Thursday, February 11, 2010

Letter to the US Trade Representative

From this Ars Technica article, the US Trade Representative, the body in charge of negotiating the Anti-Counterfeiting Trade Agreement (ACTA), has a section open for comments about the treaty. Let them know that you favor an open Internet and balanced copyright laws. I've copied my comment to them (which was too long to fit in the form, blah) here.

Letter to the USTR

I am very concerned with not only what leaked content I have been able to find about the Anti-Counterfeiting Trade Agreement (ACTA), but also with the lack of public involvement and visibility of this far-reaching and dangerous piece of legislation. For something that has a profound potential to damage the way the Internet, free speech, and the development of culture work, it is unacceptable that the only interested parties who have been allowed into the negotiations are those with the greatest interest in restricting new outgrowths of culture that the Internet can provide.

The companies that make up the Motion Picture Association of America (MPAA), Recording Industry Association of America (RIAA), the Business Software Alliance (BSA), the Association of American Publishers (AAP), and others currently have far too large of a sway over the decisions involved in the treaty, and represent a very lopsided view of copyright and patent laws. Despite their widely-publicized claims of the massive damages caused by piracy, there is little actual evidence in support of, and a great deal of evidence against their conclusions. The movie industry, especially, has been posting record profits, so the claim that innovation and creativity will be destroyed by Internet-based piracy is clearly false.

These industries' support of stricter copyright and patent laws is little more than an attempt to persuade the government, Internet service providers, or website operators to foot the bill to support their own business model. After having realized that directly suing their own customers was prohibitively expensive and generated a huge amount of bad will, these companies decided that they could pass the unpopular job of attacking music and movie consumers to someone else.

It is not the US government's job to support the business model of these few companies. The small, if any, increase in economic activity that might result from stricter laws would not nearly be worth the chilling effects on cultural expression, free speech, and innovation. Many of the companies most heavily involved in the ACTA negotiations are products of an industrial age in which reproducing and transporting information is a difficult task and requires a large, organized infrastructure in order to be feasible. With an increasing number of homes now having high-speed Internet access, this difficult problem of transporting information no longer exists. As such, a large component of the reason for these companies to exist has disappeared as well, leading to an existential crisis for them. These companies are the candle makers and buggy-whip manufacturers of the 21st century, and rather than allowing them to continue to seek economic rent from the former audience turned participants in culture who no longer need them. Even if the claimed "worst case" does come to pass, and these large media companies do fold under the weight of Internet piracy, they will not be missed, as their once-useful purpose has been superseded.

It is extremely important for the future of Democracy, freedom of speech and expression, and the continued development of our culture that the negotiations be opened to the public, the full and current version of ACTA be made available, and those companies whose interests run counter to those of the country at large not be given undue influence over one of the most important treaties of our time.

Monday, February 8, 2010

Copyright: You know the drill

This is a reply to my father's reply to my earlier post about copyright. Here, for your viewing pleasure, is my full response.

Let me first just say that I think we agree much more than we think we do, but the verbiage that we are using construes our positions as being more antagonistic than they actually are.

Well, I cannot accept the premise. If one does not own one's intellectual property, then there is a major disincentive to creating it.

Theoretically, but how does this play out in practice? There is a certain amount I would need to expect to be able to earn off of something in order for me to prefer that activity to some other form of employment. One way of trying to hit that minimum of income would be to grant me exclusive use of the thing that I have created, and allow me to sell it. This is the strategy behind copyright. On this point, we (and essentially everyone else) agree.

There are various other methods of inciting me to do the work, such as patronage, donations, and grants. Many of these have worked in the past (and continue to work) for certain kinds of works (music comes to mind), but do not appear to be general solutions to the problem of inciting people to produce the kinds of works that we as a society are interested in.

