Unless you've been living under a rock, you've heard about Napster, and the peer-to-peer file sharing process that revived interest in digital music. Unfortunately, most of what you've heard in the mainstream press on this issue only presents one point of view: much of what is being said is often quoted verbatim from recording industry sources. And much of it is either an oversimplification of the complexities of this issue or bald-faced lies. This misinformation is being promulgated by the recording industry's media hype machinery in an attempt to use the cry of "piracy" to gain public sentiment to bolster their attempts to steal digital intellectual property rights from consumers. The industry has always coveted these rights, but has never legally or physically enjoyed them under their traditional distribution systems based on the production and distribution of physical storage media (tapes, records & CDs). Should they be successful, it will be at the expense of the traditional rights purchasers of intellectual property have enjoyed since the founding of this nation. Conversely, if they fail in this attempt, artists and consumers stand to gain much, and the recording industry, as it now stands, will cease to exist! That threat of extinction is at the heart of the recording industry's assault on our freedoms.
An informed public, aware of what they and their progeny have to lose if the recording industry wins, can turn the tide of this battle. Accordingly, in an attempt to illuminate the facts in this complex issue, this article will discuss the basics of peer-to-peer file sharing; the creation, amazing growth, and death of Napster, and the renewed interest in music it created; a brief overview of the RIAA lawsuit; the open-source and serverless file-sharing systems Napster inspired; and the future challenges we face.
Peer to Peer. Despite all the recent hype in the press, peer-to-peer file sharing (shortened to the acronym P2P by us techies!) is not a new concept. The very earliest computer networks were all P2P implementations, where first mainframe and then personal computers were networked together to share files amongst themselves without any centralized control. The need by academics and scientists to share data and documents drove the creation of the very first attempts at networking. However, it was quickly found that, as the number of connected computers on a network grew, it took longer and longer for one computer to find another, and the "client-server" networks that are most popular today evolved: a central server (or servers!) manages where files are stored on the network and controls which computers and users can access them. This client-server networking paradigm became so popular that today, when we talk about a "network" in a corporate setting, the client-server model is usually assumed. But as the client computers, and the operating systems that ran on them, became more powerful, it was still also possible to set up P2P connections between two or more systems on the same client-server network. Today, a typical corporate network has a very formalized client-server infrastructure, controlled by corporate networking and security offices, but it also has numerous P2P connections between client computers, controlled by the individual computer users. Both networks ride on the same wires. So not only is P2P not new, it has been in wide use for some time, since the very beginnings of networking, and is in daily use today in corporate offices worldwide.
As local peer-to-peer networks evolved into client-server networking, a parallel idea was being developed, again being driven by academics and scientists, to connect local networks together over long distances, to allow geographically-separated research centers to collaborate on projects. Funded by the US Department of Defense's Advanced Research Projects Agency (DARPA), this network began as a limited number of dedicated conduits connecting key research centers in universities, government agencies and defense contractors. As this network grew in size and complexity, its use was expanded to encompass more government agencies, schools, and corporations until it became what we now call the Internet. The Internet has created powerful new opportunities for P2P because, where once, except under test conditions, P2P connections were limited to two computers on a local network, now any two Internet-connected computers anywhere in the world can find and talk to each other. If they can find each other, that is.
Napster. Enter Shawn Fanning. In January 1999, Shawn was a freshman at Northeastern University in Boston, Massachusetts, and like most college students at that time had discovered MP3 Audio. But he was frustrated that most attempts at finding songs he was looking for out on the Web usually ended in failure. His idea was simple but revolutionary: everyone out there probably has a few files that someone else wants, so figure out a way to help two computer users find each other, share what they have, and chat about their favorite music. On the face of it, this seemed like a harmless, perfectly legal concept: after all the Audio Home Recording Act of 1992 made the noncommercial sharing of music between friends legal. Obsessed with the idea, he quit school and worked almost every waking minute programming a Windows client program and a Unix server program that would become the core components of the Napster file sharing system, named after his high school nickname. And the Napster music community was born.
