Shameless Self Promotion

Check this out. The Spanish blog Derecho y Nuevas Tecnologias ("Straight and New Technologies" according to Babel Fish, but it's really "Law and New Technologies according to its author) has linked to my blog, putting me in the same sentence as Eric Goldman and John Battelle. That's pretty good company. Thanks Pablo!

Thanks are also due to Ron Coleman of Likelihood of Confusion and Siva Vaidhyanathan of Sivacracy.net for their links and shout-outs.

And, for those keeping score at home, The Google Copyright Blog passed 1,000 page loads sometime Monday, October 23. A minor milestone I know, but it's still pretty neat.

Inside Scoop on the GooTube Deal

After Google's acquisition of YouTube was announced, word leaked out that Big Content companies like Universal were given ownership stakes in YouTube totaling $50 million. If that didn't sound sketchy enough, Mark Cuban has dug up an anonymous insider with some intimate details that makes the deal sound even worse. According to the source:

• Nearly $500 million of the $1.65 billion purchase price is sitting in escrow for the purpose of dealing with impending lawsuits against YouTube for copyright infringement.

• Google essentially gave YouTube a blank check to come to terms with the major record labels and even threw in a "most favored nation clause" to ensure no label got screwed and to avoid anti-trust issues.

• The record companies were offered equity rather than licensing deals because, as the anonymous source puts it, "if monies were received as part of a license to Youtube then [the labels] would contractually obligated to share a substantial portion of the proceeds with [the artists]." Since it's equity, the labels can keep the lot of it and not share a dime with the artists they're protecting.
  

• The deal with the record labels was a quid pro quo in two respects. The labels agreed not to sue YouTube for 6 months and agreed to sue YouTube's competitors in the mean time, thus cementing YouTube's (and Google Video's) dominance in this space. This already played out when Universal sued Grouper and Bolt in the days after the YouTube deal was announced.

Techdirt calls this a bribe by Google to have the labels go after video competitors and a sleazy deal overall. To be fair, the source is anonymous and admits that some of its conclusions are just speculation. But the whole thing does seem to make sense and, if you trust Mark Cuban (who happens to have an agenda in backing up his claim that YouTube will be sued out of existence), the source is of the reliable sort.

Taking the source at full value then, it appears that YouTube has only months before it faces a category 5 legal storm from the major labels. Tim Wu argued in Slate that YouTube's copyright problems isn't that bad, as compared to Napster, because of DMCA 512, which provides websites with a series of "safe harbors" for sites that host potentially infringing material. However, the difference between YouTube having to pay the major labels millions of dollars for infringement and YouTube having to remove the offending content lead to the same result: no more YouTube. The company relies on the belief that any video clip can be found there, and once that belief is lost, YouTube will lose its relevance and also its traffic.

Outside of fronting some serious cash for a legal defense, some clever lawyering, and a sympathetic judge, it does seem that YouTube's only chance for survival is someone finding a good enough way to flag infringing content and prevent it being uploaded onto the site. If that happens though, it might not save YouTube anyway since it relies heavily on that content to begin with. Even with the $500 million legal war chest (though that covers a lot of billable hours) and Google's cash surplus, it appears as if the whole affair of saving the YouTube we've grown to love is an uphill battle.

Google Custom Search and CDA 230

I questioned aloud the other day whether Google's new Custom Search Engine (CSE) service will make users similarly liable as oft-sued Google over its search engine. Here are some thoughts on CSE and CDA 230.

The technology behind what is going on is key to any lawsuit involving the Internet, so it's important to be clear about what is happening when a person uses CSE to build their own search engine. So, how does Google custom search work? Take my Google Copyright Engine (GCE), for example. The CSE service is nothing more than a front end to Google. I have created a page, with Google's help, onto which I have placed a form for users to enter search queries. A person interested in learning about Google's legal affairs will enter something into this form and that info will be sent from my server to Google. Google will then take that info and, based on how I have configured my engine, will calculate the results and display them. In my case, my GCE displays the results as part of my GCE homepage hosted by Google. Another option is to have the search results displayed directly on a page within Google Copyright Blog, but I don't think Blogger supports that option (though Blogger Beta probably does).

