Google Creating A 3D Earth

Last month, Google bought @Last and its powerful 3D modeling software called SketchUp. Today Google announced it was making a version of SketchUp free to download.

The free version of SketchUp allows users to draw 3D objects from "houses, sheds, decks, home additions, woodworking projects - even space ships." What's neat about SketchUp is that Google makes it possible for you to design a 3D version of your house and uphold it into Google Earth. As you can see here, this software has lots of possibilities.

Ars has a good look at this new Google tool, coming to the conclusion that:

Google may have looked at the success of mod communities, mashups, and open APIs, and thought that encouraging a massive influx of modeling help for Google Earth could help drive traffic to the mapping tool. Right now, I don't see any way to make your own 3D version of the street you live on an official feature in the public Earth maps, but something tells me that it's coming sooner or later. What you can do is make your model, enter information about it, and upload it to the 3D center for everyone to use. Mapping data could be added fairly easily, and if you built your model on top of a Google map to begin with, mapping coordinates and the like could be part of the model's metadata already. It would then be a simple matter of approving models for official use.

I agree. Imagine a map of a neighborhood from a street level view that included all the buildings and other landmarks on the street. Some of the current Google Map hacks are pretty cool (there a some links to a few on the right) and mapping the planet in 3D from a street level perspective would require a massive amount of user participation, a la Wikipedia. I don't see any other way to go about it.

The legal issues should be minimal. "Architectural works" qualify for copyright protection under 17 USC 102, but that protection is primarily focused towards the plans and drawings. How far that protection goes to the actual building is a little fuzzy. However, 17 USC 120 provides that "pictorial representations" of a copyrighted building are permitted if the building is located in or is ordinarily visible from a public space. Thus, no one can "prevent the making, distributing, or public display of pictures, paintings, photographs, or other pictorial representations" of the building. So, while I can't make a copy of a building's blueprint, I can look at it and take a picture of it, or even draw it, without a problem. That people are going to be making 3D models of the buildings shouldn't fall outside of the words of section 120.

The X-factor may be a trademark claim against anyone modeling a building without permission. To be a valid trademark, a mark must create a separate and distinct commercial impression that identifies the source of the merchandise to the customers. In Rock & Roll Hall of Fame v. Gentile Products, the court said that a photographer who took pictures of the Hall for posters he was making did not violate the Hall's trademark. The decision rested on the notion that taking a picture of the building did not create any confusion among customers, but the court did not make that decision as a matter of law for all like cases: the Hall merely failed to meet the proper evidentiary requirements in that case. It remains to be seen whether another building could prove its case, but the R&R Hall of Fame case sets a high bar in claiming trademark protection for a piece of architecture.

However, in a situation such as on Google Earth where inclusion of these user generated buildings could be used to locate and contact businesses, I could see a court having difficulty in following the R&R Hall of Fame decision without throwing in a whole lot of confusion. Google is not at the point yet of making user created building available on the public Google Earth, but before it goes that route, it's going to have to figure out a way to supervise the system so it doesn't get sued when people start posting unflattering models of businesses (like an Apple Store) or well-known parts of the skyline (who decides who gets to model the new World Trade Center?).

It's a cool (and free) tool however that is probably going to end up with a cult like following. It will also lead to the creation of more content online, making Google all the more relevant in terms of finding stuff. Just like Blogger, by giving people the tools to make more stuff and get it easily online, Google at the same time makes itself all the more necessary in making sense of everything available out there. All I have to say is keep it coming Google.

Sawafi: The Arabic Google

A dedicated Arabic language search engine will debut later this year under the name "Sawafi." There is currently no Arabic search engine on the Internet and with 65% of Arab Internet users unable to read English, it would appear that the success of Sawafi is a foregone conclusion. Sawafi is hoping to follow the example set by Baidu, the local Chinese language search engine, which is exploding with over 100 million web surfers. Baidu is doing so well in fact that Google had to change its Chinese name to better compete. (You can also see the Chinese Online Search Survey showing stats between Baidu and Google here). Sawafi's goal is to help Arabs get online and find what they need, making profits along the way.

According to CNET, there are only 100 million web pages in Arabic (that's only 0.2% of all the web pages out there), generated by more than 16 million internet users. And all of those Arab speaking users that can't read English have to deal with a web where 70% of the total content is in English. So here's the rub:

There is not enough Arabic content available on the Internet. But there's no motivation to put more Arabic content on the Internet as long as you don't have a system to find the content.

It's interesting to watch the Internet creep into new areas and see how people deal with it. Clear here is the enormous value that a good search engine provides to all the content out there. It even encourages the creation of more content because people will know that someone can find theirs! Wow.