To be perfectly correct, Napster is not a pure P2P implementation. While all the files being shared reside on the individual users' computers, and the actual file sharing and chat works over direct P2P connections, a server is used to store an index of what files are where, to speed up access. However, the Napster server is just that, a list server, nothing more. Community members connected to the system provide all the files being shared. (Note: for a detailed explanation of the Napster protocol, check out the Annotated decompiled source code for the Napster protocol. Non-programmers needn't waste their time!).
Within weeks of its release, Napster use took off (to put it mildly!). Fed at first by word of mouth between college students, then jumping into the corporate world as file-download sites like Tucows and Download.com began featuring it, news of Napster spread quickly over the Internet. And within a month, as noted in Shawn's bio on Napster.com's company page, " . . . a beta version of Napster was named 'Download of the Week' and received over 300,000 hits at Download.com". After another month or so, over one million music lovers world-wide belonged to the Napster music community. Napster quickly developed a Napster for Macintosh, and the Open Source software community followed right behind with a Linux Napster Client. There was even a Napster "clone" developed for BeOS, as well as several Java-based implementations that could run on most any OS that support Java.
From listening to the RIAA rhetoric, you would think that interest in music file sharing via Napster was leading to the death of CD sales. In fact, Napster spurred a renewed world-wide interest in music, and has been credited for an 8 percent increase in CD sales over the previous year! None the less, all this interest in MP3 file sharing spawned by Napster gained the attention of the recording industry, who had already begun its fight to control digital music by trying to get MP3 audio players outlawed (see my MP3 Audio page).
The RIAA lawsuit. The Recording Industry Association of America — a trade group which, it is important to remember, represents only the five largest recording companies (in case you didn't know, that's why the hundreds of smaller non-member record labels are called "independents") — filed suit against Napster, claming piracy. In its defense, Napster claimed that it was innocent, since no files actually resided on its servers, and that, in any event, noncommercial file sharing was legal under the Audio Home Recording Act of 1992.
I won't dwell on the details of this case. (Those who enjoy reading legal documents can find all they'd ever want to read about this case, the RIAA, Napster and FindLaw.com have put up links to all the legal documents submitted in this case.) But in briefest form, here's the results: a federal district-court judge in California, obviously wowed by the RIAA reported billions of dollars in damages due to piracy (an argument that paints as illegal or immoral file-copying and sharing activities that have been common practice for years), the judge ruled that Napster had to be shut down. Napster won a stay pending review by a three-judge appeals court panel, who ruled that the original ruling was to broad, and threw it back to the district-court judge to modify her ruling. The district-court judge then ruled that Napster had to block access to all files the RIAA identified as being their copyrighted works. As Napster began to attempt to comply with this nearly-impossible task, they also submitted a request to have their case heard by the entire appeals court. Meanwhile, Napster community members have been renaming files using intentional misspellings, wording shifts, etc., to allow their files to be seen past the filters. Napster finally lost their case and folded their operations.
OpenNap and other File-Sharing Services. While all the Napster hype was flying in the popular press, the open source software community was not standing idly by. Once the Napster protocol had been decompiled and analyzed, it wasn't very long before an open-source Napster server software solution was developed called OpenNap. This allowed users to stand up servers compatible with the Napster clients everyone already had. It also prompted the creation of a neat little server navigation tool called Napigator, that allows users to surf from server to server looking for files! So as the RIAA was busy trying to shut down Napster, OpenNap servers were fully open for business, and relatively immune to the RIAA's attacks, since they are all purely non-profit: unlike Napster, who has been trying to turn its service into a business, OpenNap is a strictly volunteer operation. The RIAA has made an attempt to go after the ISPs that host these servers, but under the Digital Millennium Copyright Act, ISPs aren't accountable for what it's customers use their Internet accounts for.
As if OpenNap weren't enough, other file sharing services quickly emerged, most of which are "serverless" systems that are pure P2P implementations. At first, the most popular of these were based on the open-source Gnutella protocol, which was developed by two Nullsoft programmers and then squashed by Nullsoft when the RIAA went after Napster. However, the protocol and beta software Nullsoft developers were working on had already been released onto the Internet, and quickly spawned new, fully-implemented software solutions, the best of which are called Bearshare and Limewire. Being serverless and totally noncommercial, Gnutella-based systems are a very tough nut for the RIAA to crack, forcing them — if they dare! — to go directly after the millions of music lovers that are the core of their customer base. (For a good overview of software and resources developing around the Gnutella protocol, visit Wego Systems' Gnutella site.)