As one can see, I don't have any control of what is included in the Google database nor do I have control over Google's PageRank. This fact should preclude my being liable in suits such as KinderStart's, which is suing over being demoted in Google's search results. But what about what actually does show up in Google's rankings?

Fortunately, section 230 of the Communications Decency Act should provide a wealth of protection for me and other Google CSE users. CDA 230(c) specifically provides that:

No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.

Frequently raised in the online defamation context, this provision generally grants immunity to those sites/services that link to or reproduce defamatory material that originated with another source. Whether or not a "Google Bomb" can be considered defamatory (the ADL says no because Google rankings are automated), my use of CSE won't make me liable for Google's search results because my GCE is providing information from another information content provider. Julie Hilden provides a compelling argument for why you can't sue Google for these kinds of things. But the extent to which CDA 230 covers more than defamation and similar speech torts is unclear. In one case, Randall Stoner v. eBay, eBay was held immune under CDA 230 for the actions of users selling bootleg tapes, but a copyright claim was not made by the plaintiff.

The issue of liability gets trickier when looking at the level of customization that CSE offers. Google's CSE allows users to add or exclude entire websites, parts of websites, or individual pages into their CSEs. Users are also able to calibrate their engines to favor certain results by specifying certain keywords that will provide a custom bias to what results are returned. While PageRank is still relevant in all of this, CSE offers a level of customization that may override much of Google's relevance calculations.

That Google search results are displayed as part of an automated system goes a long way in arguing that Google is not responsible for what gets displayed as a relevant search result, as the ADL has said. Google itself hasn't divulged the extent to which it manually edits its rankings, but it does reserve the right to under its Terms of Service. For Google, the more it is seen as editing it's results, the less it is providing information from other providers, thus coming closer to being an information provider itself under CDA 230.

At a certain point, the question might become whether my GCE is sufficiently customized and biased towards certain results that it is no longer automatic nor relying on Google enough to warrant protection under CDA 230. If my GCE, for instance, is biased to return one of my law professor's profiles as the top result when a user searches for "beats his wife" or "notorious pedophile," and that result does not occur in Google's standard rankings, I could be in trouble for providing that information myself. Automation is probably not a good defense if the a CSE is automated to be defamatory.

Lastly, it's always worth noting what the Terms of Service say for any online service. Few people actually read these, whether because they're lazy or realize their lack of bargaining power and see their only other option is to not use the service. The Custom Search Engine Terms of Service offer some interesting tidbits. Section 1.4 on Appropriate Conduct is worth close reading, especially this:

(n) engage in any action or practice that reflects poorly on Google or otherwise disparages or devalues Google's reputation or goodwill. Further, the Site [where the CSE code is placed] shall not contain any pornographic, hate-related or violent content or contain any other material, products or services that violate or encourage conduct that would violate any criminal laws, any other applicable laws, Service policies, or any third party rights.

When read with the Indemnification clause in Section 5 [you will indemnify, defend, and hold Google harmless in the event of a lawsuit against Google stemming from a breach of the Terms], it would suggest that a CSE result that defames, or otherwise results in a lawsuit, makes the CSE user wholly liable. It could be argued that this combination of clauses prohibits a white-supremacist website from putting the blame on Google's automatic results if a lawsuit challenges the site's "HateTheJews" CSE. That's an extreme case, but the issue would be the same for a child protection site that creates a CSE to scour the Internet for info on pedophiles and returns a false positive on someone. If a lawsuit is brought under "any applicable law," the site providing the CSE is obliged to indemnify Google.

As an aside, I would also point out CSE's ToS section 1.5 on Exclusivity, stating that by using Google CSE the site owner agrees that "Google will be the exclusive provider of Internet search services on the Site."

