Peer to Peer Opportunities:
Keeping an Open Mind on File-Sharing Networks
by Jamal Le Blanc (jamal@benton.org)

Introduction

Earlier this year, the House Minority Staff Special Investigations Division conducted a special investigation of Internet file-sharing applications at the request of Representatives Henry A. Waxman (D-CA) and Steve Largent (R-OK). The point of the investigation was to ascertain whether, and to what degree, pornography was available to minors through peer-to-peer file-sharing applications. The result of that investigation is a report titled Children's Access to Pornography Through Internet File-Sharing Programs.

While the report is useful in warning parents of the potential for exposure to adult materials through programs such as Music City Morpheus and Kazaa, its overall utility is limited. The report and its accompanying publicity constructs too narrowly the questions one might ask about Internet file-sharing applications and peer-to-peer networking. Instead of presenting Waxman, Largent and the public with a fully realized portrait of file-sharing applications, it feeds into a checkered congressional record regarding Internet pornography, online censorship and the welfare of children versus all other competing concerns. This article is an attempt to frame the concerns expressed by Reps. Waxman and Largent, commissioners of the report, within the broader developing sphere of Internet file-sharing.

Worthy Goals, Competing Interests

Regarding matters of speech, legislators have long criss-crossed the line between attempting to protect the public's safety and effectively curtailing the civil liberties that the public enjoys. Too often legislators have erred on the side of safety, only to find their well-intentioned, but over-reaching laws struck down after judicial review. The Communications Decency Act (CDA), Communications Online Protection Act (COPA) and Children's Internet Protection Act (CIPA) are examples of this point. To varying degrees, all have met with court challenges specifically highlighting how they run contrary to online free speech. The most recent piece of anti-porn legislation, CIPA, is currently being challenged by the American Library Association.

Given its history of pornography-related Internet legislation and the popularity of file-sharing programs such as Napster, it is not surprising that Congress would eventually turn its attention to file-sharing applications. Indeed, Children's Access to Pornography Through Internet File-Sharing Programs is as much the result of Napster's demise as it is the product of congressional inquiry. When Napster ceased operating, its 50 million former users began searching for alternative file-sharing services. The free services that they found were essentially a second generation of popular peer-to-peer networking services with capabilities beyond Napster's limited trading of MP3s. These new services allow users to not only trade music, but also to exchange any type of computer data -- from text documents to music to video files.

What the Special Investigations Division found was what the former Napster users already knew: many of the files available on these networks contained pornographic material. And although there are not hard numbers, the Special Investigations Division's research repeatedly turned up anecdotal evidence that teens and children made up a significant portion of the user groups for these file-sharing programs. The report quotes a December 2000 Pew Internet and American Life study that found that 53% of Internet-using teens ages 12-17 had downloaded music from the Internet -- a figure that amounts to around seven million children. The Special Investigations Division also noted that searches run on such familiar names as Britney Spears, turned up explicit text, image, and video files.

The prevalence of pornography in search results and the potentially high numbers of children making use of these programs, combined with current filtering software's inability to filter file-sharing programs, have caused much concern for Reps. Waxman and Largent. The question now, therefore, is whether this concern will lead to yet another unconstitutional overreach in the name of protecting children.

Peer-to-Peer Networking in a Nutshell:
Membership Has its Privileges

Napster, an online service that allowed users to exchange music files through proprietary software, is probably the most widely recognized of the myriad file-sharing services that exist. Directly responsible for the rise of a new class of consumer electronic device -- the portable MP3 player -- the Napster service has alternately been called revolutionary and criminal.

Indeed, Napster's legal troubles with the Recording Industry Association of America has made the company a household name. And while Napster's difficulties stemmed from the copyright implications of allowing an estimated 50 million users to trade copyrighted music, its troubles and eventual court loss are an important starting point in grasping the potential of peer-to-peer networking.

Unlike Napster, which made use of a central database to coordinate the searches of its members, many of the newer file-sharing programs (most notably Gnutella) do not possess a single central server. Instead, these second-generation file-sharing programs make use of true "peer-to-peer" communications: any computer with an Internet connection can communicate with any other Internet-connected computer without the intervention of an intermediary server. This subtle, yet important, difference makes peer-to-peer networking -- and concerns over the content traded in a peer-to-peer file-sharing network -- a very different animal than the ones Congress has previously dealt with. Because of the distributed network structure of a peer-to-peer file-sharing network, any legislation of such a structure would have to take into account individual user responsibilities for content in the network; the responsibility of the users' ISPs for content traded between users without the mediation of ISP servers; privacy and free speech concerns; and the potential public interest applications of a peer-to-peer networks

So what is peer-to-peer file sharing really all about? Traditionally, the Internet has operated on what is known as a "client-server" model. An Internet user's computer, whether at home, at work or in school, acts as a "client" in an Internet transaction when it sends a request to another Internet-connected computer -- a "server" -- for information. Servers are typically, but not always, computers run by Internet Service Providers (ISPs) with dedicated connections to the Internet that host Web sites, email applications and so forth. (They are also often the targets of Internet worm and virus attacks, as has been the case with the recent Code Red worms.) When a server receives a client computer's request for information, such as a Web page, it then checks to see if it is hosting that piece of information. If the information is located, the computer hosting the information "serves" the information back to the requesting computer. Hence, the name client-server.