Another interesting serverless file-sharing system to emerge is called the Free Network project, or FreeNet for short. It was developed in response to the RIAA's attacks on Napster, and broader attempts by content providers to limit free speech rights on the Internet. It was specifically designed to allow the "uncensorable dissemination of controversial information" between computers on the Internet, a goal far beyond the simple sharing of music files.
One of the more ingenious of the new peer-to-peer file sharing services to spring from the Napster idea was called Aimster. This service was unique in several ways: first, it integrates with instant messaging services — it was first developed for AOL Instant Messaging or AIM, hence the name, but was quickly adapted to work over ICQ, Yahoo Messenger and MSN — and used these services as its transport, rather than a separate dedicated file service. Secondly, it was primarily designed for sharing of files between you and members of your "buddy list," not the world in general like Napster, and all messages, searches, and file transfers are encrypted for privacy, so no one (including Aimster and the RIAA!) will know what you're sharing! Unfortunately Aimster was dealt a one-two puch: first by America Online (AOL), who sued them to stop the use of the name "AIM", which is AOL's registered trademark, which forced Aimster to change their name to Madster. Then the RIAA sued them under circumstances similar to the Napster case. They were recently ordered by a judge to shut down.
The latest and most-active Napster replacement is a shared system with software provided by Dutch software company KaaZa, West Indies software house Grockster, and US company Music City (who later developed their own Gnutella based system called Morpheus, which has since been shut down by RIAA legal action!). Like Gnutella, this shared system is serverless: the numerous users' computers on system serve files to each other. However, unlike Napster, which was limited to MP3 audio files, this system allows users to share any file. KaaZa, has quickly taken over the file sharing scene, and watchers note that they currently boast having twice the number of users that Napster did in its heyday. KaaZa recently sold the rights to the file-sharing system to Sharman Networks, a small company based in the pacific island nation of Vanuatu. The RIAA has been going after them to shut them down, and although a US judge ruled that they can stand trial in a California court , it is still unclear whether a US court can effectively enforce any judgement on a foreign entity!
It is interesting to note the logic used by the Appeals court panel in the Napster case, when it threw the case back to the district court to bar sharing of copyrighted files. That ruling was based in-part on the following idea (emphasis is mine): "Napster may be vicariously liable when it fails to affirmatively use its ability to patrol its system and preclude access to potentially infringing files listed in its search index. Napster has both the ability to use its search function to identify infringing musical recordings and the right to bar participation of users." In the view of legal experts, this language may well shield from prosecution serverless systems like Gnutella, Aimster and the shared KaZaa/Music City/Grokster system, which are incapable of determining what is being shared, and maintain no index of files available on the system. But we'll only know that if a case ever actually comes to trial. So far all leagl actions to close down file traders other than Napster have been pretrial injunctions, or little guys running scared from Big Media lawyers!
Challenges for the Future. The legal venue is not the only front in this war. The recording industry — and their pals in the motion picture industry who are similarly trying to hijack our rights vis-à-vis digital video, and have already been partially successful (see my DVD & DeCSS page!) — are trying to come up with ways to harness technology to digital media so they can control where, why, how many times, and for how long we can enjoy digital media once we buy it! The potential technologies that could be applied in this manner are legion, but we'll discuss here a few that, due to popularity or strong industry backing, harbor the most threat to our rights:
So how do we stop the technological hijacking of our Fair Use rights? Vigilance! Most tech news sources (including those listed on my Credits page) have e-mail newsletters you can subscribe to. Scan them regularly looking for attempts to apply technology to deprive you of your rights, and call or write your Congressmen and industry leaders, and let them know about your displeasure!