CDA 230 only addresses defamation and other similar speech torts, a type of case that Google faces rarely compared to copyright cases. At some point soon I will address how DMCA 512 relates to Google CSE.

New Color Scheme

Hope you like. And hope no one gets confused.

Bush Uses "the Internets" to Reach "the Google"

ROTFL. President Bush uses "the Internets" to reach "the Google." While Google has sent warning letters in trying to police the use of its name to avoid genericide, Bush's use doesn't seem to run afoul of the approved uses of Google's TM since he's still using it as a proper noun rather than a verb.* But it's still funny as a grammatical error. I should try and avoid both the inference that this means he doesn't use Google all that much and thus doesn't seek out answers to questions (or have questions) as often as he should, and the question of whether it's appropriate for the leader of the "free world" to be ignorant regarding the Internet, but it's too irresistible.

* Reproduces an article from The Independent which is now behind a pay wall.

Political Google Bombs

Another campaign season means more another look at how Google bombing is being used as a campaign tactic. This isn't new, as the NYT would have you believe, especially since the NYT reported on the tactic in 2004 (do they even bother to Google their own papers before writing new articles?). It brings to mind a piece on Slate by John Hiler, about blogs ruining Google, that discussed the practice as mostly harmless, given that Google bombs were usually motivated by humor, ego, money, or justice. He worried that the practice could do permanent damage to Google's search results, but Google wasn't too worried back then in 2002:

"We love the Weblog community," said Google's Peter Norvig. "We don't see any problems with Google Bombs yet. You would need a concerted effort to abuse Google. What we're seeing now is independent nodes acting alone."

And, though the tone is changed, it's still on message now:

“We don’t condone the practice of Google bombing, or any other action that seeks to affect the integrity of our search results,” said Ricardo Reyes, a Google spokesman. “A site’s ranking in Google’s search results is automatically determined by computer algorithms using thousands of factors to calculate a page’s relevance to a given query.”

Google's hands-off approach seems fair here. Since a Google bomb is created through linking, remedying one would have to depress a page's ranking somehow. And since there's nothing "wrong" with linking to other pages on the Internet, it doesn't seem that its Google's job to make sure that people are linking in good faith. In fact, what people say and what they link to is precisely what makes Google work in the first place.

BUT, the potential here for a defamation claim against someone seems too obvious. For instance, a Google search for Santorum, as in Sen. Rick Santorum R-PA, returns as the top result a curious linguistic reference created by Dan Savage in response to anti-homosexual remarks by Santorum. That example hews the line of political speech a little too closely and isn't a great example for defamation, but you can see what can be done to someone. It would make sense to pursue such a claim against the originator of the bomb, but that's not necessarily an easy thing to find out. And given that people like to sue Google, it may only be a matter of time before someone comes to Google decrying their search results, demanding that something be done to clear their good name. I can also think of trademark claims against the tactic (dilution anyone?). If Google bombing becomes more widespread, it will no doubt cause numerous headaches for Google, both technically and legally.

Now if only I could manipulate PageRank so that this site comes up on top when you search for Google copyright.

The Google Copyright Engine

This looks too cool not to risk getting sued over, just like Google does, so I started my own engine: the Google Copyright Engine. I need to update my blogroll since I've just started blogging again and have more sites to add, but they will all be included in the engine. Also, Google allows users to include single pages into the engine, which seems similar to making bookmarks searchable. If anyone would like to volunteer in adding sites to my engine, I'd be happy to have your help. Email me at ngdaly at gmail dot com and I'll add you as a contributor.

As an aside, it seems that the service is ripe for "search-squatters" right now since the service is new. Better get on that lawyers!

Get Sued Like Google

Google has released their Custom Search Engine, a custom search tool that allows users to create their own search engines using Google technology and which can be hosted on a user's website. Now that everyone can build their own search engine - not that they couldn't before, just that the process has been Googlicized - does that mean that we've all availed ourselves to being sued over what our search engine turns up?