I know it's not that profound of an idea, but its simplicity seems lost on those seeking to enforce their copyrights against Google and other search engines. There is hardly a market for digital content without a good search engine, yet search engines get sued for copyright infringement under the theory that they are appropriating the value of copyrighted work. I would say that maybe we have forgotten how annoyingly little content was on the Internet 10 years ago, before search got really good, and if we have in fact forgotten, I would remind people to watch things like Sawafi and Baidu take off and watch the populations they serve become Internet savvy rather quickly.

I would also remind people that copyright is supposed to promote the progress of the useful arts and that extending copyright law in a way that restricts what search engines can do does not promote anything that looks like progress.

Bonus Addition (9:23pm):
Do you know what Sawafi means? It's actually surprisingly tough to figure out Arabic to English on the Internet (for me at least), but Sawafi is a version of al-safiyah, which is the:

Share taken by the leader of an army before division of booty, such as horse or a sword. The term was later used as sawafi (singular, safiya) lands or possessions which the sultan (ruler) appropriated exclusively for himself. It also means all lands which their owners have abandoned or the owners of which have died leaving no heirs.

That means its in the public domain, right? What a nifty play on words.

A Tribute Not Taken

Quick. Do you know who Joan Miro is?

I didn't until Google told me.

On Thursday, Google displayed this special logo on its home page to celebrate the anniversary of Spanish surrealist artist Joan Miro's birth in 1893 (who, by the way, died in 1983). Google has a habit of paying such tributes to artists and other intellectuals, with previous tributes to luminaries such as Einstein, Michelangelo, Frank Lloyd Wright, van Gogh, Warhol, Albert Hitchcock, MC Escher, Ray Charles, Gaston Julia, da Vinci, Monet, and Piet Mondrian. What a happy birthday for Miro. To not only be included in such an exclusive group, if you will, but to also have interest in you and your work kindled by such blatantly free advertising seems to be a blessing for an artist not well known outside of certain art circles.

Halfway through the day, however, Miro's family and the Artist's Rights Society asked Google to remove the logo. According to the Mercury News:

The Artists Rights Society, a group that represents the Miro family and more than 40,000 visual artists and their estates, had asked Google to remove the image early this morning.

"There are underlying copyrights to the works of Miro, and they are putting it up without having the rights," said Theodore Feder, president of Artists Rights Society.
...

"It's a distortion of the original works and in that respect it violates the moral rights of the artist," Feder said.
...

Feder said the society receives hundreds of requests each day from media organizations who are interested in reproducing a copyrighted work in some form. He said the authorization process is simple: all Google needed to do was send an e-mail asking permission to use the images.

"We would have asked the estate or the family, and they would have said yes or no," he said.

Google agreed to remove the image as requested, but stated that it did not believe that the tribute constituted copyright infringement.

ARS is upset because Google did not ask for permission before creating the logo, which incorporated pieces from Miro's work (see here, here, and here). Perhaps Google could have seen this coming since they and the ARS bumped heads before regarding Google's tribute to Salvador Dali in 2002. ARS asked Google to remove that logo, and it complied, and the logo is now unavailable on the Google Doodle Archive Pages. But the key point is that, in cases like this, Google doesn't have to ask for permission.

Google only has to ask for permission if it is violating one of the exclusive rights of a copyright owner (ie. making copies, distributing copies, etc.). If Google makes use of the material in one of the ways that the law considers "fair use," then there is no infringement of a creator's right, thus no need to ask for permission. Copyfight has a good post on this and what looks to be a clear case of fair use:

But fair use, as U.S. courts recognize it, eliminates the need to ask permission. Fair use saves us from the sanitized world where only authorized tributes or commentary are permitted. Moral rights, applied in many European countries but not the U.S., protect the "integrity" of artists' works -- but even that was hardly under threat [in this case].

I really like that bit about "authorized tributes," and this post is exactly right: it is not copyright infringement if the use of the copyrighted material is deemed fair. Interestingly enough, the ARS website has a section for Copyright Information that covers Copyright Basics and Other Rights, neither of which contain a single mention of fair use. It's as if fair use doesn't exist or is something an artist (or his family) shouldn't worry knowing about.

A quick run-through of fair use appears favors Google. This is probably why the ARS wants to pretend "fair use" is not part of the law.

The first prong, the purpose and character of the use, is in Google's favor. First, the logo is very transformative, taking bits of several paintings to create an entirely new Google logo whose function, identifying Google and paying tribute to Miro, is different than the function which Miro's art serves, namely aesthetics. Second, the commercial aspect of Google's use is not "highly exploitive" (ie selling the copyrighted content for profit) and falls under the less burdensome rubric of "incidental commercial use."

The second prong, looking at the nature of the copyrighted work, might only slightly favor Miro. The paintings are highly creative and thus the kind of work copyright is supposed to protect the most. However, Miro's paintings Google relied on were originally published in the early 1940s and they're available online (see above). On this point, Miro barely ekes out a win.