- SDMI and copy protection. In response to the explosive popularity of MP3 audio, the recording industry started up the Secure Digital Media Initiative, an attempt to apply copy protection to digital media, in an effort to wrest control of digital media from the masses (so they can continue to control the distribution of audio at our expense!). The idea is to use so-called "digital watermarking" to control what we do with the file once we own it. Before we discuss the potential this has to restrict fair use rights, we first have to review a bit of history to point out why copy protection is such a stupid idea! This is an example from the software industry, but the similarities dovetail very nicely in this instance, as you will see.
In the early and mid 1980's, software for PCs was extremely expensive: the single-unit cost for mainstream business applications like WordStar, Lotus 1-2-3 or dBase II was $300-500 (the equivalent of $700-1000 in today's market!), well out of reach of the typical PC hobbyist's pocketbook. As a result, although "deep-pocket" corporate accounts willingly paid the cost, software piracy was nearly universal among hobbiests. So the industry, perceiving lost profits from software piracy, began to copy-protect software rather than address the cause of the problem (overpriced software!). All sorts of ideas were tried, some of them quite ingenious: special disk encoding like intentional bad sectors the MS-DOS Diskopy command (all we had at first!) couldn't reproduce, modified boot sectors, custom "copy-proof" disk sector formatting schemes, etc. However, around the same time, the Supreme Court, in Sony Corp. v. Universal City Studios, Inc., 464 U.S. 417 (1984), reaffirmed that, in their own words, " Any individual may reproduce a copyrighted work for a 'fair use'; the copyright owner does not possess the exclusive right to such a use." This case was quickly embraced by rogue elements of the software industry to justify selling software copying utilities that could get around the copy protection schemes, under the justification of allowing legal archive copies to be made, and the piracy continued! Of course, this led the software industry to develop even more-inventive copy protection schemes, which led to even better copying utilities . . . you get the picture! In the process, the copy protection schemes being developed became so exotic that they would often result in software that would be subtly incompatible with some disk drives, leading to a revolt in the corporate market led by popular computer magazine journalists to boycott copy-protected software! As a result, to this day, copy protection is virtually unknown in the software industry, except for a few high-value vertical applications. And software vendors, driven primarily by the Shareware market, finally realized that if they cut the cost of their software they'd sell a lot more of it!
So what does this have to do with digital audio? Once again we have intellectual property that is being marketed in a manner that garners customer contempt, resulting in blatant "piracy" (the RIAA's view, not mine or the Audio Home Recording Act's!). As a result, rather than address the factors that lead to that contempt, the industry is attempting to use copy protection technology to protect the market they are mismanaging! In a Web article discussing this subject, noted technology guru Andreas Pfeiffer questions why, when copy protection was such a dismal failure for the software industry, both technically and popularly, the recording industry thinks it will work any better for them?! One quote he makes is particularly telling: "If users don't accept copy protection on a professional program, why would they do so for something priced at a few dollars?" Why indeed?! And just like with software, copy protection is doomed to failure since anything one person can invent another can figure out how to circumvent!
Which, by the way, is exactly what has happened to SDMI before it was even deployed: although the SDMI officials deny it, a group of collegiate researchers, responding to a $10,000 challenge from the SDMI, successfully circumvented the SDMI's "digital watermarking" technology!
Now that we've viewed a little history to show why copy protection is a waste of effort, we must return to the Fair Use arguments. If Sony Corp. v. Universal City Studios, Inc. applies here (and, the RIAA not withstanding, most intellectual property experts say it must!), then any technology that deprives the owner of a legally acquired digital file the right to make non-commercial copies is a violation of our Fair Use rights. And SDMI is specifically intended to create a system whereby the content provider can encode into the file whatever controls they choose (copy 4, 1 or no times; expiration dates after which the file is unusable; etc.), then program the controls into the playback mechanism itself to enforce the encoded rules! This post-sale control of the content is a gross violation of Fair Use rights!
However, what finally did the SDMI in was similar to why software copy protection failed: as Eric Scheirer, Technology Correspondent, MP3.com, said in his article titled The End of SDMI:
What the RIAA and its members failed to realize is that to make this story real, they had to keep the technology companies happy. The tech industry could see what was happening in the online music space, and they wanted to make sure they maintained the opportunity to make real businesses--hardware, software and services--from it.
So once the RIAA lost the technology industry, the standard was doomed.
