality of a public library's use of Internet filtering software, we must first identify the appropriate level of scrutiny to apply to this restriction on patrons' access to speech. While plaintiffs argue that a public library's use of such filters is subject to strict scrutiny, the government maintains that the applicable standard is rational basis review. If strict scrutiny applies, the government must show that the challenged restriction on speech is narrowly tailored to promote a compelling government interest and that no less restrictive alternative would further that interest. United States v. Playboy Entm't Group, Inc., 529 U.S. 803, 813 (2000). In contrast, under rational basis review, the challenged restriction need only be reasonable; the government interest that the restriction serves need not be compelling; the restriction need not be narrowly tailored to serve that interest; and the restriction "need not be the most reasonable or the only reasonable limitation." Cornelius v. NAACP Legal Def. & Educ. Fund, 473 U.S. 788, 808 (1985).

Software filters, by definition, block access to speech on the basis of its content, and content-based restrictions on speech are generally subject to strict scrutiny. See Playboy, 529 U.S. at 813 ("[A] content-based speech restriction . . . can stand only if it satisfies strict scrutiny."). Strict scrutiny does not necessarily apply to content-based restrictions on speech, however, where the restrictions apply only to speech on government property, such as public libraries. "[I]t is . . . well settled that the government need not permit all forms of speech on property that it owns and controls." Int'l Soc'y for Krishna Consciousness, Inc. v. Lee, 505 U.S. 672, 678 (1992). We perforce turn to a discussion of public forum doctrine. 1. Overview of Public Forum Doctrine The government's power to restrict speech on its own property is not unlimited. Rather, under public forum doctrine, the extent to which the First Amendment permits the government to restrict speech on its own property depends on the character of the forum that the government has created. See Cornelius v. NAACP Legal Def. & Educ. Fund, Inc., 473 U.S. 788 (1985). Thus, the First Amendment affords greater deference to restrictions on speech in those areas considered less amenable to free expression, such as military bases, see Greer v. Spock, 424 U.S. 828 (1976), jail grounds, see Adderley v. Florida, 385 U.S. 39 (1966), or public airport terminals, see Int'l Soc'y for Krishna Consciousness, Inc. v. Lee, 505 U.S. 672 (1992), than to restrictions on speech in state universities, see Rosenberger v. Rector & Visitors of Univ. of Va., 515 U.S. 819 (1995), or streets, sidewalks and public parks, see Frisby v. Schultz, 487 U.S. 474 (1988); Hague v. CIO, 307 U.S. 496 (1939). The Supreme Court has identified three types of fora for purposes of identifying the level of First Amendment scrutiny applicable to content-based restrictions on speech on government property: traditional public fora, designated public fora, and nonpublic fora. Traditional public fora include sidewalks, squares, and public parks: [S]treets and parks . . . have immemorially been held in trust for the use of the public and, time out of mind, have been used for purposes of assembly, communicating thoughts between citizens, and discussing public questions. Such use of the streets and public places has, from ancient times, been a part of the privileges, immunities, rights, and liberties of citizens.

Hague, 307 U.S. at 515. "In these quintessential public forums, . . . [f]or the State to enforce a content-based exclusion it must show that its regulation is necessary to serve a compelling state interest and that it is narrowly drawn to achieve that end." Perry Educ. Ass'n v. Perry Local Educs. Ass'n, 460 U.S. 37, 45 (1983); see also Int'l Soc'y for Krishna Consciousness, 505 U.S. at 678 ("[R]egulation of speech on government property that has traditionally been available for public expression is subject to the highest scrutiny."); Frisby, 487 U.S. at 480 ("[W]e have repeatedly referred to public streets as the archetype of a traditional public forum."). A second category of fora, known as designated (or limited) public fora, "consists of public property which the State has opened for use by the public as a place for expressive activity." Perry, 460 U.S. at 46. Whereas any content-based restriction on the use of traditional public fora is subject to strict scrutiny, the state is generally permitted, as long as it does not discriminate on the basis of viewpoint, to limit a designated public forum to certain speakers or the discussion of certain subjects. See Perry, 460 U.S. at 45 n.7. Once it has defined the limits of a designated public forum, however, "[r]egulation of such property is subject to the same limitations as that governing a traditional public forum." Int'l Soc'y for Krishna Consciousness, 505 U.S. at 678. Examples of designated fora include university meeting facilities, see Widmar v. Vincent, 454 U.S. 263 (1981), school board meetings, see City of Madison Joint School Dist. v. Wisc. Employment Relations Comm'n, 429 U.S. 167 (1976), and municipal theaters, see Southeastern Promotions, Ltd. v. Conrad, 420 U.S. 546 (1975).