Third, as to the amount and substantiality of the portion used, Google comes out ahead. Google is not copying any of Miro's paintings en masse, but is merely taking small, but noticeable, pieces from several of Miro's works. If any of this even counts as "copying," it is certainly no more than what is necessary for Google's intended use.

Lastly, one has to look to the effect upon the potential market for the copyrighted work, which appears to favor Google here as well, but it can be a little tricky. Google's use of Miro's art is not taking away from or competing with the sale of Miro's works. Google has, if anything, actually improved this market by bringing the attention of the masses to Miro. ARS might argue that Google took away Miro's right to license his work to be incorporated into Google's logo. This would be a market of one, probably outside the intended scope of this fair use prong. It's worth noting that the mere requirement of having to ask Miro, let alone pay his estate, would have the potential effect of destroying this market anyway.

Looking at the four fair use factors, Google wins 3-1 (or maybe 2-1, because the nature of the work prong could easily be a seen as a tie as well) and if a court agreed with my analysis, Google would be said to be using Miro's work "fairly" and thus could not be blamed for failing to ask for permission to use Miro's work in its logo.

ARS also claims that Miro has moral rights in his works and the right to prevent Google from distorting his reputation. First, US law doesn't grant the same kind of moral rights to works that Europe does, but what might be the closest is section 106A's bitty on Rights of Attribution and Integrity. That section gives a copyright owner the right to claim or disown authorship of a work and to prevent the use of his name as author of a work that would distort his reputation. Unfortunately, subsection (b) of that section clearly says that "Only the author of a work of visual art has the rights conferred by subsection (a) in that work, whether or not the author is the copyright owner." Thus if the author has been dead 23 years, he can't quite exercise this right now can he?

Second, even if he were alive he couldn't satisfy the requirements. No one would assume that Miro was endorsing Google (have I mentioned he died in 1983?) and it would be difficult to say that the logo is "prejudicial to Miro's honor or reputation" as it could easily be argued it enhances it by looking at the select company he is joining, as I mentioned earlier. Further, it certainly is not intentionally prejudicial as the folks over at Google are merely paying tribute to someone they are fond of.

I also wonder whether or not Google's logos would count as parody and whether that would be a better position for Google to take. A court might be more sympathetic to a claim of parody, but at the same time to do so might just confuse the court.

Simply put, the Artist's Rights Society is engaging in copyright abuse and I find it to be offensive. It's really too bad that the holders of Miro's copyright can and feel like they are entitled to wield this kind of power over others. It's also too bad that Google decided to cave in and remove it. Google acted in a noble fashion by voluntarily taking down the logo, but it's important to realize that it was not required to do so by law and that it would have been justified to have refused to listen to ARS's request. Every time Google refuses to take a stand on issues like this it is protecting its business interests and leaving the rest of us to dry. Technically, I too am infringing Miro's copyright by posting the logo above, and that his estate or anyone else could so me for this or similar acts will prevent me (assuming I cared about being sued) from doing these kinds of creative things.

Perhaps there are ways that, even if the laws regarding copyright stay the same, the rest of us can strike back and cause copyright owners to think twice before they threaten with a bogus claims. Take for instance Miro's page on Wikipedia. Perhaps, as it becomes easier for people to connect and share ideas, and it will becomes easier for people to publicly humiliate others, those who seek to abuse their copyright powers will be called out on it. While Miro has a life's worth of accomplishments, his Wiki page's largest section relates to the dispute with Google and doesn't make him appear to be all that likeable a figure.

It makes me wonder whether ARS is really looking out for Miro's best interests here or merely looking to protect its percentage of the profit from Miro's work at Miro's expense.

Defining Distribution

Ars has a nice article on Elektra v Barker, one of the RIAA's 19,000 lawsuits against people using p2p technology. Based on the arguments presented by both sides of the case, it looks as if the parties are jockeying to get the court to pass a binding ruling as to whether or not digital transmissions count as "distribution" under the Copyright act.

EFF Files Brief in Viewfinder Case

The EFF has filed a friend of the court brief arguing that the First Amendment commands the rejection of a French copyright law ruling against an American company and the decision thus cannot be enforced here.

The case, Sarl Louis Feraud International v Viewfinder Inc, has been bouncing around for a while now as the original French action beginning back in January of 2001. The claim is one of copyright infingement because a photographer for Viewfinder, a fashion ezine, took pictures of some dresses at a fashion show and posted them on their website without paying or asking permission to do so. The French court ruled in favor of the plaintiff, not because of the merits of its claims, but because Viewfinder never showed up. The two sweetest words in the English language weren't sweet enough, however, as Sarl's attempt to have his victory enforced in America was denied last September by Southern District of New York Judge Gerald Lynch (Read that decision here). Sarl is currently appealing that decision and the EFF has filed a brief, with the Center for Democracy and Technology and the ACLU, supporting Judge Lynch's decision in favor of Viewfinder.