- Digital Rights Management. Similar to copy protection, these are controls built into the file itself that control how the file is distributed. For example, in it's vendor's own words, ContentGuard, the most well-know product of the genre, can be used for "content protection and integrity, rights specification and enforcement, usage tracking, and interfaces with complementary solutions for reporting and royalty management." Let's analyze these concepts, in light of Fair Use:
- "Content Protection and Integrity." This means making the file impossible to modify, and the content impossible to edit. On the face of it, this sounds benign. However, when documents are digital, this could limit or make impossible the fair-use copying of portions of the content for non-commercial use, a common, legal practice in academia; and the increasingly common practice of "sampling," one artist using, with permission, snippets of other artist's works.
- "Rights Specification & Enforcement." Again, as with SDMI, the content provider would be able to encode into the file the rights they wish to allow you to have, regardless of the law, and use the technology to force you to live with them!
- "Usage Tracking." In other words, when you use the file, someone is notified! Can you say "privacy violation?" Even worse, this technology could be set by the publisher to limit the number of copies that could be made, down to and including none!
- "Royalty Management." This means "pay before play," i.e., tying the use of the file to proof of purchase in some manner. Or, even worse, requiring payment each time the file is used! Again, set by the publisher!
- Microsoft's Windows Media Rights Manager, used for setting rights into its WMA file specification, is another example of Digital Rights Management.
- Content Protection for Recordable Media (CPRM) and other content controls built into storage specifications. This is one of the scariest ideas put forth so far. Intel, IBM, Toshiba and Matsushita, calling themselves the "4C Entity,"; have developed a technology called Content Protection for Recordable Media, supposedly intended for only for removable media, that could prevent the copying of copyrighted files! However, this technology could easily find its way into hard drives as well, since it was submitted to the T13 group of the National Committee for Information Technology Standards (which sets standards for storage media like CD-ROMs, flash memory and hard drives), as a possible inclusion in the next-generation of the AT Attachment Interface, the so-called "EIDE" interface used to connect hard drives and CD-ROMs to the computer!
The biggest fear over this technology is that, by building it into the actual drive interface, it could be used to control what users can and cannot save on their computer! Which is what Big Media wants: if, as they see it, all content should be "pay per view," why should you be saving anything? The 4C Entity members have made numerous assurance that the technology will only be used to control removable media (which is bad enough!), but by building it into the EIDE interface itself, they have the capability to restrict what we can save to our hard drives as well. If they are given the capability to censor what we save, do you want your only protection against such censorship to be their assurance that they won't turn on a switch they'll already have? Neither do I!
Conclusion. Whether or not our children, grandchildren, and future generations of Americans retain the right to control what they listed to, how often they listen to it, and what they can do with it once they're finished with it depends in large part on the results of the recording industry's current attempts to use the media, the courts, influence in Congress, and technological excess to effectively invalidate the "fair use" rights granted to us by the US Constitution, laws, regulations, and case law developed throughout the history of this country. These rights allow the purchaser of any intellectual property the right to use it wherever and whenever they wish, as often as they wish; make copies and share them with others as long as no fee is charged; and, under a concept called the First Sale Doctrine, loan or rent them to others, and sell or give them away once they no longer want them. However, the industry, in the name of "stopping piracy" is attempting to ensure that, in the future, no one will be able to own music, only rent it every time they want to listed to it, by trying to kill off all forms of file sharing they don't control, while encouraging the expansion of "Internet radio stations" and other streaming media solutions that disappear from your computer after you play them. That way, since we don't own them, fair use laws don't apply! Our challenge is to see that they aren't successful. To do this, citizens need to become involved! You can do this in several ways:
- Write your state and federal Senators and Congressmen and let them know that you are unhappy with the way the recording industry has been attacking digital music distribution, and let them know you want to see our traditional Fair Use rights fully applied to digital media, both legally and technologically.
- Any time you read or hear members of the press quoting recording industry propaganda as fact, call them on it! Letters to the editor, phone calls to TV news reporters, and e-mail replies to network news or Web journalists to set them straight on the facts will hopefully steer some of them to reporting this issue from the consumer's point of view!