The third category, nonpublic fora, consists of all remaining public property. "Limitations on expressive activity conducted on this last category of property must survive only a much more limited review. The challenged regulation need only be reasonable, as long as the regulation is not an effort to suppress the speaker's activity due to disagreement with the speaker's view." Int'l Soc'y for Krishna Consciousness, 505 U.S. at 679. 2. Contours of the Relevant Forum: the Library's Collection as a Whole or the Provision of Internet Access?

To apply public forum doctrine to this case, we must first determine whether the appropriate forum for analysis is the library's collection as a whole, which includes both print and electronic resources, or the library's provision of Internet access. Where a plaintiff seeks limited access, for expressive purposes, to governmentally controlled property, the Supreme Court has held that the relevant forum is defined not by the physical limits of the government property at issue, but rather by the specific access that the plaintiff seeks: Although . . . as an initial matter a speaker must seek access to public property or to private property dedicated to public use to evoke First Amendment concerns, forum analysis is not completed merely by identifying the government property at issue. Rather, in defining the forum we have focused on the access sought by the speaker. When speakers seek general access to public property, the forum encompasses that property. In cases in which limited access is sought, our cases have taken a more tailored approach to ascertaining the perimeters of a forum within the confines of the government property. Cornelius v. NAACP Legal Def. & Educ. Fund, Inc., 473 U.S. 788, 801 (1985).

Thus, in Cornelius, where the plaintiffs were legal defense and political advocacy groups seeking to participate in the Combined Federal Campaign charity drive, the Court held that the relevant forum, for First Amendment purposes, was not the entire federal workplace, but rather the charity drive itself. Id. at 801. Similarly, in Perry Education Association v. Perry Local Educators' Association, 460 U.S. 37 (1983), which addressed a union's right to access a public school's internal mail system and teachers' mailboxes, the Court identified the relevant forum as the school's mail system, not the public school as a whole. In Widmar v. Vincent, 454 U.S. 263 (1981), in which a student group challenged a state university's restrictions on use of its meeting facilities, the Court identified the relevant forum as the meeting facilities to which the plaintiffs sought access, not the state university generally. And in Christ's Bride Ministries, Inc. v. SEPTA, 148 F.3d 242 (3d Cir. 1998), involving a First Amendment challenge to the removal of advertisements from subway and commuter rail stations, the Third Circuit noted that the forum at issue was not the rail and subway stations as a whole, but rather the advertising space within the stations. Id. at 248. Although these cases dealt with the problem of identifying the relevant forum where speakers are claiming a right of access, we believe that the same approach applies to identifying the relevant forum where the parties seeking access are listeners or readers.

In this case, the patron plaintiffs are not asserting a First Amendment right to compel public libraries to acquire certain books or magazines for their print collections. Nor are the Web site plaintiffs claiming a First Amendment right to compel public libraries to carry print materials that they publish. Rather, the right at issue in this case is the specific right of library patrons to access information on the Internet, and the specific right of Web publishers to provide library patrons with information via the Internet. Thus, the relevant forum for analysis is not the library's entire collection, which includes both print and electronic media, such as the Internet, but rather the specific forum created when the library provides its patrons with Internet access. Although a public library's provision of Internet access does not resemble the conventional notion of a forum as a well- defined physical space, the same First Amendment standards apply. See Rosenberger v. Rector & Visitors of Univ. of Va., 515 U.S. 819, 830 (1995) (holding that a state university's student activities fund "is a forum more in a metaphysical than a spatial or geographic sense, but the same principles are applicable"); see also Cornelius, 473 U.S. at 801 (identifying the Combined Federal Campaign charity drive as the relevant unit of analysis for application of public forum doctrine). 3. Content-based Restrictions in Designated Public Fora