The issue that most people have with copyright is what happens when the incentive to create a particular work rises well above what its value is to society? If I am to write a program, and it will cost me $50,000 worth of opportunity cost to do so, what happens when I am paid $500,000 for it? Clearly, if I was willing to write the program for $50,000, then someone is over-investing and is wasting money. The problem that people have is that the costs imposed by the current copyright laws are higher than what is required to incite the works produced.

There is also the added "cultural morality" notion that applies mainly to things like movies, music, and books, that says that works that are old enough become a part of the culture, and claiming ownership of them is unethical. A prime example of this is the fact that the "Happy Birthday Song" is still under copyright, even though it could be considered a basic part of our culture at this point. I find this argument reasonably convincing, but believe that a strong case can be made against the current scope and duration of copyright protection even without invoking matters of morality.

I do not read radiological studies for free. If I could not charge for it, I would not do it. I would (have to) find another line of work.

I agree. Nobody (well, except for the VERY fringe people) is suggesting an abolition of compensation for intellectual endeavors; they are merely saying that past a certain point, extremely long copyright terms lead to a kind of rent-seeking behavior.

An analogy to radiology would be that you wouldn't expect someone to continue paying you for the rest of their life for use of their brain after you had performed a radiological examination. They will pay you enough to cause you to perform the exam instead of seeking another line of work, but once you have performed the service, the payment relationship ends. This analogy is by no means perfect, since a third party would have no use for a diagnosis made of someone else, but the general idea applies.

So performing intellectual work, as a job rather than a hobby, must include the ability to be compensated for it.

Yes. Absolutely. I completely agree with this, and as time goes on, this fact will only become more important in an increasingly abstract-creation-based economy. The challenge is determining how much to compensate and for how long someone holds a monopoly over their creation.

The notion underlying this analysis is that new intellectual work is of such limited importance that all increments in units are of constant value. Hence formulations such as
"the term of a copyright must not be one second longer than is required to induce a creator to produce new works"

I disagree with your assessment of that statement. I believe that a copyright term should be as short as possible so that the creation can be re-integrated into the common pool of knowledge and reused without transactional friction by as many people as possible.

What I did not make sufficiently clear in that quote is what kind of "new works" to which I am referring. An ideal copyright system would assign a specific level of compensation to each individual work which would exactly match the amount of money needed by the creator to incite her to work on that creation rather than some other endeavor. Somehow, society would determine which new works were worth supporting, and grant monopoly privileges only to those which were worthwhile, or at least only subsidize an activity up to its value to the whole society.

Obviously, no system will ever be able to perfectly predict what amount of compensation is needed for each individual creator, nor will it be able to accurately predict ahead of time what the value of a new work is to society, but this is the ideal toward which we are aiming. The job of determining the value of a work to society is left to the market, which is (like anything) imperfect, but the best mechanism we currently have of determining value to a wide group. The other factor, the amount that the creator needs to prefer working on her creation rather than something else is currently solved by the promise of exclusive control over that work for 70 years after the creator's death. My argument is that life + 70 years is drastically over-estimating the compensation required to incite the creator to spend her time on the project.

I am not at all interested in the minimum incentive that will get one more copyrightable one-column post on a newspaper site. The sort of thing that could be thrown off in the time it takes to type may get copyright protection, but does not need it. No one cares whether it is published, and no one has any reason to use it at all.

I agree. One of the hardest problems is that the compensation required to incite a one-column newspaper article and the development of a distributed, soft-real-time database are totally different, yet the laws to support both of them are essentially the same.

I am very interested in the incentives that would lead to important and substantive contributions. These may require sustained long term effort by people who are able to provide this only because they derive their income from this work.

Absolutely. One thing to consider, though is that none of the work that these people are doing occurs in a vacuum. Essentially everything that is done today (and this trend is only accelerating with the increased ubiquity of the Internet and accessible programming tools) is built off of the work of people who came before. If the cost of integrating and building off of existing work is lower, then each person can do more with less. The ubiquitous availability of blogging software has meant that anyone who wants to blog no longer has to be a programmer, leading to more time spent blogging and less time spent rewriting blogging programs.

