Tim Wu from Slate Magazine discusses the development of legal protection against copyright infringement for internet-based services. He argues that Youtube may be safer from copyright litigation than many might believe. He notes that “in the early 1990s…Hollywood and the recording industry worked hard to make Internet companies responsible for any…copyright infringement that happened via the Internet.” Had this view, the idea that internet companies are liable for any infringement that occurs on their sites, been adopted by Congress, few if any Web 2.0 companies would exist today. Fortunately, Lobbyists working for firms controlling a large portion of internet traffic worked to convince Congress that “copyright law threatens to put a damper on the expression of ideas on the internet.” This forced Hollywood to settle for the Digital Millennium Copyright Act, which protects “Information Residing on Systems or Networks At Direction of Users,” or in other words, sites using user-generated content. These sites are protected by the Safe Harbor clause in the DMCA which protects content sites and ISPs so long as they comply with takedown requests submitted by copyright holders. Wu notes that Youtube may be liable if it prosecutors can prove that it is aware of specific infringing material on its site. It may also be liable due to the fact that there is a search option on its services. Wu’s most interesting argument, however, is the idea that Hollywood and the recording industry may actually be fonder of the DMCA than previously believed. He states that these industries get the best of both worlds: if they do not want a clip on Youtube, they can simply issue a takedown notice and have the offending clip removed. If the infringing clip appears to be giving a boost to TV or music ratings, however, they can simply allow the clip to remain on Youtube. Thus, Youtube is protected by the idea of “tolerated use” rather than something like “fair use.” While sites like Grokster were not covered under the Safe Harbor law, Youtube is protected by this provision of the DMCA.
Wu accurately conveys the crucial role that the DMCA plays in protecting internet content sites and ISPs, which is a crucial argument in my paper. Had the DMCA never been created, Internet sites could be just as liable for copyright infringing content as stores are for selling physical pieces of copyright infringing material. However, the lawmakers clearly anticipated the importance of user-generated content for e-commerce and thus created a protection clause for sites like Youtube. While Wu mentions that Youtube may be liable for its “search” tools, a similar clause in the DMCA may equally protect Youtube from prosecution for this feature. 512D states that sites which allow users to search for material are not liable for any copyright infringing items that users locate using search options. Thus, the DMCA successfully anticipates many aspects of these Web 2.0 sites and prevents content industries from suing services like Youtube.
This article discusses the development of the DMCA and the reasons it was drafted. The author describes in depth the protection guaranteed by the Safe Harbor law before he begins to describe the development of the DMCA. Because I have discussed Safe Harbor elsewhere, I will focus on the part of the article detailing the creation of the DMCA, beginning in section C. In 1993, President Clinton created a task force designed to “fine tune” copyright laws to make them relevant in the digital world. A preliminary report ruled that ISPs be held wholly liable for infringement. ISPs immediately retaliated, claiming that under this law they would be liable whenever their users sent infringing works across their networks. They also argued that the law would require them to monitor their users’ transmissions while trying to detect copyrighted material, thus violating their users’ First Amendment Rights. Representative Rick Boucher argued that “Congress should provide ISPs with an incentive to invest in the development of computer networks without fear of litigation.” ISPs rationalized that they should only be liable if they have knowledge of the infringements. Copyright holders, however, argued that giving ISPs freedom from liability would encourage them to purposefully allow infringement on their servers. Because of the stark contrasts in opinion between the groups, a decision was not reached immediately. After months of negotiations between ISPS and Copyright Holders, the Online Copyright Infringement Liability Act was introduced in February 1997, which sought to protect ISPs from Direct Liability and monetary damages if they were not aware of “information indicating that the material is infringing.” In House Bill 3209, the Senate created the “red-flag test” for ISPs to use to determine infringement. ISPs did not have to monitor for infringement, but if they became aware of a “red-flag,” they would lose Safe Harbor protection unless they removed the content immediately.