Unlike nonpublic fora such as airport terminals, see Int'l Soc'y for Krishna Consciousness, Inc. v. Lee, 505 U.S. 672 (1992), military bases, see Greer v. Spock, 424 U.S. 828 (1976), jail grounds, see Adderley v. Florida, 385 U.S. 39 (1966), the federal workplace, see Cornelius v. NAACP Legal Def. & Educ. Fund, 473 U.S. 788, 805 (1985), and public transit vehicles, see Lehman v. City of Shaker Heights, 418 U.S. 298 (1974), the purpose of a public library in general, and the provision of Internet access within a public library in particular, is "for use by the public . . . for expressive activity," Perry Educ. Ass'n v. Perry Local Educs. Ass'n, 460 U.S. 37, 45 (1983), namely, the dissemination and receipt by the public of a wide range of information. We are satisfied that when the government provides Internet access in a public library, it has created a designated public forum. See Mainstream Loudoun v. Bd. of Trustees of the Loudoun County Library, 24 F. Supp. 2d 552, 563 (E.D. Va. 1998); cf. Kreimer v. Bureau of Police, 958 F.2d 1242, 1259 (3d Cir. 1992) (holding that a public library is a limited public forum). Relying on those cases that have recognized that government has leeway, under the First Amendment, to limit use of a designated public forum to narrowly specified purposes, and that content-based restrictions on speech that are consistent with those purposes are subject only to rational basis review, the government argues for application of rational basis review to public libraries' decisions about which content to make available to their patrons via the Internet. See Rosenberger, 515 U.S. 819, 829 (1995) ("The necessities of confining a forum to the limited and legitimate purposes for which it was created may justify the State in reserving it for certain groups or for the discussion of certain topics."); Perry, 460 U.S. at 46 n.7 (1983) ("A public forum may be created for a limited purpose such as use by certain groups . . . or for the discussion of certain subjects.").

In particular, the government forcefully argues that a public library's decision to limit the content of its digital offerings on the Internet should be subject to no stricter scrutiny than its decisions about what content to make available to its patrons through the library's print collection. According to the government, just as a public library may choose to acquire books about gardening but not golf, without having to show that this content-based restriction on patrons' access to speech is narrowly tailored to further a compelling state interest, so may a public library make content-based decisions about which speech to make available on the Internet, without having to show that such a restriction satisfies strict scrutiny. Plaintiffs respond that the government's ability to restrict the content of speech in a designated public forum by restricting the purpose of the designated public forum that it creates is not unlimited. Cf. Legal Servs. Corp. v. Velazquez, 531 U.S. 533, 547 (2001) ("Congress cannot recast a condition on funding as a mere definition of its program in every case, lest the First Amendment be reduced to a simple semantic exercise."). As Justice Kennedy has explained: If Government has a freer hand to draw content-based distinctions in limiting a forum than in excluding someone from it, the First Amendment would be a dead letter in designated public forums; every exclusion could be recast as a limitation. . . . The power to limit or redefine forums for a specific legitimate purpose does not allow the government to exclude certain speech or speakers from them for any reason at all. Denver Area Telecomm. Consortium, Inc. v. FCC, 518 U.S. 727, 801 (1996) (Kennedy, J., concurring in the judgment).

Although we agree with plaintiffs that the First Amendment imposes some limits on the state's ability to adopt content-based restrictions in defining the purpose of a public forum, precisely what those limits are is unclear, and presents a difficult problem in First Amendment jurisprudence. The Supreme Court's "cases have not yet determined . . . that government's decision to dedicate a public forum to one type of content or another is necessarily subject to the highest level of scrutiny. Must a local government, for example, show a compelling state interest if it builds a band shell in the park and dedicates it solely to classical music (but not to jazz)? The answer is not obvious." Denver, 518 U.S. at 750 (plurality opinion); see also Southeastern Promotions, Ltd. v. Conrad, 420 U.S. 546, 572-73 (1975) (Rehnquist, J., dissenting) ("May an opera house limit its productions to operas, or must it also show rock musicals? May a municipal theater devote an entire season to Shakespeare, or is it required to book any potential producer on a first come, first served basis?"). We believe, however, that certain principles emerge from the Supreme Court's jurisprudence on this question. In particular, and perhaps somewhat counterintuitively, the more narrow the range of speech that the government chooses to subsidize (whether directly, through government grants or other funding, or indirectly, through the creation of a public forum) the more deference the First Amendment accords the government in drawing content-based distinctions.

At one extreme lies the government's decision to fund a particular message that the government seeks to disseminate. In this context, content-based restrictions on the speech that government chooses to subsidize are clearly subject to at most rational basis review, and even viewpoint discrimination is permissible. For example, "[w]hen Congress established a National Endowment for Democracy to encourage other countries to adopt democratic principles, 22 U.S.C. Sec. 4411(b), it was not constitutionally required to fund a program to encourage competing lines of political philosophy such as communism and fascism." Rust v. Sullivan, 500 U.S. 173, 194 (1991); see also Velazquez, 531 U.S. at 541 ("[V]iewpoint-based funding decisions can be sustained in instances in which the government is itself the speaker, or in instances, like Rust, in which the government used private speakers to transmit information pertaining to its own program.") (internal quotation marks and citation omitted). Although not strictly controlling, the Supreme Court's unconstitutional conditions cases, such as Rust and Velazquez, are instructive for purposes of analyzing content-based restrictions on the use of public fora. This is because the limitations that government places on the use of a public forum can be conceptualized as conditions that the government attaches to the receipt of a benefit that it offers, namely, the use of government property. Public forum cases thus resemble those unconstitutional conditions cases involving First Amendment challenges to the conditions that the state places on the receipt of a government benefit. See Velazquez, 531 U.S. at 544 ("As this suit involves a subsidy, limited forum cases . . . may not be controlling in the strict sense, yet they do provide some instruction.").