The most interesting aspect of the case is its international flavor. Fashion designs are not copyrightable here in the US (though a movement is afoot to change that), while France has some of the strictest laws fashion copyright laws around. Since the Internet is medium that knows no geographical bounds, the question of whose laws apply to any given website creates an issue when laws conflict. For two good analyses, check out William Patry's Choice of Law and Internet Copyright and Kurt Wimmer and Eve Pogoriler's International Jurisdiction and the Internet.

Also, the principle of "ordre public" says that the US would generally have to enforce a French court's decision against an American company here, unless it violates a public policy here. In this case the public policy that enforcement here would run afoul of is the First Amendment, one of, if not the most important public policy this country has.

Viewfinder originally tried their hand at two arguments. First they argued that the French decision could not be enforced here because it was inconsistent with American copyright law (in that fashion is not copyrightable here). In rejecting this claim, Judge Lynch wrote:

Intellectual property regimes are economic legislation based on policy decisions that assign rights based on assessments of what legal rules will produce the greatest economic good for society as a whole. Different countries will, at different times, reach different conclusions as to the types of creative endeavor that should receive the benefit of copyright protection and the extent of that benefit, and different conclusions as to the kinds of competitive activity that should be encouraged or discouraged by trademark law. If the United States has not seen fit to permit fashion designs to be copyrighted, that does not mean that a foreign judgment based on a contrary policy decision is somehow "repugnant to the [American] public policies underlying the Copyright Act and trademark law."

Thus, even if we think the French are nuts for placing fashion design on such a high copyright pedestal, it is not for American courts to call them out on it.

The second argument was that enforcing French copyright law here would infringe on the first Amendment. It was on this basis that Judge Lynch ruled in favor of Viewfinder. Quoting from the opinion:

Viewfinder's last, and sole persuasive, argument is that the French judgment is "repugnant to fundamental notions of what is decent and just" because Viewfinder's conduct is protected by the First Amendment. The freedoms of speech and of the press protected by the First Amendment are not mere vagaries of legal policy, matters of legal detail that might as easily have been resolved differently by our legislatures or courts. Freedom of speech is a matter of constitutional command, binding even on the will of the majority as expressed in legislation. The very Congress of the United States "shall make no law abridging the freedom of speech, or of the press." Even among the basic human rights protected by the United States Constitution, the First Amendment occupies a special place. As Justice Cardozo put it, the American legal tradition "reflects a pervasive recognition of th[e] truth" that freedom of speech is "the matrix, the indispensable condition of nearly every other freedom."

Bingo. Judge Lynch then proceeded to confirm, over Sarl's objections, that pictures contain sufficient speech-like qualities to fall under the protections of the First Amendment, that a lack of editorial comment accompanying the pictures is not fatal for "a picture is worth a thousand words," and that using the pictures in a for-profit website is not a problem because this was a "simple sale of a news photograph of a public event." Relying on the breadth of American First Amendment jurisprudence, Judge Lynch reached the conclusion that the French decision was incompatible with the American ideal of free speech in that it prevented legitimate and well recognized forms of expression.

The EFF brief generally restates Judge Lynch's argument, but is very well written and definitely worth a read for understanding the connections between copyright law and free speech.

That so much emphasis is placed on the First Amendment is odd, presumably because of the international law elements of the case, since the First Amendment is not usually so well discussed in copyright infringement cases between American parties. In Viewfinder, the court essentially ignored the copyright argument, saying that the First Amendment issue was more important. In other circumstances, the court would have said that the First Amendment issues are subsumed within the fair use defense (referring to either Campbell v Acuff-Rose Music or Suntrust v Houghton Mifflin - the Wind Done Gone case). When the First Amendment issues are considered within the context of fair use, they tend to lose the emotional impact that is so present in Judge Lynch's decision. Even odder is the notion that American courts might take free speech defenses to copyright more seriously against international plaintiffs than when an American is claiming free speech against a claim of copyright infringement by another American.

The Viewfinder case illustrates the power that more emphasis on the First Amendment could have for copyright cases generally. Using the Google Book Search lawsuit as an example, Google, acting as an intermediary for the everyone with an Internet connection, would have a significant First Amendment claim to stand on. But as copyright has developed, that argument would have to be sparsed out between the various elements of fair use, diluting the argument and not providing proper space to claim important free speech values (ie, the right to read and the right to receive information).

If a court used a similar approach as in Viewfinder to Google, I think it would come out something like the following:

Court: Google, are you infringing on someone's intellectual property rights?
Google: Maybe, but only because this is a new area of law and it could go either way. But... um... there's a good chance it might.
C: Ok, well, does Author's Guild claim of copyright infringement affect your free speech?
G: Of course it does! We're just allowing people to track down information, kind of like a giant library of Alexandria. Is that so wrong of us?
C: Is that effect so great as to be against the important public policy of free speech, and all that that policy entails?
G: Yes. Free speech entails the right to speak one's mind, but in that is the right to gather information. As has been said, speech does not exist in a vacuum.
C: I agree. Go ahead and digitize the books...
- or -
C: I am not moved by your free speech claim. I think you're scamming the authors here. But I'll tell you what, can you show me that your use is nonetheless fair?
G: Fair use!? Boy, do I have a story for you...