This article is very important for my paper because it addresses the attitudes of both ISPs and Copyright Holders before the creation of the DMCA. Without the DMCA in place, both these groups were forced to come up with their own arguments to defend themselves, something ISPs are much less likely to do now that they can simply hide behind the Safe Harbor clause of the DMCA. The interesting aspect of this article, however, is that many, if not all, of the concerns of ISPs and Copyright Holders made their way into the finalized version of the DMCA. ISPs, such as Youtube today, were worried that they would violate their users’ rights if forced to monitor for copyright violations. Thus, the DMCA was written so that the burden fell on Copyright Holders to identify infringing material and report it to the ISPs. Similarly, the Copyright Holders were worried ISPs would turn a blind eye towards infringement, so the DMCA requires ISPs to remove infringing material at the request of the Holder. Thus, it is clear that the DMCA was created to appease both ISPs and Copyright Holders.
In early 2007, Stephanie Lenz recorded a video of her children dancing to the song “Let’s Go Crazy” recorded by Prince. She uploaded the recording to Youtube and, roughly three months later, received a takedown notice from Youtube notifying her that the video infringed on a copyright held by Universal Music. Lenz issued a complaint stating that the video was actually a Fair Use of Prince’s music and should therefore be put back onto Youtube. She said her video was not taken down based “on a particular characteristic of the video or any good-faith belief that it actually infringed a copyright,” but rather Prince’s personal desire to control all of his work. The plaintiffs in this case accept that the video includes elements that are under copyright by Prince and Universal. Their argument is whether or not the Digital Millennium Copyright Act “requires a copyright owner to consider the fair use doctrine in formulating a good faith belief that use of the material in the manner complained of is not authorized by the copyright owner, its agent, or the law.” The judge in this case noted that no other court case has actually determined the merits of whether the phrase quoted above pertains to Fair Use. The judge determined that, despite no previous ruling, Fair Use is not an infringement of copyright and is a lawful use of the copyright. The court thus ordered that a brief review of potentially infringing material must be completed by content owners prior to sending a Takedown notice, to ensure whether it is a Fair Use.
This decision strengthens my paper’s argument that many potentially infringing videos on Youtube may, in fact, be examples of Fair Use. While only a small percentage of songs available on file sharing websites could be constituted as Fair Use due to the skill required to sufficiently transform songs, many videos on Youtube may be shielded from unwarranted takedown notices because of this ruling, due to the fact that transforming and mashing video clips is much easier than transforming songs. The complaint that a large portion of Youtube’s videos are copyright infringing and that Youtube encourages such videos is thus proven false. In reality, many of these “infringing” videos actually make up the user-generated content that embodies the spirit of Youtube: a community of Web 2.0 users producing unique and individual content to share with others. Had this decision not been made, unchecked takedown notices could have harmed time-sensitive and important videos that were actually examples of Fair Use. While Universal argued that this checking for Fair Use is an unnecessary waste of time, the Judge was quick to point out that the Copyright Act of 1976 established 4 simple, quick factors for determining Fair Use. This decision upholds the hard work of individuals who successfully transform copyrighted material, and it prevents large corporations and recording artists from overreaching their bounds by unfairly removing Fair Use videos. Youtube’s legitimacy as a website made up of a majority of unique material is thus upheld.
The article extensively illustrates the development of Web 2.0 and the emergence of Youtube as one of the most popular websites on the internet. The author then summarizes Youtube’s liability protection under the Fair Harbor law. My interest in this article, however, stems from its discussion of the filtering software used by Youtube. “Youtube recently unveiled a video identification service which would create digital fingerprints of material that content providers wish to have protected.” If a video is uploaded to Youtube that matches the fingerprint of a copyrighted work, the owner can request that it be removed. Extensive tests have already been conducted: in one case, the system caught 18 instances of infringement after a service uploaded over 4400 hours of content to Youtube. After a copyright owner identifies infringing work, it can either have the material pulled or, even more incredibly, have its own advertisements added to the video. This technology is very appealing to Youtube because adopting it will show courts that it is doing all it can to remove copyrighted material. However, several factors make this protection unappealing. First, the “fingerprints” rely on a library of original content with which to match against infringing content. Thus, copyright owners will have to provide an extensive library of material to Youtube before being able to find their illegally uploaded material on Youtube. It is similarly unclear whether this technology will be able to identify slightly altered versions of original clips uploaded to the website. Fair Use advocates are equally concerned that the software will remove their own Fair Use works, mistaking them for infringing material.