Even when the government does not fund the dissemination of a particular government message, the First Amendment generally permits government, subject to the constraints of viewpoint neutrality, to create public institutions such as art museums and state universities, dedicated to facilitating the dissemination of private speech that the government believes to have particular merit. Thus, in NEA v. Finley, 524 U.S. 569 (1998), the Court upheld the use of content-based restrictions in a federal program awarding grants to artists on the basis of, inter alia, artistic excellence. "The very assumption of the NEA is that grants will be awarded according to the artistic worth of competing applications, and absolute neutrality is simply inconceivable." Id. at 585 (internal quotation marks and citation omitted). Similarly, as Justice Stevens explained in his concurring opinion in Widmar v. Vincent, 454 U.S. 263 (1981), the First Amendment does not necessarily subject to strict scrutiny a state university's use of content-based means of allocating scarce resources, including limited public fora such as its meeting facilities: Because every university's resources are limited, an educational institution must routinely make decisions concerning the use of the time and space that is available for extracurricular activities. In my judgment, it is both necessary and appropriate for those decisions to evaluate the content of a proposed student activity. I should think it obvious, for example, that if two groups of 25 students requested the use of a room at a particular time one to view Mickey Mouse cartoons and the other to rehearse an amateur performance of Hamlet the First Amendment would not require that the room be reserved for the group that submitted its application first. Nor do I see why a university should have to establish a "compelling state interest" to defend its decision to permit one group to use the facility and not the other. Id. at 278 (Stevens, J., concurring in the judgment).

The more broadly the government facilitates private speech, however, the less deference the First Amendment accords to the government's content-based restrictions on the speech that it facilitates. Thus, where the government creates a designated public forum to facilitate private speech representing a diverse range of viewpoints, the government's decision selectively to single out particular viewpoints for exclusion is subject to strict scrutiny. Compare Rosenberger, 515 U.S. at 834 (applying heightened First Amendment scrutiny to viewpoint-based restrictions on the use of a limited public forum where the government "does not itself speak or subsidize transmittal of a message it favors but instead expends funds to encourage a diversity of views from private speakers"), with Finley, 524 U.S. at 586 ("In the context of arts funding, in contrast to many other subsidies, the Government does not indiscriminately encourage a diversity of views from private speakers.") (internal quotation marks and citation omitted). Similarly, although the government may create a designated public forum limited to speech on a particular topic, if the government opens the forum to members of the general public to speak on that topic while selectively singling out for exclusion particular speakers on the basis of the content of their speech, that restriction is subject to strict scrutiny. For instance, in City of Madison Joint School District No. 8 v. Wisconsin Employment Relations Commission, 429 U.S. 167 (1976), the Court held that where a school board opens its meetings for public participation, it may not, consistent with the First Amendment, prohibit teachers other than union representatives from speaking on the subject of pending collective-bargaining negotiations. See id. at 175 (noting that the state "has opened a forum for direct citizen involvement"); see also Ark. Educ. Television Comm'n v. Forbes, 523 U.S. 666, 680 (1998) (distinguishing, for purposes of determining the appropriate level of First Amendment scrutiny, a televised debate in which a public broadcasting station exercises editorial discretion in selecting participating candidates from a debate that has "an open-microphone format").

Finally, content-based restrictions on speech in a designated public forum are most clearly subject to strict scrutiny when the government opens a forum for virtually unrestricted use by the general public for speech on a virtually unrestricted range of topics, while selectively excluding particular speech whose content it disfavors. Thus, in Conrad, the Court held that a local government violated the First Amendment when it denied a group seeking to perform the rock musical "Hair" access to a general-purpose municipal theater open for the public at large to use for performances. See also Denver, 518 U.S. at 802 (Kennedy, J., concurring in the judgment) (suggesting that strict scrutiny would not apply to a local government's decision to "build[] a band shell in the park and dedicate[] it solely to classical music (but not jazz)," but would apply to "the Government's creation of a band shell in which all types of music might be performed except for rap music"). Similarly, in FCC v. League of Women Voters of Cal., 468 U.S. 364 (1984), the Court subjected to heightened scrutiny a federal program that funded a wide range of public broadcasting stations that disseminated speech on a wide range of subjects, where the federal program singled out for exclusion speech whose content amounted to editorializing. As the Court later explained: In FCC v. League of Women Voters of Cal., 468 U.S. 364 (1984) the Court was instructed by its understanding of the dynamics of the broadcast industry in holding that prohibitions against editorializing by public radio networks were an impermissible restriction, even though the Government enacted the restriction to control the use of public funds. The First Amendment forbade the Government from using the forum in an unconventional way to suppress speech inherent in the nature of the medium.

