Maybe Rupert has a point

Sunday, Nov 29, 2009 12:46 pm
William Barnes

Everybody thinks Rupert Murdoch is crazy for wanting to hide his web properties from Google, but I’ve been thinking: maybe he has a point. Bloggers and podcasters say that Rupert doesn’t understand the Internet; I think they’re probably right but I also think the bloggers and podcasters are too understanding of the Internet. They (we?) put out content for free and whenever something comes along that makes an old media guy like Rupert uncomfortable they just brush it off and laugh: “That’s the Internet for ya.”

Don’t get me wrong, I like Google and I mostly trust Google, but I can see how what they do might be called unfair. When they index websites, they make a copy. Somewhere hidden in an abandoned missile silo (I assume) they have a vast array of computers mirroring the Web. When you search, Google searches these computers and presents you with a link and a little snippet of text from the site. Everybody generally agrees that the little snippet is fair use. But what about that underground data centre? It’s a huge asset for Google and it is made up of content they copied from other people. Think about it this way: if you go to the library, look up a definition for a word and quote it in an essay, that is fair use; but if you go to the library, photocopy the entire dictionary and take it home so you can look stuff up later, what is that? Google takes home the entire library. It’s not the display that is the problem, it’s the process before the display.

But, people say, Google is providing a service. True, but they’re also capitalizing on other people’s work without paying. Further, Google’s copy of a particular website is a valuable asset whether or not Google ever sends traffic to that site. But, they continue, if you want to exclude them, you can just put a robots.txt file on your site. Also true, but if you do that then your site may as well not exist. Google won’t offer to pay to get a single site into their search engine. But, the conclude, this is the way the Internet works now, search engines index and don’t pay to do so. True again, but does it have to work this way? Why shouldn’t search engines pay sites that they index? (I can actually think of a number of extremely practical reasons but for present purposes I’ll pretend I can’t) Just because something works one way now, doesn’t mean we all just have to accept it.

[Added Mon, Nov 30, 2009] My first website went online in 1997. For 12 years, I’ve been ecstatic every time somebody read something I posted. I imagine many bloggers feel the same way. It leads us to disregard a property right that we have. When Rupert comes along with his old media view, he’s shocked. We’re all just giving Google our content; Google is making money from it and giving us nothing in return. Back in his day, people wouldn’t have stood for it and he’s not going to stand for it now. Google provides a commercial service on the back of what may be massive copyright infringement, it’s something worth thinking about.

In trying to figure out where the value comes from, ask yourself this question: who would last longer, Google without content, or content without Google?

Ownership in the digital world

Friday, Oct 9, 2009 1:07 pm
William Barnes

My latest IP Osgoode post is up. In it I discuss the difference between what consumers expect when they buy digital content or a network connected device and reality. This post expands and clarifies that article.

Most people, I believe, think of all their physical possessions similarly. There is no difference between a couch and a novel as far as ownership goes. The incidents of ownership are the same: you possess the object; nobody can take it from you without permission; you can dispose of it as you please. There is, however, one significant difference between the two: the novel is a copy of a work in which the author retains rights. But physics isolates the consumer from this fact. They have a single copy and will probably never have more than a single copy. If they sell that copy, then they lose it at the very instant another person gains it. As long as the consumer treats the book as an ordinary possession, they will never have to think about IP rights. For most practical purposes, therefore, all physical possessions can be treated the same.

With digital content, however, all this changes. Rather than being given a physical object, consumers are granted a license to use the work in certain ways. This license will, at the very least, allow the consumer to possess and access copies of the work for personal enjoyment. Note that I said copies in plural. With a physical object, the owner has a single copy. The fact that they own a copy of Cryptonomicon (great book) does not give them the right to possess just any copy; it has to be that specific one. With digital content, however, you can have as many copies of the work as you want. In fact, if you buy a song from iTunes you will be told that it is your responsibility to make copies. And it shouldn’t matter where your copy comes from. If Alice and Bob both buy a song from iTunes and Bob’s hard drive crashes, then Bob should be able to just obtain a copy from Alice (at least in theory). In a real life example, if a user deletes a book from an Amazon Kindle (or lose the device), Amazon will let them download a new copy for free. This doesn’t happen with hardcover books; Bob can’t photocopy Alice’s copy of Cryptonomicon and Chapters won’t give him a new copy for less than the cover price.