I don't think such an approach would be that wrong, and in fact, that kind of heavy reliance on the free speech implications for an internet copyright dispute would help return the proper balance between the public's right to access and the copyright owner's limited rights of control.

Of course, Judge Lynch may just be one of those French-hating freedom fry loving kind of guys. But that would be ok too. I'm a big fan of the law of unintended consequences.

The Blogosphere is Blowing Up

Technorati's founder Dave Sifry presents the State of the Blogosphere and by his estimation, it is strong. According to the report:

• Technorati now tracks over 35.3 Million blogs
• The blogosphere is doubling in size every 6 months
• It is now over 60 times bigger than it was 3 years ago
• On average, a new weblog is created every second of every day
• 3.9 million bloggers update their blogs at least weekly
  
• 19.4 million bloggers (55%) are still posting 3 months after their blogs are created
• Technorati tracks about 1.2 Million new blog posts each day, about 50,000 per hour

The growing number of blogs and people posting to them regularly illustrates that this form of speech means something more than generating income to many people. Blogging is a form of self-expression and a tool with which to shape one's personal self-image, and it appears that this idea is quickly catching on. Although the actual percentage of how many blogs turn a profit, to the point that they can substitute as a second job, is elusive, it is a small percentage.

Wouldn't you agree that there would still be hundreds, if not thousands, of small newspapers in every city if it weren't so expensive to run one? Now that the Internet has brought those costs down almost to zero everyone with an Internet connection is a potential newspaper.

This leads to the question: if a disproportionate number of people do not blog with profits in mind (83% of bloggers characterize their entries as personal ramblings and 20% mostly publish lists of links to interesting stuff, source), should copyright law be in the business of allowing copyright owners to engage in what is essentially censorship of bloggers' online diaries? Courts have yet to consider the freedom of speech in online copyright infringement defenses outside of the free speech elements implicit in the fair use defense. As blogging develops and becomes more widely utilized, it seems clear that copyright law must adapt to fit the ways in which users engage in and experience the Internet and not the other way around.

If blogs can serve as a way to re-inject the First Amendment into copyright analysis, then a tool like Google enables this kind of speech. Without Google and the ability to find information and sources to write about, how would anyone be able to blog? What Google does in making information available it does for a profit, but it does this for the rest of us as well.

How Do You Say Google in Chinese?

Gu Ge (pronounced goo-guh). It means "song of the harvest grain" in Chinese. Google apparently has been having problems with its name in China because people weren't sure how to pronounce it. Google sounds a lot like "doggy," "old hound," "fruit fruit," or even "check check." In response, Google changed its name in China. This is the first time Google has changed its name in a foreign country, while running 139 country specific Google search sites.

That this change is motivated by consumer confusion may be a red herring. Google has been criticized for being a sloppy company, but they take great care to protect their image as a benevolent figure on the Internet. Not only does this help build trust with users, it also serves to keep litigation and legislation away. After all, which sounds better: a marketing and advertising company named "old dog" or a search engine that conveys the sense of a fruitful and productive search in a poetic fashion?

States Taxing Digital Media

Slashdot tips that states are looking for ways to tax digital media. It would be better for these kinds of decisions to be made in the open, rather than by state officials reinterpreting the tax code, and Congressional action on this can't be too far off.

One interesting bit is that Kentucky justifies its taxation of digital music because "music downloads fit the definition of personal property." I thought that a (legit) music download was considered the purchase of a license to use rather than a purchase of personal property. Buying the cd itself is one thing, buying the sound is another. No?

Another bit worth mulling over is that media downloads should be taxed like software. Washington defines software as "a set of coded instruction designed to cause a computer to perform a task" and officials have determined that the same definition applies to music and even books and movies too. Like software, data files cause a computer to perform a task in order for the music to be heard. I can't think of much that can't fit under this theory of taxable software. Such an interpretation could make possible a tax for merely visiting a web page.

The problem with these kinds of changes to the tax law is that it effectively skews our understanding of intellectual property without Congressional authority and, presumably, without much knowledge of what intellectual property is all about. How much do state tax agencies know about copyright? Probably not much.

I don't want to pay taxes for downloads. Who does? Congress could decide on a federal scheme for internet taxation and that would be ok. We already pay taxes for cable tv. But applying local taxes to a decidedly unlocal thing like the Internet just sounds like a bad idea. Even worse if it serves to further confuse people about what exactly intellectual property is.