Velazquez, 531 U.S. at 543.

In sum, the more widely the state opens a forum for members of the public to speak on a variety of subjects and viewpoints, the more vulnerable is the state's decision selectively to exclude certain speech on the basis of its disfavored content, as such exclusions distort the marketplace of ideas that the state has created in establishing the forum. Cf. Velazquez, 531 U.S. at 544 ("Restricting LSC attorneys in advising their clients and in presenting arguments and analyses to the courts distorts the legal system by altering the traditional role of the attorneys in much the same way broadcast systems or student publication networks were changed in the limited forum cases . . . ."). Thus, we believe that where the state designates a forum for expressive activity and opens the forum for speech by the public at large on a wide range of topics, strict scrutiny applies to restrictions that single out for exclusion from the forum particular speech whose content is disfavored. "Laws designed or intended to suppress or restrict the expression of specific speakers contradict basic First Amendment principles." United States v. Playboy Entm't Group, Inc., 529 U.S. 803, 812 (2000); see also Denver, 518 U.S. at 782 (Kennedy, J., concurring in the judgment) (noting the flaw in a law that "singles out one sort of speech for vulnerability to private censorship in a context where content-based discrimination is not otherwise permitted"). Compare Forbes, 523 U.S. at 679 (holding that the state does not create a public forum when it "allows selective access for individual speakers rather than general access for a class of speakers") (emphasis added), with Police Dep't of the City of Chicago v. Mosley, 408 U.S. 92, 96 (1972) ("Selective exclusions from a public forum may not be based on content alone, and may not be justified by reference to content alone.") (emphasis added).

We note further that to the extent that the government creates a public forum expressly designed to facilitate the dissemination of private speech, opens the forum to any member of the public to speak on any virtually any topic, and then selectively targets certain speech for exclusion based on its content, the government is singling out speech in a manner that resembles the discriminatory taxes on the press that the Supreme Court subjected to heightened First Amendment scrutiny in Arkansas Writers' Project, Inc. v. Ragland, 481 U.S. 221 (1987), and Minneapolis Star & Tribune Co. v. Minnesota Commissioner of Revenue, 460 U.S. 575 (1983), which we explain in the margin. 4. Reasons for Applying Strict Scrutiny

1. Selective Exclusion From a "Vast Democratic Forum"

Applying these principles to public libraries, we agree with the government that generally the First Amendment subjects libraries' content-based decisions about which print materials to acquire for their collections to only rational review. In making these decisions, public libraries are generally free to adopt collection development criteria that reflect not simply patrons' demand for certain material, but also the library's evaluation of the material's quality. See Bernard W. Bell, Filth, Filtering, and the First Amendment: Ruminations on Public Libraries' Use of Internet Filtering Software, 53 Fed. Comm. L.J. 191, 225 (2001) ("Librarians should have the discretion to decide that the library is committed to intellectual inquiry, not to the satisfaction of the full range of human desires."). Thus, a public library's decision to use the last $100 of its budget to purchase the complete works of Shakespeare even though more of its patrons would prefer the library to use the same amount to purchase the complete works of John Grisham, is not, in our view, subject to strict scrutiny. Cf. NEA v. Finley, 524 U.S. 569 (1998) (subjecting only to rational basis review the government's decision to award NEA grants on the basis of, inter alia, artistic excellence). Nonetheless, we disagree with the government's argument that public libraries' use of Internet filters is no different, for First Amendment purposes, from the editorial discretion that they exercise when they choose to acquire certain books on the basis of librarians' evaluation of their quality. The central difference, in our view, is that by providing patrons with even filtered Internet access, the library permits patrons to receive speech on a virtually unlimited number of topics, from a virtually unlimited number of speakers, without attempting to restrict patrons' access to speech that the library, in the exercise of its professional judgment, determines to be particularly valuable. Cf. Rosenberger v. Rector & Visitors of Univ. of Va., 515 U.S. 819, 834 (1995) (applying strict scrutiny to viewpoint-based restrictions where the state "does not itself speak or subsidize transmittal of a message it favors but instead expends funds to encourage a diversity of views from private speakers"). See generally supra Section IV.C.