This is an important article because it discusses Youtube as a company increasingly working for the Copyright holding companies rather than for its own users. Youtube is constantly in danger of copyright litigation: even the DMCA will not protect the company if plaintiffs can prove that Youtube is directly benefitting financially from copyrighted content. By signing deals with content owners that allow the owners to add advertisements to any of their content that was illegally uploaded, Youtube has cleverly created a way to profit from illegal content. Youtube also signed agreements with content owners to provide studio shows and clips on its services. This mitigates the temptation for users to upload illegal videos, especially if they can watch the legal version on the exact same website. However, by blindly implementing filtering software that automatically flags seemingly copyrighted material, Youtube may be dooming Fair Use works. Rather, Youtube should alter the filtering software so that it only flags videos that are either entirely made up of one video clip or contain a part of a copyrighted video with the corresponding audio from that clip playing as well. Many Fair Use artists will take the video but not the audio portion of a clip and mix it with other clips. Youtube can thus appease the studios and courts while still emphasizing the importance of its community of users, whom it built the website for in the first place.
Mark Cuban, creator of Broadcast.com and outspoken opponent of Youtube, directly compares Youtube to the original Napster website in this blog entry. He attributes Youtube’s quick success to two specific sources: “Free Hosting from any 3rd Party Site” and “Copyrighted music and video.” He goes on to make direct comparisons between Grokster, Napster, and Youtube. Napster was “the first to tell you it [pirating] wasn’t illegal.” He argues that the only reason Youtube hasn’t been brought to court multiple times already is that the studios are not sure what having so many clips available illegally means for them financially. Similarly to Napster, once the lawsuits begin, they will not stop until the service is forced to shut down. He observes that Youtube is remarkably similar to Napster, because users can simply open as many Youtube pages containing copyrighted songs as they want, and then listen to the songs as they would on Napster. Youtube will be hurt not just by lawsuits, but also by the wide availability of copyrighted content in legal online channels, such as NBC making clips available on its own site. Cuban states that as soon as Youtube is sued by copyright holders, it will be forced to find and remove all infringing content. This will leave the site, he argues, devoid of most appealing content.
While Cuban is correct in noting that there is a large amount of copyrighted material available on Youtube, he fails to take into account several key details. First, he states that Youtube will be sued for inducing others to commit infringement, just as Napster and Grokster were sued. Unlike Youtube, however, Napster and Youtube advertised themselves as sites which allowed users to download any music they wanted. They actually did induce users to visit the site for the purpose of downloading infringing material, whereas Youtube encourages users to visit its site to host user-generated content, evident from its slogan of “Broadcast Yourself.” Cuban also suggests that after copyrighted material such as TV shows is widely available in other locations and once copyright holders begin ordering their content to be removed, Youtube would be devoid of any content to set it apart from competitors. However, sites like Hulu, Joost, and services run by major Television studios have been online for over a year and Youtube is as popular as ever. This debunks the argument that Youtube would be unappealing once its copyright material was removed and other legal video-viewing services were established. Rather, users still visit the site for non-copyrighted material, and it continues to thrive, having just signed several deals itself with major content creators and TV Studios. Cuban’s main oversight is in the DMCA. He completely fails to take into account the fact that the DMCA Safe Harbor law removes Youtube from direct liability for any infringing videos that are posted on its service, so long as it removes them upon request of the copyright holder.
The article by Mr. Cloak discusses the evolution of online copyright cases in the pre-DMCA era. Before the DMCA, there was very little protection from liability for online user-generated services. The first case to test website liability in copyright infringement was Playboy Enterprises, Inc. v. Frena. Frena owned an online message board service that allowed its users to post messages and pictures to share with others. One user uploaded over 100 images copyrighted by Playboy. Playboy consequently sued Frena for copyright infringement, despite that fact that Frena had not actually uploaded the pictures himself. The court ruled that Frena was liable for the infringement, failing to note that Frena simply ran a service allowing others to upload images. The case of Religious Technology Center v. Netcom On-line Communications Services advanced the Frena ruling when the Church of Scientology sued Netcom for allowing unpublished Scientology writings to be posted on its services. The court judged that Netcom was not directly liable for this infringement because it had not directly facilitated the infringement. Rather, its software had automatically uploaded the infringing material without the knowledge of the operators. While Netcom was not liable for direct infringement, the court noted that it might be liable for another type of infringement. A website could be liable for Vicarious Liability if it receives direct financial benefit from the infringing material and has the ability to control infringement. Similarly, it could be guilty of Contributory Infringement if it has subjective knowledge of the infringement and substantially participates in the infringement.