As another aside, while digital media is growing rapidly and states are looking for ways to increase their tax base, it seems misguided to increase taxes on Internet use while cutting taxes elsewhere. An overall tax policy that fails to provide for all of a government's responsibilities is bad, but instituting bad taxes to cover for budget shortfalls doesn't strike me as a workable or sustainable solution.

Voice Search

Via Slashdot: Google looks to be entering the realm of vocal search in which Google listens to your spoken query and returns the results. According to Google's patent, Patent #7,027,987:

A system provides search results from a voice search query. The system receives a voice search query from a user, derives one or more recognition hypotheses, each being associated with a weight, from the voice search query, and constructs a weighted boolean query using the recognition hypotheses. The system then provides the weighted boolean query to a search system and provides the results of the search system to a user.

This would prove infinitely useful for Google Mobile because it will make life easier (assuming it works) and has the potential to put a real dent in the number of text-related injuries.

Another potential use for voice search would be to make searching for audio easier. This isn't what the patent claims to be for, but it seems it would be easier to match audio wavelengths than to go through writing transcripts or tagging all the audio out there. Making audio searchable is sure to be on the horizon, especially with the increasing popularity of podcasts, and the ad revenue that such search might produce could be huge.

Such a search service would run into tricky copyright issues, similar to those faced by the Google Book Search project, because they would have to make copies of all the audio to make to make a searchable database. Since podcasts are generally given away for free online anyway, the problems may be minor.

Whether Google were able to make music searchable in this way would be a different story. If the technology developed to make this possible it would make lyric searching easier without having to write up the lyrics separately.

I wonder, however, what effects voice search would have on copyright law. I do not know the answers to these questions right now, but is posting the lyrics of a song online a fair use? I'm going to guess no, but if a song is online and a Google bot finds it, can match an audio search string I provide to the audio of the song, and tell me the name of the song I have stuck in my head after hearing it on the radio, it would appear to skip over some of the copyright hangups with Internet search. You could count on the RIAA suing Google, but it's one of those things that shows how potentially useful the Internet and search can be, forcing us to think about is the potential public benefit great enough to justify reigning back the reach of copyright.

Book Search Basics

In preparing to go to a panel discussion on Google Book Search and the Author's Guild lawsuit I was looking for a quick refresher on the issues. As a Book Search primer you must watch Lawrence Lessig's online lecture on Book Search, supporting Google's claim to fair use in making books, at least those in copyright and out of print, searchable online. Great stuff and I will report on the panel later.

Google Cells

Via Slashdot. BusinessWeek has a nice write up on how Google is poised to organize the globe. What the article makes clear is that a company whose mission is "to organize the world's information and make it universally accessible and useful" has a lot of work on its hands.

Describing the eclectic mix of Google services, the article describes the company as:

being built entirely of stem cells: able to grow and develop into whatever form it sees fit.

It's a good metaphor as stem cells inevitably lead down a slippery slope to cloning. Just take a look at all the Google spin- offs out there. (This one is not much of a hack, but touchingly hilarious.

A Precise Topic! Finally!

My posts over the last two weeks have been infrequent, to say the least, because as I navigate the cases and commentary on fair use as applied to the Internet and search I increasingly feel like I'm trapped in the Labyrinth without a string. A large part of my difficulties has been a lack solid justification for my paper in arguing that Google's various lawsuits would set bad precedent if Google loses. Now, thanks to two professors, who I won't yet implicate in this project, I have found a legitimate basis on which to criticize current copyright law, which certainly beats just calling it dumb.

The First Amendment and the Copyright Clause of the Constitution are in tension. While a central tenet of our society is encouraging a robust public discourse on matters important to the country, the Constitution also restricts speech in certain ways as an incentive to create more speech.

Freedom of speech has often been interpreted to stress and favor political speech. However, this has been more of a response to technological limits of a given medium than anything else. The Internet poses new challenges for the First Amendment, and offers the opportunity to rethink what freedom of speech means in a world where each person with Internet access can be heard world-wide with almost zero costs. Perhaps it is time to shift our understanding of free speech to one that emphasizes speech's effect on shaping our world through participation than simply making snarky comments.

Free speech has always been described as an essential element in fostering a sense of personal autonomy. What is different these days is the way copyright law is working as applied to the Internet. It is a battle of words and metaphors in a public relations effort to convince people that they should support the rights of content creators. But supporting the rights of content creators the way copyright law does now betrays the fundamental purpose of copyright law.

A service such as Google is representative of the debate. It is a technology that is infinitely useful in organizing the massive amounts of info on the Internet. It's utility is good for the public. Also, as lawsuits are decided and precedents are set, the rules that courts apply to Google will apply to everyone else as well. If Google can't link to other sites, I won't be able to link on this blog. And if a majority of what Google does is deemed to be copyright infringement, it has profound implications for how we will be allowed to read, write, and interact with the world around us. Google is on the front line of the battle between greater Internet utility and content providers seeking to force users to consume their goods in the manner they dictate.