While there is two-way communication between clients and servers, the line of demarcation is clear -- clients request, but do not host information, while servers host information, but do not request it. In comparison, the peer-to-peer technology used by file-sharing programs such as Napster or Gnutella breaks down the traditional client-server relationship. In a peer-to-peer network, computers that might have once been seen only as clients take on the role of servers as well. So when one computer in a peer-to-peer network tries to search for information, it has the ability to interact with all the other individual computers in the network rather than relying on a single server for tracking down the information it needs. On a Gnutella-based network, for example, a user interested in locating an MP3 file would use Computer #1 to query 10 other computers on the network (let's call them Computers #2 through #11) to see if any of them had a copy of the MP3 file in question. If one of those computers had the it would inform Computer #1 and makes the file accessible to it. But if computers #2 through #11 don't have the MP3 file, they each query 10 other computers in the network to see if those might have a copy of it. Each computer in the network contributes to the effort, searching its own archive of files then passing on the request to other computers if it can't find what Computer #1 is looking for. When the MP3 file is located on Computer #15 (or on Computer #150, #1,000, or #1,000,000 for that matter) Computer #15 would be given instructions on how to link up directly with Computer #1 and pass on the MP3 file to it.

Napster, the service that kicked off the peer-to-peer revolution, was in reality only a semi-peer-to-peer network, since computers participating in the network would search for the locations of MP3 files on Napster's centralized server before communicating with the computers in the network that actually hosted the music files. Many of the new peer-to-peer programs, including Gnutella, differ from Napster because there is not one central entity that is responsible for transactions over the network. These new programs are totally independent of centralized control. Any computer user who downloads and installs the peer-to-peer software on their computer instantly becomes part of that file-sharing community.

Grasping Peer-to-Peer's Potential

The potential for peer-to-peer networking is yet to be realized. To date there have been only hints to its public interest potential. But what is apparent is that peer-to-peer networks allow communities of users to exchange information in new and creative ways independently of traditional information hierarchies. The potential for use in organizing, advocacy, and information sharing is enormous. To date however, file-sharing networks have been put to less-than-civic uses, raising the specter of regulatory intervention and other complicating factors. One of those complicating factors can be seen in the challenges to Napster (and now Kazaa) from the Recording Industry Association of America and various Hollywood studios. These industry interests have alleged that file-sharing networks are used to pirate copyrighted media works. This debate is far from settled, of course, but it should not stifle the identification and development of public interest applications. Peer-to-peer networks have much potential in the public interest arena, and should be considered a proving ground for enabling noncommercial communities of users to exchange information in new and creative ways.

In terms of public interest applications, Napster was, in a sense, a type of groupware: software that allows for remote collaboration between individuals in different locations. Ray Ozzie, the creator of Lotus Notes, recently commented in Technology Review:

"From the personal perspective, "online collaboration," or knowledge work with others toward a common objective, facilitated by technology, is becoming the rule as opposed to the exception... From a personal perspective, our goal has been to create an environment that securely brings together the right people, the relevant information, the appropriate tools to manipulate that information, at the right time -- whether spontaneously or over a long period of time. A multimedia, multi-temporal tool that is so natural and easy to use that interactions naturally migrate into it, toward the goal of more productive and effective interaction."

Ozzie has developed a proprietary application called Groove (www.groove.net). Groove is a peer-to-peer application that enables some of the best parts of peer-to-peer networking and file sharing: the exchange of images and files; text chat; voice chat; threaded discussion; drawing, co-browsing of the Web and shared calendaring. One module of Groove is specifically designed to allow a sports coach to develop a schedule for a team and share that information with all of the players.

While Groove is proprietary and specifically designed for groupware uses, even the current, free file-sharing services have the potential for enabling the equivalent of a Wide Area Network among its users. For nonprofit and mission-driven organizations, the ability to have a reliable network without the capital outlay costs could lead to greater levels of collaboration and organizing.

Unfortunately, Children's Access to Pornography Through Internet File-Sharing Programs does not communicate this potential of peer-to-peer networks. Indeed the report does not attribute a single positive thing to file-sharing networks. Admittedly, the identification of the potential of file-sharing networks was not the primary focus of the researchers. However, the document was commissioned to inform members of Congress so that they could best act in the public's interest. As written, it is difficult to see how this report will help them do so. Currently, the lack of balance does a disservice to anyone hoping to reach a wise and informed decision.

©2006 Consortium for School Networking. All rights reserved.