In this fashion, licensing sounds pretty good. But it can give rise to strange circumstances that dealing in the physical world would not lead one to expect. Last month, Amazon remotely deleted two books from Kindle e-book readers when they learned that they did not have the right to sell the book. It sounds weird but consumers who purchased the book from Amazon didn’t purchase anything and Amazon did not take anything from them when they deleted the book; Amazon can’t give what they don’t have. Nobody would expect an employee from Chapters to come to their house and take back a hardcover book if it turned out to be counterfeit. There’s the practical difficulties, of course, but there’s also a physical item which the consumer owns. The consumer would be well within their rights to refuse to hand it over. In the case of an e-book, however, there is no interest in anything other than the license. The consumer may have other causes to complain, but they cannot claim that they had any right to keep their copy of the e-book.

The Amazon issue leads us away from ownership of digital content and into a related (and even more problematic) issue: ownership of network-connected devices. Unlike an MP3 or an e-book, the Kindle is an actual physical thing. Traditionally, consumers don’t expect manufacturers to interfere with their physical possessions after sale. Connected devices, however, routinely phone home for updates and content. This is very valuable behaviour but it isn’t always used to enhance the device. It can be used, for example, to remove the ability to skip commercials from a PVR; to block websites on a smartphone; or delete books from a Kindle. What this means is that, for better or for worse, the device someone buys is not in their own control. Why would a company want to downgrade a device? I think the best reason is to avoid liability. They have sold a device (or, in Amazon’s case, content) that has a defect of some kind and don’t want to be sued. The problem with this is that it shifts the burden of the company’s mistakes onto its customers. Amazon could have refused to delete the books but they would have been sued. Instead, they saved themselves by depriving others—although they did refund the purchase price.

That said, I don’t think that forbidding such practices is the way to go. It could very easily end up banning a lot that is useful about network-connected devices. I also think that over time an etiquette will develop over when such action is permissible and what compensation is required. Partially this will be regulatory, but the market will help define what people are willing to put up with. And I actually think that in Amazon’s case, it is defensible. If Alice sells Bob a stolen car, Bob can’t reasonably expect to keep it. Likewise with the example I used in my article of a defective reclining chair. It’s better to lose a function than a leg. Maybe companies should have to offer to buy back devices that they downgrade or offer a reasonable monetary payment.

I don’t think that we should try to force new forms of property into awkward physical molds (the music industry tried that with DRM and look where it got them). A balance has to be struck between the surety of the old ways and the flexibility of the new.

In defense of the white pages

Sunday, Sep 27, 2009 9:05 pm
William Barnes

I apologize if this is hopelessly I-just-read-CCH of me. No doubt it’s all been said before, but when has that stopped me?

Copyright only protects “original works”. But original work, it turns out, is a tricky concept to define. On the one hand, you might say that original simply means “not copied”. This would give rights to anybody who puts effort into creating something. On the other hand, you could require a little bit of novelty or creativity. In CCH, the Supreme Court of Canada set out a (possibly) different originality standard that required more than mere effort, but not actual creativity. This standard requires that a work be the product of “skill and judgement”. A lot has been said about that judgement, but I think that I have discovered the truth behind it: people don’t like phonebooks.

Whenever a person talks about the difference between the sweat-of-the-brow (effort/UK) standard and the creativity (US) standard, they always boil it down to that one thing—the UK standard protects phonebooks. And when originality is discussed in class, students always include the disclaimer “of course, I don’t believe phonebooks should be copyrightable”. Why all this hostility? There must have been a wave of heavy-phonebook-fell-on-my-toe incidents in the pasts of most lawyers. Their first tort.

Seriously, though, why is it that people are so sure that phonebooks shouldn’t be copyrightable? They provide valuable information. They might even take quite a bit of effort to put together. Should someone be able to just photocopy the thing and sell it? OK, I admit that phonebooks are free, but imagine another completely non-creative collection of facts that someone would be willing to pay for. There’s no incentive to create such a thing if you have no rights at all and that is what all this talk of originality seems to say. I’m just not sure it makes sense to deny copyright at the point of originality—except on the broadest definition of the term. There must be some way that the phonebook author can have the bare minimum protection. This would prevent him from being so directly undercut and ensure that any copier would produce a slightly different phonebook. In this scenario, the public would even have two different versions to choose from. Everybody wins.

I’m still working this out in my mind, but I think there are other concepts in copyright that can do the work that originality does in North America.

Software Patents ~= Literary Patents?

Sunday, Sep 13, 2009 9:00 am
William Barnes

Stuart Freen posted a reply on IPilogue to a Cato Institute article claiming that software patents are analogous to literary patents. I thought I agreed with Stu, but I’m not so sure.