The aforementioned types of liability have been mentioned in multiple copyright cases, such as the Napster case. With the advent of the DMCA and the Safe Harbor law, a service like Youtube is increasingly safe from direct liability. Youtube, however, could be held Vicariously Liable: it receives ad revenue in proportion to the number of users; if the number of copyrighted videos increases, its ad revenue will increase. Youtube has also developed filtering software, giving it the ability to control copyrighted material. For this reason, it could still theoretically be vicariously liable. Youtube could also be liable for Contributory Infringement because it has knowledge of infringement on its site (evident from the numerous takedown notices it receives). However, by demonstrating sufficient non-infringing uses of its services, like Betamax, it can escape Contributory Liability by being labeled a “staple article of commerce.” As Mr. Cloak states, “subjecting YouTube to liability from copyright owners could destroy a major facilitator of American creativity.” At the same time, authors have the right to control their works. Thus, the DMCA establishes a perfect medium by protecting online services from direct liability for allowing their users to upload anything, while also allowing copyright owners to request their work be removed if they wish. Subjecting Youtube to liability for simply allowing creativity to thrive on its servers should not be allowed, unless Youtube is proven to be guilt via vicarious or contributory means.
On March 13, 2007, Viacom International Inc. filed a class action lawsuit against Youtube claiming massive copyright infringement by the defendant. Viacom filed the suit after sending takedown notices to Youtube demanding over 150,000 copyrighted videos be removed from its servers. In its complaint, Viacom notes “millions have seized the opportunities digital technology provides to express themselves creatively.” However, Viacom argues that Youtube has “harnessed technology to willfully infringe copyrights on a huge scale.” Youtube, the complaint urges, has built a library of infringing video clips in order to increase profit. Rather than attempting to remove all infringing videos, Youtube “has decided to shift the burden entirely onto copyright owners to monitor the Youtube site…to detect infringing videos and send takedown notices to Youtube.” Viacom claims that Youtube increases its own value at the expense of copyright holders through the following methods: displaying advertisements above infringing videos, allowing users to embed infringing files onto other websites to draw users to Youtube and subsequently increase ad revenue, and permitting users to keep copyrighted videos hidden from the public. Viacom also notes that Youtube hosts the videos on its own servers, rather than simply acting as a conduit through which users pass files. This, in Viacom’s interpretation, makes Youtube the primary copyright infringer as it is the entity that is actually “performing” the copyrighted footage.
Youtube is one of the more influential websites in the development of Web 2.0. The website has essentially ushered in a new age of internet democratization by giving all users the ability to create and host content. Viacom’s complaint fails to take several important copyright issues into account, however, decreasing the lawsuit’s validity in several key issues. First and foremost, it assumes that Youtube has a clear intention of hosting copyright infringing content. While the court decided that Grokster, in MGM Studios v. Grokster, did not have sufficient non-infringing uses to escape liability, Youtube was developed as a website where average internet users can upload home videos. When asked about a memory associated with Youtube, users will typically discuss a humorous home movie they saw rather than an illegal movie clip. Similarly, Viacom assumes that Youtube is responsible for policing its site for all copyrighted material, failing to mention the DMCA once in the lawsuit. The Safe Harbor clause of the Digital Millennium Copyright Act, however, removes service providers from liability for any copyrighted material that users upload to their servers, specifically if the content provider removes material that a copyright holder insists is infringing. Youtube immediately removes material upon receipt of a takedown notice, typically without even ensuring that the entity which issued the notice is actually the copyright holder. Youtube is similarly protected by the Inducing Infringement of Copyrights Act, which protects sites which do not induce others to commit copyright infringement. Rather, Youtube encourages users to produce their own works.