If freedom of speech can be harnessed as a lens through which to view the Internet and copyright law, then the laws can be discussed in a manner that people can at least understand. We may decide that strong protection of authors' rights is a good thing, much in the way that some prefer a Mall to Main St. USA, but that should be an educated decision people make through public debate, not through a legislative coup by interested parties at the exclusion of the public.

So this is the gist of the project. Any ideas or suggestions are most welcome. Exercise your freedom of speech and leave a comment.

TorrentSpy: MPAA Might As Well Sue Google

TorrentSpy, a BitTorrent search engine, was one of six torrent sites sued (read the complaint here) by the MPAA in February for copyright infringement for directing users to websites that host copyrighted content available for download. TorrentSpy filed its motion to dismiss last week, in which it suggested that the MPAA would just be better off suing Google:

There is nothing alleged to distinguish defendants' website from that maintained by Google... Everything alleged about defendants' website is true about Google, and even more so, because Google outperforms the allegations in the complaint.

In fact, the difference between Google and a site like TorrentSpy might just come down to marketing: Google simply sounds friendlier than TorrentSpy and doesn't help people download music or movies for free, thus, it can't be the same kind of nefarious web entity. Right?

Almost. Websites such as TorrentSpy operate a lot like Google in that they don't actually generate or host any content, they merely organize the info out there to make it searchable, providing links to requested information depending on user requests (you can even find the torrent files available through TorrentSpy in Google by typing the search string "filetype:torrent"). According to Wired:

Torrentspy is a search engine that helps people find torrents on the web. It crawls dozens of third-party trackers to see which torrents they are serving. Then, when you run a search, your results are made up of the hits that your search returned from those outside trackers. TorrentSpy doesn't host any torrents on its servers, but only points its users to torrents hosted elsewhere that they can download and use. It's a bit of a loophole, and the MPAA is actively trying to close that loophole.

A bit of a loophole is right, but it can also be said that this is exactly the kind of site (or behavior) the Supreme Court was trying to preserve as legitimate in the Grokster case. Everyone can essentially agree that search is a good thing when it comes to organizing all the stuff on the Internet, and at a certain point responsibility for infringing material online is going to have to lie in the providers of that content and not the engines that merely locate it.

The EFF predicts that this case will be the point where the world of copyright law collides with the world of search:

The complaint gives little guidance about what the studios think separates TorrentSpy from any other index. It alleges that "the predominant use" of the index is for infringement (shades of MGM v. Grokster!). It claims that "indexing files according to specific titles of copyrighted television programs" is evidence of inducement. It argues that TorrentSpy "favorably compare[s] its website to other peer-to-peer services widely used for infringing activities." I'm sure the plaintiffs will further develop their "TorrentSpy is different" themes as the case goes forward.

But that's the important question raised by the TorrentSpy lawsuit: what's the difference between a "good" index and a "bad" index, and is that a distinction that copyright law can effectively make? In 1998, when Congress passed the DMCA's "safe harbor" provisions, it seemed to be saying that indexes should be shielded from copyright claims, so long as they implemented a "notice-and-takedown" procedure. The TorrentSpy suit (as well as the MP3Board.com lawsuit) suggests that the entertainment industry wants to renegotiate that bargain in court. The result could have important implications not just for torrent indexes, but for all online index and search services.

And CNet refers to an EFF attorney:

Electronic Frontier Foundation attorney Fred von Lohmann said that the courts had not yet ruled on whether search tools could be held liable for copy infringement.

Most relevant cases, such as record labels' suit against MP3Board several years ago, have been settled before the issue has come to trial, he said. "We haven't had a case that really tests the case of whether providing an indexing service by itself an infringement," von Lohmann said.

And here is where the sticky issue with this case. A detailed analysis of a fair use defense could hypothetically come out differently for Google than for TorrentSpy, namely a major difference in the nature of the use and impact on the potential market. Also, Google might qualify for the safe harbors under Section 512 of the DMCA while TorrentSpy might not (wouldn't the knowledge threshold be a lot lower for TorrentSpy than Google?), based primarily on how the two sites position themselves and their corporate image (even though the technology is very similar).

There are decisions on filesharing. There are decision for internet search. Now it looks like there will be a decision on searching for shared files.