So I would not equate an improved design for a hybrid engine, which might save billions of dollars in transportation costs with some rant posted on the FoxNews website.


If one takes the "not one second longer" criterion, and the structure of the legal system requires that the duration be the same for all, then you would wind up with zero for the duration of protection.

This is where the heart of the problem lies. My position is that the copyright duration for new hybrid engines should be "not one second longer" than is required for the new hybrid engine to be produced, while the copyright duration for a rant should be "not one second longer" than is required to incite the production of the rant. Essentially, I want to do away with the "and the structure of the legal system requires that the duration be the same for all." This is not an easy problem to solve! However, I think that the current term is far longer than is needed for even the most involved of production.

People will produce trash for free, and a huge amount of it. If you make the standard the minimum required to produce that, then this will drown the much smaller number of truly productive accomplishments.

Yes, with the caveat that numerically speaking, there will always be much more trash than useful stuff, but you always have the luxury of ignoring the trash. This is not a dismissal of the concern of the loss of quality, involved creations, but a side-note against the distress that people have when they see a lot of junk. For instance, a common criticism of Twitter, Facebook, and blogs in general is that "most of the stuff there is crap." Yes, this is absolutely true, but especially in the Internet age, it is trivial to completely ignore the crap and only look at the worthwhile content. What would be a problem is if a great deal of crap is produced to the detriment of the production of worthwhile content.

If you cannot protect your creations legally, then you will be forced to hide them. Instead of patenting your hybrid design, and selling it. Making the servicing and repair widely available would benefit society. The alternative is to put the engine in a sealed compartment, and require them to shipped to the factory for all service. You would have to build it so that unauthorized attempts to open the unit would destroy it. That way no one can buy one, take it apart and copy it, since this would be legal. Of course, the benefit to society of the more efficient engine would be reduced by the costs of replacing them, shipping them back to the manufacturer, etc. You could get to this situation although you still had huge numbers of short articles published online, so you can induce A creator to produce ONE more work, in her spare time, for free.

Buried in there is an argument for more open sharing of information. You mention that "making the servicing and repair widely available would benefit society." This is the key balance to find. You want there to be enough of an incentive for the manufacturer to produce the engine, but it is also more efficient if anyone can service the engine without needing special permission. Obviously, there is the matter of being qualified to service the engine, but this is a matter of certifying someone's expertise, and not of restricting the knowledge or use of the knowledge.

Also, you are making a "benefit to society" argument here, and the only area in which we differ is specifically how to tweak things to achieve that maximum benefit. Again, our positions are not as different as they seem.

This is also more of a patent issue than a copyright one, and while the two issues are similar (and derive from the same section in the Constitution), there are key differences which make it dangerous to conflate the two.

If you cannot keep something as a trade secret, and there is no legal impediment for others to copy them, then it has no value to the creator. Even though it may have huge value to society. So no one will do it.

I wouldn't say that it has _no_ value to the creator. There have been numerous businesses (including my beloved Open Source companies) which are able to find value in giving things away for free. An (admittedly contrived) example would be that the engine manufacturer gives away the designs for the engines for free, without copyright protection, but by employing the engineers who designed the engine, would be in the best position to service the engine when it needed work. In that case, the engines would essentially be loss-leaders for the repair service.

Now, this is not likely to work for the auto industry specifically, but this is how a great many Open Source companies (including Red Hat and IBM) sustain themselves. The software is essentially given away for free, and the money is made off of customizing and managing that software installation.

Consider the tragedy of the commons. Say you would grow much more grass and hay in the pasture if someone were to fertilize, water and rotate crops on the land. But you insist that the only claim on the result is the minimum that will incent anyone to do anything. One may well say "once my cattle are fed I will NOT set the field on fire. Therefore there is currently enough incentive to generate the lowest possible level of socially useful activity. Therefore there should be no further payment to the person who will care for the field." Of course under these circumstances, no one will fertilize or water, society will be worse off, but the "not one second longer" standard would be upheld.