Google Gets Sued A Lot

Thanks to the NYT for telling the world that Google gets sued a lot. From Geico to Perfect 10 to Belgian publishers to AFP to the Authors' Guild to KinderStart to Rescuecom to inheriting Robert Tur's against YouTube for posting the Rodney King video, the article neatly summarizes a veritable "best of" when it comes to Google's legal problems. Unfortunately the article is light on substance and takes the "unbiased" approach, making some of the cases (ie. KinderStart) sound more meritorious than they really are:

Last spring, KinderStart, a small search engine in Southern California that focuses on information for parents of young children, sued Google after it noticed that its site had been removed from Google’s search results — leading to a loss of traffic and revenue for the company.

Google said in court filings that an area of the site that permitted visitors to add links had been full of pointers to low-quality or pornographic sites, indicating that it was poorly maintained or was an effort to manipulate Google’s search results. KinderStart said the removal was unfair and unjustified and that Google’s guidelines on ways to avoid such punishment were too vague.

A federal judge in San Jose dismissed the first version of the complaint, in essence agreeing with Google that the company is free to shape its search results in any way it chooses. KinderStart has filed a second, amended complaint, which is scheduled to be heard by the same judge on Friday.

“We’re not against innovation at all,” said Gregory J. Yu, a lawyer for KinderStart. “But Google should not dictate what we should or should not see and find on the Web. They can knock off these small Web sites and there’s nothing the small Web sites can do.”

What this doesn't make clear is that KinderStart really is nothing but a link farm, pastes other people's content into their own window, and refreshes the page every few seconds (I'm presuming to boost page loads), all of which I've said before. None of which comes up in the NYT piece.

But it's a nice little summary nevertheless, if that's what you're looking for.

JASRAC Opts-Out of YouTube

YouTube has removed nearly 30,000 videos that were flagged as infringing copyright by the Japan Society for Rights and Authors, Composers, and Publishers (JASRAC), a Japanese equivalent of the RIAA. But don't worry too much, JASRAC actually went through the trouble of identifying all those videos themselves and asked YouTube to remove them, following the opt-out policy inspired by Google and encouraged by the DMCA safe harbor provisions. This actually helps legitimize the policy by showing it's really not that hard to do, but whether the cost of flagging that many videos is worth removing them from YouTube remains to be seen.

Google News: Good for Democracy?

A News.com op-ed rips Google over its Google News service in light of the successful lawsuit by Belgian news publishers, which even forced Google to post the ruling on its Belgium homepage (good summary of the affair by SearchEngineWatch here). The op-ed calls Google "immoral" for free-riding on the work of others and blasts the whole Web 2.0 thing.

Again, there should be no problem with linking to material that is freely accessible online. Unless I restrict access to what I put online, it is proper to think of me as having given everyone else an implied license to link to and use my material, up to a point. Plagiarism and full copies are right out, but headlines have yet to be fully recognized by copyright law, as with ledes, so I have few problems with the behavior of Google News. If it infringes copyright, so does just about every blog. TechDirt responds similarly to the article.

One argument that always arises in defending Google is that its News site boosts the traffic of the articles it links to, but here's another: the benefit of users being confronted with unexpected and varying viewpoints.

I have recently read Cass Sunstein's Republic.com, which fears that too much choice online is bad for democracy as people can filter their experiences as they wish and that niche sites will lead to group polarization, calling for affirmative steps to be taken to encourage democratic deliberation online. Dan Hunter has highlighted several problems with Sunstein's argument (disclosure: Hunter is my Prof. for Cyberlaw right now), but the desire for people to be confronted with views they ordinarily wouldn't encounter is addressed by Google News.

The reason to read Google News is that Google links to several sites for each story it covers, giving me a quick roundup of the various views on any given topic. I could be given links to as diverse sources as CNN, FOX News, and the Guardian. As such, this particular application of linking to copyrighted material without permission (making the outlandish assumption that it truly is infringement) addresses important social needs that don't appear to be satisfied elsewhere in the market. While we may disagree as to the prevalence of group polarization online (which Sunstein sees as rampant), Google News seems to offer an ideal check to fringe echo chambers. And copyright is supposed to supplement democracy, not detract from it, right?