Sources:
MPAA Lawsuit
TorrentSpy's Motion to Dismiss
CNet: MPAA Sues Newsgroup, P2P search sites
Rothkin Law Firm (TorrentSpy's Lawyers and probably a good place to go for updates)
Macworld: Sue Google, Not Us, TorrentSpy Tells Hollywood
Softpedia: MPAA Wants to Ban the .torrent Format
Wired: MPAA Versus TorrentSpy
EFF: Copyright v. Indexing
BBC: BitTorrent Search Site Hits Back
Techdirt: TorrentSpy Says MPAA Can't Reinterpret The Supreme Court on Filesharing
MGM v. Grokster, and analysis by EFF

Revising Section 108 of US Copyright Law

Slashdot mentions that The Section 108 Study Group is inviting comments in its effort to improve Section 108 of the US Copyright Laws, which deals with the reproduction rights of libraries and archives. From Slashdot:

The Section 108 Study Group, a group of copyright experts, has been meeting to discuss Section 108 of the U.S. Copyright Law. It is focusing on preservation of websites and access to digital copies of library materials. Representatives of Internet Archive, including Brewster Kahle, went to the group's public roundtable sessions in March. Google did not register to attend the roundtable sessions even though the findings of the Section 108 Study Group may impact Google's Library Project. The Section 108 Study Group seeks written comments through April 17, 2006, according to this Federal Register notice.

Revisiting Section 108 is an important step that needs to be taken. The statute concerns itself with the number of copies made for archival use, causing problems in a digital age where there are no longer physical restraints on making copies. Does making one digital copy count as one of the permissible number of copies made, or does every page view (which could be an unlimited number) count as a copy? Also, the statute says that it is not infringement if "reproduction or distribution is made without any purpose of direct or indirect commercial advantage." A search engine might want to one day scan the contents of prestigious libraries to make them searchable, but the words of the statute would only allow it for purely altruistic reasons.

Interestingly, Google has decided not to join the discussion even though it would appear to be in their best interest to do so. Such a decision might more of a public relations decision than anything else.

The four issues the group is focusing on are:
1) Eligibility for section 108 exceptions;
2) Amendments to the preservation and replacement exceptions in subsections (b) and (c), including amendment to the 3 copy limit and off-site access restrictions;
3) Proposals for a new exception to permit the creation of preservation-only/restricted access copies in limited circumstances;
4) A new exception to permit preservation of websites and other online content.

The group notes that part of their motivation for Section 108 revisions is "the fact that digital preservation cannot be conducted without making multiple copies, and the lack of statutory copyright exceptions that clearly permit activities for effective preservation." In short, libraries are concerned over DRM and fear that such technology could essentially destroy many of the social and cultural functions that libraries serve when what used to be in books is now in a digital format.

The Group is still seeking comments, so if you have thoughts on the topic, submit them here.

Google as a Leech

Here is a nice explanation of the motivation behind suing Google and other search engines:

The latest attack, brought about by search's phenomenal financial success, is that search is sucking all the value out of millions of Web sites by scraping content in bite-sized pieces and doling it out to searchers, at the same time monetizing traffic that's being driven by content created by the site owners, not the search engine.

The article Is Search a Leech on the Internet? by Gord Hotchkiss, posted on Mediapost.com, continues with an insightful description of how search marketers decide to divy up their advertising funds. It's a vicious cycle, where bid prices are relatively constant until a company is able to significantly improve its conversion rate, at which point it can bid more for ad placement, driving up the cost of such ads for all marketers competing for the same eyeballs and clicks. All of this happens without the search engine doing a single thing, effectively making it richer through all the hard work of the marketers.

Hotchkiss also touches on the issue of search engines acting more like answer engines. This is problematic because it, in effect, sucks all the value out of a website. Here is a great example:

They need a quick answer to a question (i.e., what movie had Brad Pitt's first starring role?) and often they don't have to go any further than the search results page to find it. The answer is often contained in the snippet of text that comes from the listed site. If you did the search above, you'd find that the site tiscali.co.uk has the answer ("Dark Side of the Sun," by the way). The content came from that site, answered your question and you're merrily on the way, arguing about who played the corpse in "The Big Chill." But that's far from the experience the owners of tiscali.co.uk wanted. They pay for their site by selling advertising. On the page that snippet came from are three separate ads. You didn't see any of them, but you did see the ads the engine chose to show you.

This kind of criticism is the central complaint behind the suits against Google brought by newspapers and book publishers. Through the process of making information searchable, which is in the public interest, Google has the ability to control the flow of web traffic based on how it displays its results. It can make money as well as deprive others of potential revenue.

So what are we to do? Hotchkiss proposes a new Golden Rule for understanding the internet:

While I understand the frustration, I think it's time for a reality check. The fact is, it's the new User Rule for online: Those that have the users, make the rules.

This is probably not the answer that most content providers hope for, but practically speaking, it allows for search engines to continue developing better search technology without needing to waste money defending themselves from lawsuits. Search engines do in fact suck value from websites, but perhaps the better solution is for websites to concentrate on keeping customers rather than advertising for new ones.

"Stickiness" was a term thrown around a lot in the past, but Jakob Nielsen thinks that the real value in websites is not their ability to have a visitor spend three hours there, but in their ability to make users come back.