Judge Stanton ruled in favor of Viacom in some aspects of his decision and in favor of Youtube in others. In favor of Youtube, he denied Viacom access to Youtube’s search code, noting that it is a trade secret that cost Youtube thousands of man-hours to produce and that it will not help Viacom determine the extent to which Youtube is liable. This decision came after numerous programming experts testified that there is currently no search code in existence with the ability to distinguish between copyrighted and non-copyrighted works. Similarly, the judge denied Viacom access to the Video ID Program. The judge also denied Viacom’s request for access to all videos currently available on the Youtube servers. Viacom claimed this would help them determine how much knowledge Youtube had relating to infringing videos, but Youtube’s response that they have been entirely accommodating to Viacom’s requests was favored by the judge. The judge stated that there is “no compelling need…to justify the analysis of millions of pieces of information.” The judge similarly denied access to the Advertising Schema, stating that this was both a trade secret and not necessary information. However, the judge favored with Viacom in many aspects, in an attempt to allow them to research how much power Youtube has over infringing videos on its website. He mandated that Youtube produce information about all videos that have already been removed so as to determine the amount of copyright infringing videos that have been available in the past. Most interestingly, he allowed Viacom access to all information about who has viewed which videos and how many times they have been viewed. This includes IP addresses, screen names, and videos viewed for every user. Viacom states that this will allow them to know, proportionally, whether copyrighted videos are typically viewed more often or less often than non-copyrighted videos. The judge also allowed Viacom access to the Google Video Content database so as to allow Viacom to determine Youtube’s knowledge of infringing activity.
This decision is interesting because it details the opinions of a judge who has considered both Viacom and Youtube’s opinions. He allows Youtube to retain several of its valuable coding secrets, but makes large concessions to Viacom to allow them to determine Youtube’s knowledge of infringing material. The reason for this decision can likely be linked to the relatively young age of cases like this. The DMCA has only been active for 10 years and many aspects of website liability for users infringing on copyrights are still uncertain. By allowing Viacom access to Youtube video records, the court is essentially hoping that Viacom will either show that Youtube is guilty of indirect liability or that Youtube has no control over the infringement beyond its current efforts. Thus, the impact of this court decision will likely come from Viacom’s analysis of Youtube video information. In my paper, I plan to further examine the same topic: whether or not Youtube is completely free from liability for infringing material.
The author, in this entry from a Web 2.0-centric blog, details Youtube’s recent efforts to both appease copyright holders and to promote creativity amongst its users. In January 2007, Youtube unveiled plans for a Revenue Sharing program which would give certain Youtube users a portion of ad revenue Youtube receives based on the number of hits their videos garner. Youtube will give even higher exposure to users labeled as “Directors,” people who are allowed to upload films greater than 10 minutes in length. Similarly, Youtube will share revenue with some copyright holders based on ad money they receive for the viewing of infringing videos. The author discusses the possibility that Youtube will have to increase the number of ads it shows to make up for the profit lost from the Revenue Sharing Program. This leads to the dilemma of Youtube losing viewers if advertisements begin to show up before minute-long clips. To increase the effectiveness of heightened advertising, Youtube may have to adopt a TV style model in which “an advertiser pays Youtube (and thus the content creator) X amount for every viewing.” To appease advertisers, Youtube’s new Audio Fingerprinting technology could be used to prevent inappropriate videos from being paired with reputable brands. This would be similar to Google Adsense which provides targeted advertising to firms. The problem relates to copyright because if Youtube adopts targeted advertising, which it has recently begun to do, it will be receiving revenue for ads placed in front of infringing videos for which it does not have deals settled with the copyright holders, thus increasing the possibility of them being vicariously liable. The solution, the author notes, is to use Audio Fingerprinting to detect copyrighted material and then inform the copyright holder, who will have the option to either remove the material or share revenue gained from the video with Youtube.
This system could potentially solve the problem of both Youtube and the copyright holder losing money from various transactions. Youtube loses money when it devotes bandwidth and time to a video only to have the video deleted due to a takedown notice. Similarly, the holder loses money wasting man hours filing takedown notices and finding the actual infringing material. If both groups work together, as Youtube intends, companies will be much less likely to sue Youtube, especially if they are actually making money from infringing videos posted online. Similarly, Youtube decreases its chance of liability because it is increasing its promotion of original works by paying some users. By offering directors a part of the revenue earned from their original and creative works, Youtube is encouraging users to make their own films rather than simply splicing together copyrighted material (which leads to zero profit for users). Thus, with the adoption of the revenue sharing plan detailed above, Youtube has simultaneously appeased the copyright holders and expanded its promotion of original material, showing courts that there are indeed significant “non-infringing” uses for Youtube.