My position is not that the duration should be set to "generate the lowest possible level of socially useful activity," but to generate the highest. Going in the opposite extreme, I could say "in order to feed my cattle, I will maintain exclusive use of the fields for one year, extendable to 70 years after my death." This might benefit me (assuming I took well enough care of the fields that my cattle could survive), but would be under-utilizing the common resource of the field. I do not need exclusive use of the entire field in order to raise my cattle, and by preventing other people from using it, I am hurting everyone else for little to no benefit to myself.

This does not begin to address the moral element. We do not have laws against stealing only because an empirical analysis of the economy suggests that those with laws against stealing grow faster than those without. I suspect this is true, but it is possible that societies with fewer laws, or lax enforcement, grow rapidly. We have laws against stealing because stealing is wrong. We would still have laws against stealing, even if one could show that stolen money has higher velocity and provides more fiscal stimulus. At least I hope so.

This is a very different argument, and I hope that it does not get conflated with the other points I have made, because I believe they are able to stand on their own without invoking the moral aspect.

That said, I would say that, even though it is hard to imagine, if a society with ubiquitous stealing actually was better for everyone than one without, I would certainly prefer the former. Again, it is difficult to even conceive of a situation in which this might be true, but if it were, I would choose the option which left me better off. If I were somehow better off being able to freely steal from others, with the knowledge that they could freely steal from me, then I would choose that system. I don't care so much about the particular items and property which I own, but what I am able to do with them. If I can do more with the same amount of property by letting others "steal" it from me, then that would be the obvious choice, to me.

In the case of digital music, one could certainly argue that this has already happened. The rewards to issuing a new CD have fallen. Although the cost of simply recording music and burning it onto a CD also has fallen, there is simply less reason to do it.

I'm not actually sure about this. It is hard to get data on this, but one large (possibly even primary) reason for the decline in CD sales (aside from the shift to online music) is that most of the people who are interested in purchasing back-catalog music on CDs have already done so. When they first came out, there was a windfall of people repurchasing the old music they owned on cassettes or LPs on CDs, which boosted the sales numbers. This was clearly unsustainable, as eventually people would have purchased all of their old music on CDs, and would only purchase new music from that point forward.

The record industry has not clearly shown statistics of the relative sales of new releases versus back catalog music over the years, but the general attitude among those who follow this more closely is that the decline in CD sales is mostly attributed to the end of the back-catalog windfall, rather than piracy.

One outcome, which I contend is already underway, is for musical performers to release fewer CD's, and leave fans to the much lower quality of bootleg recordings from live events. If you cannot be paid for creating a studio performance, why do it?

Well, one of the main reasons for artists creating fewer CDs is that the current state of recording industry deals means that artists stand to make vastly more money off of live shows than from recorded music. An artist makes about 7x as much money from live shows as they do from recorded music , and that ratio is increasing.

Hence the greater focus of musicians on live touring and much less on recorded music. As time goes by, the rational performer would never create a recording of a full song. They might record the refrain and use it for marketing, but a full CD? Are you crazy? The mere existence of such a recording would be equivalent to giving it away.

Yes, that does seem to be where things are headed, though this is largely the result of the recording companies taking 90% of the revenue of recorded music, leaving 10% to the artists, versus the 90% of the revenue which goes to the artists from live shows.

So if you want to hear this performer, you had better show up in person.

I am personally kind of bummed by this, since I'm not all that interested in live shows.

Now I am not clear on why society is better off as a result of this, and it is clear that vigorous enforcement of laws against stealing would have prevented this from occurring.

Not really, and actually the opposite seems to be true. It is primarily the recording companies who are seeking increasingly strict copyright enforcement, whereas artists seem to do better with less enforcement (which allows their recorded music to be more easily promoted, serving as an advertisement for their live shows).

Fortunately for me, I am not looking for anyone to create new music. Essentially all the music I am interested in hearing was composed long ago. And once there is one recording available for legal purchase, I can buy it and listen to my heart's content. If I were a fan of contemporary music this would be a disaster.