In those cases upholding the government's exercise of editorial discretion in selecting certain speech for subsidization or inclusion in a state-created forum, the state actor exercising the editorial discretion has at least reviewed the content of the speech that the forum facilitates. Thus, in Finley the NEA examined the content of those works of art that it chose to subsidize, and in Arkansas Educational Television Commission v. Forbes, 523 U.S. 666 (1998), the public broadcaster specifically reviewed and approved each speaker permitted to participate in the debate. See id. at 673 ("In the case of television broadcasting, . . . broad rights of access for outside speakers would be antithetical, as a general rule, to the discretion that stations and their editorial staff must exercise to fulfill their journalistic purpose and statutory obligations."); Finley, 524 U.S. at 586 ("The NEA's mandate is to make esthetic judgments, and the inherently content-based 'excellence' threshold for NEA support sets it apart from the subsidy at issue in Rosenberger which was available to all student organizations that were 'related to the educational purpose of the University . . . .'") (quoting Rosenberger, 515 U.S. at 824); see also Cornelius v. NAACP Legal Def. & Educ. Fund, 473 U.S. 788, 804 (1985) ("The Government's consistent policy has been to limit participation in the [Combined Federal Campaign] to 'appropriate' voluntary agencies and to require agencies seeking admission to obtain permission from federal and local Campaign officials. . . . [T]here is no evidence suggesting that the granting of the requisite permission is merely ministerial."). The essence of editorial discretion requires the exercise of professional judgment in examining the content that the government singles out as speech of particular value.

This exercise of editorial discretion is evident in a library's decision to acquire certain books for its collection. As the government's experts in library science testified, in selecting a book for a library's collection, librarians evaluate the book's quality by reference to a variety of criteria such as its accuracy, the title's niche in relation to the rest of the collection, the authority of the author, the publisher, the work's presentation, and how it compares with other material available in the same genre or on the same subject. Thus, the content of every book that a library acquires has been reviewed by the library's collection development staff or someone to whom they have delegated the task, and has been judged to meet the criteria that form the basis for the library's collection development policy. Although some public libraries use "approval plans" to delegate the collection development to third-party vendors which provide the library with recommended materials that the library is then free to retain or return to the vendor, the same principle nonetheless attains.

In contrast, in providing patrons with even filtered Internet access, a public library invites patrons to access speech whose content has never been reviewed and recommended as particularly valuable by either a librarian or a third party to whom the library has delegated collection development decisions. Although several of the government's librarian witnesses who testified at trial purport to apply the same standards that govern the library's acquisition of print materials to the library's provision of Internet access to patrons, when public libraries provide their patrons with Internet access, they intentionally open their doors to vast amounts of speech that clearly lacks sufficient quality to ever be considered for the library's print collection. Unless a library allows access to only those sites that have been preselected as having particular value, a method that, as noted above, was tried and rejected by the Westerville Ohio Public Library, see supra at 46-47, even a library that uses software filters has opened its Internet collection "for indiscriminate use by the general public." Perry Educ. Ass'n v. Perry Local Educs. Ass'n, 460 U.S. 37, 47 (1983). "[M]ost Internet forums including chat rooms, newsgroups, mail exploders, and the Web are open to all comers." Reno v. ACLU, 521 U.S. 844, 880 (1997).

The fundamental difference between a library's print collection and its provision of Internet access is illustrated by comparing the extent to which the library opens its print collection to members of the public to speak on a given topic and the extent to which it opens its Internet terminals to members of the public to speak on a given topic. When a public library chooses to carry books on a selected topic, e.g. chemistry, it does not open its print collection to any member of the public who wishes to write about chemistry. Rather, out of the myriad of books that have ever been written on chemistry, each book on chemistry that the library carries has been reviewed and selected because the person reviewing the book, in the exercise of his or her professional judgment, has deemed its content to be particularly valuable. In contrast, when a public library provides Internet access, even filtered Internet access, it has created a forum open to any member of the public who writes about chemistry on the Internet, regardless of how unscientific the author's methods or of how patently false the author's conclusions are, regardless of the author's reputation or grammar, and regardless of the reviews of the scientific community. Notwithstanding protestations in CIPA's legislative history to the contrary, members of the general public do define the content that public libraries make available to their patrons through the Internet. Any member of the public with Internet access could, through the free Web hosting services available on the Internet, tonight jot down a few musings on any subject under the sun, and tomorrow those musings would become part of public libraries' online offerings and be available to any library patron who seeks them out.