Darwin Goes Online, With a Catch

In what must be assumed to be a slap in the faces of Intelligent Designers, the entire works of Charles Darwin are now available online. And searchable too. This is great news:

John van Wyhe, director of the project run by Cambridge University, said the collection is so comprehensive it will help dispel the "many misconceptions and myths" about the naturalist

He told the BBC that nothing can now come between the man and his writings.

Except for maybe copyright law. At the bottom of each scanned page of Darwin's original work is one of these:


So what exactly is being copyrighted here? We know it can't be Darwin's original work since it's in the public domain. Claiming copyright to the scanned image of the original works is sketchy because while a new edition or version of a public domain work can be copyrighted (as in its particular expression in the new version), we are dealing with just scans here. The whole point is to replicate the original as closely as possible: there is no "new" expression. Photocopying stuff doesn't give me any copyrights in the copies I've made and scanning books is no different.

This recalls BookSearch's PDF downloads that include Google watermarks on the pages of public domain books. At least the Darwin people are not trying to restrict the use of Darwin's works like Google does with its PDFs.

Perhaps the copyright notice at the bottom of each page means nothing at all, but was simply tacked on out of an abundance of caution because copyright is all in the news right now and it's better to be safe than sorry. If that's the case, how much worse off will we all be when people feel it's appropriate to slap copyright notices on everything they put online? I doubt people will trust/follow copyright notices that appear arbitrarily, so I hope the Darwin collection has a solid basis for doing this and would certainly enjoy someone telling me what that reason is. Seriously.

That Explains It

Yesterday I posted on how Google/YouTube avoided getting sued by Universal. Well, now it seems that the answer to why not has come:

Three of the four major music companies — Vivendi’s Universal Music Group, Sony and Bertelsmann’s jointly owned Sony BMG Music Entertainment, and the Warner Music Group — each quietly negotiated to take small stakes in YouTube as part of video- and music-licensing deals they struck shortly before the sale, people involved in the talks said yesterday. The music companies collectively stand to receive as much as $50 million from these arrangements, these people said.

You read that right. To make sure Big Music wouldn't sue over "pirated" music videos and amatuer lipsyncs, they were given ownership stakes in YouTube so they could reap the windfall of Google's $1.6 billion purchase of the company. A combined $50 million is paltry compared to $1.6 billion, but it helps explain why Bolt and Grouper were unable to strike deals with the music companies before being sued. In fact, I'm not even sure that YouTube has brought in $50 million, but such a deal sets a high bar for other such video and music sharing sites if $50 million and a share of ownership is the going rate for avoiding a lawsuit. It all sounds fishy.

I wonder if Google requires YouTube not to be evil?

Google Not Sued

Universal announced that it is suing video sharing sites Bolt.com and Grouper (which, remember, was recently bought by Sony) for copyright infringement. The announcement comes after Google's purchase of YouTube, which Universal threatened with a similar suit weeks ago. The threat may have been a bargaining strategy - YouTube signed a licensing deal with Universal shortly before being bought by Google - but it's not clear that YouTube is totally safe should Universal and others grow unhappy with how things turn out. There has been lots of discussion as to whether YouTube will cause legal trouble for Google, so that Google not being sued at this point must come as a relief to the company.

Copyright owners have a monopoly right over their content and can choose who they license it to, but if we are really in the beginning of the "video revolution" as Google CEO Eric Schmidt has claimed, does it makes sense for copyright holders to allow only YouTube to host videos while suing other competing sites out of existence? Such a strategy by Universal seems a bit too abusive, monopolistic, and shortsited by putting all of its eggs into YouTube's basket, decidedly so if YouTube tanks upon implementing whatever its ad strategy is going to be.