I do worry somewhat about a shift away from recorded music, though that doesn't seem to be happening (since obscurity is a far greater threat to artists than piracy).

And all because of a de facto legalization of music stealing.

No, more like a set of increasingly artist-unfriendly contracts for recorded music, coupled with an increasing interest in live shows (which I don't get, but still).

These arguments also do not address the "clean up the streets" argument. If you lock up those whose respect for property rights is so weak that they feel no impediment to stealing music then you leave a more honest set of people walking around free. If you encourage stealing, then those who steal will prosper, and you will have a less honest society.

This is back to the separate moral argument, but I'm not that convinced by the "clean the streets" argument. To me, it seems a bit like (a more moderate version of) declaring something common and generally accepted as illegal, and then using that as an excuse to lock everyone up. Imagine if you made "saying mean things about the president" illegal, as is the case in Indonesia. You could then say "we need to clean up the streets of all those people who have so little respect for the president that they feel no impediment to saying mean things about him." Just because something is illegal, doesn't mean it's wrong. It also doesn't mean that every instance of law-breaking is an act of protest in the form of civil disobedience.

I could not care less about what a bunch of ignorant, racist, sexist, alcoholic, landed gentry who never did a lick of work in their lives (the framers of the Constitution) thought about the purpose of copyright. Not a one of them could have created anything of value. The idea of doing useful work was so alien to them that I am sure someone had to be brought in to explain the concept. Work was for servants. Laws were to protect the interests of the aristocracy.


Copyright in Context

This is my response to this Ars Technica article.

If you look at all of the foundational arguments for copyright, they are about a benefit to society, while only secondarily being about the authors or creators. Since the beginning, copyright, or a monopoly on ideas, has been regarded as an acceptable evil in the pursuit of the creation of new works. Copyright, like any enforcement strategy in a social-contract-based form of government, is a deal between the governors and the people which is only acceptable if the restrictions on personal autonomy are outweighed by the benefits that the restrictions bring.

Currently, there is little thought given to whether a copyright term that extends 70 years after the author's death actually makes the author more likely to produce a new book. Whether or not "Steamboat Willy" is still restricted by copyright has little bearing on whether Walt Disney produces more cartoons, because he is dead. According to the constitution and every western (possibly others as well) system of copyright ever, the term of a copyright must not be one second longer than is required to induce a creator to produce new works. The author's "property right" has never been a basis of copyright law and was even explicitly rejected by the Supreme Court of the US in 1985.

The problem nowadays is that wealthy corporate interests (not to beat the "rah, corporations r teh evil!!!11" drum too much) have a large vested interest in extending the copyright term as long as possible. Since their primary motive is profit, they don't care whether the money they make is on new creations or the back catalogue; they are perfectly content to resell "Snow White" over and over indefinitely, regardless of whether doing so encourages the production of new works. Indeed, a compelling argument could be made that continuing to sell 50+ year-old works actually decreases the incentive to produce new works, as it crowds out any new content and reduces the money available to finance new creations.

Friday, February 5, 2010

Just to make your lives difficult...

I finally bit the bullet and renamed my GitHubaccount from "dhax" to "haxney". I did
this to increase the consistency of my online names, so hopefully things will be less confusing overall.

Unfortunately, this means that all existing URLs to my old "dhax" repositories are broken, and GitHub unfortunately doesn't offer redirection for changed usernames. Though, to be fair, the Git and SSH protocols probably don't support any fancy redirection anyway, so there isn't much they could do. Plus, just offering the ability to change an account name while still preserving all the links and info is noteworthy in and of itself, so they should be commended for that. Yay GitHub!!

Anyway, if you find any place that has a broken link to my old "dhax" account, please let me know so I can fix it.

I figured I would do this now, before anything of mine gains much in the way of popularity so that in the unlikely event that anything should gain some degree of usefulness, I will have avoided a much more painful switch later. We'll see if it pays off.