In providing its patrons with Internet access, a public library creates a forum for the facilitation of speech, almost none of which either the library's collection development staff or even the filtering companies have ever reviewed. Although filtering companies review a portion of the Web in classifying particular sites, the portion of the Web that the filtering companies actually review is quite small in relation to the Web as a whole. The filtering companies' harvesting process, described in our findings of fact, is intended to identify only a small fraction of Web sites for the filtering companies to review. Put simply, the state cannot be said to be exercising editorial discretion permitted under the First Amendment when it indiscriminately facilitates private speech whose content it makes no effort to examine. Cf. Bell, supra, at 226 ("[C]ourts should take a much more jaundiced view of library policies that block Internet access to a very limited array of subjects than they take of library policies that reserve Internet terminals for very limited use."). While the First Amendment permits the government to exercise editorial discretion in singling out particularly favored speech for subsidization or inclusion in a state-created forum, we believe that where the state provides access to a "vast democratic forum[]," Reno, 521 U.S. at 868, open to any member of the public to speak on subjects "as diverse as human thought," id. at 870, and then selectively excludes from the forum certain speech on the basis of its content, such exclusions are subject to strict scrutiny. These exclusions risk fundamentally distorting the unique marketplace of ideas that public libraries create when they open their collections, via the Internet, to the speech of millions of individuals around the world on a virtually limitless number of subjects.

A public library's content-based restrictions on patrons' Internet access thus resemble the content-based restrictions on speech subsidized by the government, whether through direct funding or through the creation of a designated public forum, that the Supreme Court has subjected to strict scrutiny, as discussed above in Section IV.C. Although the government may subsidize a particular message representing the government's viewpoint without having to satisfy strict scrutiny, see Rust v. Sullivan, 500 U.S. 173 (1991), strict scrutiny applies to restrictions that selectively exclude particular viewpoints from a public forum designed to facilitate a wide range of viewpoints, see Rosenberger v. Rector & Visitors of Univ. of Va., 515 U.S. 819 (1995). Similarly, although the state's exercise of editorial discretion in selecting particular speakers for participation in a state-sponsored forum is subject to rational basis review, see Ark. Educ. Television Comm'n v. Forbes, 523 U.S. 666 (1998), selective exclusions of particular speakers from a forum otherwise open to any member of the public to speak are subject to strict scrutiny, see City of Madison Joint School Dist. No. 8 v. Wis. Employment Relations Comm'n, 429 U.S. 167 (1976). And while the government may, subject only to rational basis review, make content-based decisions in selecting works of artistic excellence to subsidize, see NEA v. Finley, 524 U.S. 569 (1998), the Supreme Court has applied heightened scrutiny where the government opens a general-purpose municipal theater for use by the public, but selectively excludes disfavored content, see Southeastern Promotions, Ltd. v. Conrad, 420 U.S. 546 (1975), where the government facilitates the speech of public broadcasters on a virtually limitless number of topics, but prohibits editorializing, see FCC v. League of Women Voters of Cal., 468 U.S. 364 (1984), and where the government funds a wide range of legal services but restricts funding recipients from challenging welfare laws, see Legal Servs. Corp. v. Velazquez, 531 U.S. 533 (2001). Similarly, where a public library opens a forum to an unlimited number of speakers around the world to speak on an unlimited number of topics, strict scrutiny applies to the library's selective exclusions of particular speech whose content the library disfavors. 2. Analogy to Traditional Public Fora

Application of strict scrutiny to public libraries' use of software filters, in our view, finds further support in the extent to which public libraries' provision of Internet access promotes First Amendment values in an analogous manner to traditional public fora, such as sidewalks and parks, in which content-based restrictions on speech are always subject to strict scrutiny. The public library, by its very nature, is "designed for freewheeling inquiry." Bd. of Education v. Pico, 457 U.S. 853, 915 (1982) (Rehnquist, J., dissenting). As such, the library is a "mighty resource in the free marketplace of ideas," Minarcini v. Strongsville City Sch. Dist., 541 F.2d 577, 582 (6th Cir. 1976), and represents a "quintessential locus of the receipt of information." Kreimer v. Bureau of Police for Morristown, 958 F.2d 1242, 1255 (3d Cir. 1992); see also Sund v. City of Wichita Falls, 121 F. Supp. 2d 530, 547 (N.D. Tex. 2000) ("The right to receive information is vigorously enforced in the context of a public library . . . ."); cf. Int'l Soc'y for Krishna Consciousness, Inc. v. Lee, 505 U.S. 672, 681 (1992) ("[A] traditional public forum is property that has as 'a principal purpose . . . the free exchange of ideas.'") (quoting Cornelius v. NAACP Legal Def. & Educ. Fund, 473 U.S. 788, 800 (1985)).