A story has words, literary devices and a plot. A computer program has variables, control structures and algorithms. Literary devices/control structures manipulate words/variables in accordance with the plot/algorithm. There are multiple ways of expressing any plot or algorithm. So you have two things of value: the expression and… I’m having trouble thinking of a good word. Function is too slanted towards programming. But there is something purposive shared by programs and stories. Just like you (probably) wouldn’t run a program that does nothing, you (probably) wouldn’t read a story where nothing happened.

What does the plot of a story do? It holds the readers interest. I may be a weird case, but I seldom appreciate the writing in a book. People tell me how they love the descriptiveness in a certain author’s writing, but when I read it I never notice. I tend to skim the adjectives because I want to know what happens. I don’t care how beautifully George R. R. Martin describes the landscapes of Westeros, I just want to know what happens to Neddard. I would probably find a much more poorly written book that hit all the same plot points to be just as satisfying.

So the argument goes: if you allow a programmer to patent a certain algorithm that solves a computer problem (e.g., compressing a video file) why not allow a writer to patent a certain plot that solves a—the—literary problem (e.g., making the story compelling). I’m characterizing the problem solved by a particular combination of plot points by their effect on the reader. So I imagine the purpose of a literary patent to be a method of sustaining reader interest rather than, for example, a method for explaining how the thief got into the locked room. Stu separates the literary value as being about expression while the computer is about function, but I accidentally convinced myself that both kinds of value are present in either medium.

This doesn’t mean that I don’t agree with his conclusion. The fact that we don’t allow literary patents does not mean we have to disallow software patents for consistency’s sake (I’m stating the Cato article’s point much more plainly than I think the author does). We could just accept it as inconsistency and say that we just don’t want to patent stories. Sorry authors, you’ll just have to live with copyright. But let’s say we do want to entertain the notion of a literary patent analogous to a software patent. The software patent is easy: you express an algorithm in abstract terms and say what it does—find out if a number is odd by dividing by two and seeing if you get a fraction. Somebody could follow that abstract and create a function of equal value (though expressed in different form perhaps). But what would the abstract terms in a literary patent look like? It couldn’t be anything as general as “two star-crossed lovers meet and get secretly married and then commit suicide”. That description wouldn’t allow another person to write a story that has the same effect on the reader as Romeo and Juliet. For that, we would need a description of incredible detail. And if a story infringes at that level of detail, I think copyright would be comfortable handling it (I’ll have to ask my IP prof on Tuesday). And I’m not even getting into a discussion of obviousness or prior art. Therefore, I think we can safely say either that we don’t want literary patents for some reason of cultural values or because they would be impractical and useless.

I hope all that made sense. It was somewhat stream of consciousness (hey, that’s a literary device). I need to start editing these blog posts. By the way, none of this is to say that I’m completely in favour of software patents. It’s one thing to patent a compression algorithm and another to patent a user interface element or simple method (Amazon 1-Click comes to mind).

The Commons

Friday, Jul 17, 2009 7:51 am
William Barnes

I just wanted to add something to my recent post on copyright on 200 year old paintings. The reasoning behind copyright is that giving a time-limited monopoly encourages people to create new things. And why do we care that people create things? Because after the time limit is up those creations enter the commons and become the property of everybody (and I ignore the contentiousness of that phrase). People understand the monopoly bit, but rarely the commons bit.

What bothers me is that museums are given works of art for safekeeping. They control access to the physical items and (in the interests of preventing damage perhaps) restrict the ability of people to take copies (flash photographs, for example). Eventually, these items are supposed to pass into the commons. But what good is a work in the commons if nobody can get a copy of it?

It’s really difficult to work out what’s right and wrong here in any general way. I don’t think we want to impose a duty to provide the public with copies of work in the commons. And I’m not sure we want to take away economic incentives to reproduce such works. It can be very expensive to restore and reproduce art. Maybe laws need to be made that address the reproduction of privately held commons works. Then again, the practical difficulties of providing copyright protection to copies of commons materials could be prohibitive. Eventually, those digital copies will themselves enter the commons. Does the first person to copy them at that point receive a copyright interest? And how do we manage the fact that dozens of people may own copyright in what is essentially the same thing?

In the case of museums it’s simpler than with private owners. Museums exist for the purpose of protecting and sharing cultural knowledge. They receive public funding, tax benefits, and physical possession of the works based on this. They should not be removing works from the commons. I’m not saying the museums are evil. They are, of course, constrained by financial considerations. They are underfunded and need to take advantage of everything they have. I’m just saying that something is wrong and needs to be fixed.