We acknowledge that the provision of Internet access in a public library does not enjoy the historical pedigree of streets, sidewalks, and parks as a vehicle of free expression. Nonetheless, we believe that it shares many of the characteristics of these traditional public fora that uniquely promote First Amendment values and accordingly warrant application of strict scrutiny to any content-based restriction on speech in these fora. Regulation of speech in streets, sidewalks, and parks is subject to the highest scrutiny not simply by virtue of history and tradition, but also because the speech-facilitating character of sidewalks and parks makes them distinctly deserving of First Amendment protection. Many of these same speech-promoting features of the traditional public forum appear in public libraries' provision of Internet access. First, public libraries, like sidewalks and parks, are generally open to any member of the public who wishes to receive the speech that these fora facilitate, subject only to narrow limitations. See Kreimer, 958 F.2d at 1260 (noting that a public library does not retain unfettered discretion "to choose whom it will permit to enter the Library," but upholding the library's right to exclude patrons who harass patrons or whose offensive personal hygiene precludes the library's use by other patrons). Moreover, like traditional public fora, public libraries are funded by taxpayers and therefore do not charge members of the public each time they use the forum. The only direct cost to library patrons who wish to receive information, whether via the Internet or the library's print collection, is the time spent reading.

By providing Internet access to millions of Americans to whom such access would otherwise be unavailable, public libraries play a critical role in bridging the digital divide separating those with access to new information technologies from those that lack access. See generally National Telecommunications and Information Administration, U.S. Department of Commerce, Falling Through the Net: Defining the Digital Divide (1999), available at http://www.ntia.doc.gov/ntiahome/fttn99/contents.html. Cf. Velazquez, 531 U.S. at 546 (invalidating a content-based restriction on the speech of federally funded legal services corporations and noting that given the financial hardship of legal services corporations' clients, "[t]he restriction on speech is even more problematic because in cases where the attorney withdraws from a representation, the client is unlikely to find other counsel"). Public libraries that provide Internet access greatly expand the educational opportunities for millions of Americans who, as explained in the margin, would otherwise be deprived of the benefits of this new medium.

Just as important as the openness of a forum to listeners is its openness to speakers. Parks and sidewalks are paradigmatic loci of First Amendment values in large part because they permit speakers to communicate with a wide audience at low cost. One can address members of the public in a park for little more than the cost of a soapbox, and one can distribute handbills on the sidewalk for little more than the cost of a pen, paper, and some photocopies. See Martin v. City of Struthers, 319 U.S. 141, 146 (1943) ("Door to door distribution of circulars is essential to the poorly financed causes of little people."); Laurence H. Tribe, American Constitutional Law Sec. 12-24 at 987 (2d ed. 1988) ("The 'public forum' doctrine holds that restrictions on speech should be subject to higher scrutiny when, all other things being equal, that speech occurs in areas playing a vital role in communication such as in those places historically associated with first amendment activities, such as streets, sidewalks, and parks especially because of how indispensable communication in these places is to people who lack access to more elaborate (and more costly) channels."); Daniel A. Farber, Free Speech without Romance: Public Choice and the First Amendment, 105 Harv. L. Rev. 554, 574 n.86 (1991) (noting that traditional public fora "are often the only place where less affluent groups and individuals can effectively express their message"); Harry Kalven, Jr., The Concept of the Public Forum: Cox v. Louisiana, 1965 Sup. Ct. Rev. 1, 30 ("[T]he parade, the picket, the leaflet, the sound truck, have been the media of communication exploited by those with little access to the more genteel means of communication."). Similarly, given the existence of message boards and free Web hosting services, a speaker can, via the Internet, address the public, including patrons of public libraries, for little more than the cost of Internet access. As the Supreme Court explained in Reno v. ACLU, 521 U.S. 844 (1997), "the Internet can hardly be considered a 'scarce' expressive commodity. It provides relatively unlimited, low-cost capacity for communication of all kinds." Id. at 870. Although the cost of a home computer and Internet access considerably exceeds the cost of a soapbox or a few hundred photocopies, speakers wishing to avail themselves of the Internet may gain free access in schools, workplaces, or the public library. As Professor Lessig has explained: The "press" in 1791 was not the New York Times or the Wall Street Journal. It did not comprise large organizations of private interests, with millions of readers associated with each organization. Rather, the press then was much like the Internet today. The cost of a printing press was low, the readership was slight, and anyone (within reason) could become a publisher and in fact an extraordinary number did. When the Constitution speaks of the rights of the "press," the architecture it has in mind is the architecture of the Internet.