indessed roscivs

My reasons for disliking the GPLv3 are simple: the requirements it introduces go above and beyond protection of Stallman’s original four freedoms, and attempt to control other things beyond the scope of these freedoms. For example, some corporations were taking copyleft software and putting it on hardware devices that would detect modifications to the software and refuse to run any modified code. To Stallman, this is a violation of Freedom 1, since adaptations of the program will not work on that device. However, you are perfectly free to run your adapations on another device or on a modifed device, so I consider Freedom 1 intact. The restrictions of the GPLv3 are attempting to enforce further freedoms above and beyond the original four—hardware-related freedoms, not software-related freedoms.

Furthermore, the GPLv3 has aggressive “patent retaliation” clauses.^[1] When originally developing the General Public License, Stallman was worried about the situation where someone might hold a software patent for a bit of Free Software that someone else wrote. This would in everyone losing the Four Freedoms (including the original author of the software!) except for the patent holder, who could do whatever they wanted with the software, including charging money to allow others to use it.

Stallman wisely placed a clause into the GPLv2 stating that “any patent must be licensed for everyone’s free use or not licensed at all.”^[2] In other words, if I have a software patent that covers something in Free Software that somebody else wrote, I am not allowed to distribute that software unless I provide a free license for anyone.

The GPLv3 goes much further than this, causing the distributor to “lose their license and any patent licenses that accompanied it.” For example, let’s say I’m a large software company with many patents, some of which could possibly apply to some Free Software that somebody else wrote, which I distribute. Meanwhile, a competing large software company writes some totally different software that obviously violates my patents, and I decide to sue them. The competing company decides to point out this piece of Free Software that I’m distributing, and claims that my patent applies to that piece of Free Software, and hence I must provide a patent license “for everyone’s free use or not at all.”

If the court decides that the patent in question does indeed apply to this piece of Free Software, then if the software is licensed under the GPLv2 I can simply stop distributing the software, pay copyright violation damages to the author of the software, and continue on with my lawsuit against my competitor. If, however, the software is licensed under the GPLv3, then if the courts decide that I “knowingly relied” on the patent license, then I automatically lose my patent and cannot enforce it against anyone, ever again.

This is exactly what Stallman wants, because he believes that software patents are a blight on humanity and should be abolished. I happen to agree on this point, but I disagree that such retaliatory features should be part of a general-purpose software license. This goes way beyond trying to protect the four freedoms for a particular piece of software. Instead, it uses the license as a weapon against an unrelated battle against software patents in general.

While I agree with the battle, and will fight the good fight, I do not choose to use the software I write as a weapon in that war. That is why I will not use the GPLv3.

In 1999, Richard Stallman penned an essay that is now known as The Free Software Definition. In it, he details four requirements, or “the four freedoms,” which are requisite for software to be considered “free” (as in liberty, not as in price):

• The freedom to run the program, for any purpose (freedom 0).
• The freedom to study how the program works, and adapt it to your needs (freedom 1).
• The freedom to redistribute copies so you can help your neighbor (freedom 2).
• The freedom to improve the program, and release your improvements to the public, so that the whole community benefits. (freedom 3).

(Access to the software’s source code is a prerequisite for freedoms 1 and 3.)

There are two types of Free Software licenses. The first type simply provides the recipient of these software with these four freedoms. The second type requires that anyone who distributes this software must do so with a license that guarantees the four freedoms. The second type is called “copyleft” (a pun on “copyright”).

The difference is easy to explain. In the 1990s, Microsoft took the source code to some Free Software called BSD, and copied some of that code into Windows in order to allow Windows to connect to the Internet. But when Microsoft distributes Windows, they do not provide the recipient of Windows with those four freedoms. This is a license of the first kind.

In 2002, Apple took the source code to some Free Software called KHTML, and copied some of that code in order to make Safari, the default web browser on Apple systems. When Apple distributes Safari, they are required to provide those four freedoms for the part of the KHTML code they copied. In other words, they provide the source code, and allow everyone the freedom to do with that source code the things listed above.

I really like the Four Freedoms, and I distribute much of the software I write on my own time under a copyleft license that Richard Stallman wrote, called the General Public License (or GPL) version 2.0. However, on June 29th, Richard Stallman released a new version of this license, the GPL version 3.0, which he will now release all his software under. This license is much longer, much more complex, and has many more requirements than version 2.0, and many people in the Free Software community are questioning the wisdom of the license—including me. I will likely not use the license for any code I write myself, and will be very hesitant to contribute source code to GPLv3 projects.

In my next post, I’ll talk about my reasons why.

A recent poll revealed that about half of Americans don’t “believe in evolution”. (Interestingly enough, a quarter of Americans believe that both evolution and creationism are correct.)

But such polls always make me wonder how they’re worded. “Evolution” is a very vague word that means a lot of different things to a lot of different people. I’d much prefer a different sort of poll where people were given statements like these and asked whether they agreed or disagreed with each one. I wonder if we’d see a considerably different picture of the American populace?

• The earth is (at least) millions of years old
• Life has been present on this earth for (at least) millions of years
• Strata dating is generally accurate
• Radiometric dating is generally accurate
• Fossils of species show up in different eras of the fossil record,
  with more complex species appearing later
• All life on earth is based on the same DNA=>RNA=>Protein transcription
  process
• All change between generations within a species is due to mutation,
  crossing over, and similar DNA recombination (i.e. other theories such
  as Lamarckism [heritability of acquired characterists] are false)
• Within a single species, there can be different “subspecies” which could
  interbreed but do typically not
• The ability of subspecies to interbreed is not transitive; in other
  words, there exist subspecies A, B, and C such that A and B can
  interbreed, B and C can interbreed, but A and C cannot interbreed
• Over time, two subspecies that could previously interbreed lose (for
  various reasons) the ability to interbreed and become different species
• The more recently two species split, the more of their DNA they generally share
• Vestigial structures (such as limbs on snakes and whales or degenerate
  eyes on cave-dwelling fish) give us clues to the ancestry of species
• It is theoretically possible to explain all the complexity of life
  (e.g. eyes, bat sonar, etc.) through some process of genetic variation
  and selection
• The theory currently with the most evidence that explains the origin
  of the species is that all species of life on the planet evolved from
  single-celled organisms

Do you “believe in evolution”? Are there any statements above that you disagree with?

In a recent book called the Blank Slate, Steven Pinker writes about how what society considers moral and immoral can change quite easily over the years—”moral emotions … can be turned on and off like a switch. These mental spoinks are called moralization and amoralization …They consist in flipping between a mindset that judges behavior in terms of preference with a mindset that judges behavior in terms of value.”

Certainly a lot of recent changes have been about things becoming morally acceptable. Pinker lists, for example, some things that have become talked about in terms of “lifestyle choices” rather than moral sins:

divorce, illegitimacy, working motherhood, marijuana use, homosexuality, masturbation, sodomy, oral sex, atheism … Similarly, many afflictions have been reassigned from the wages of sin to the vagaries of bad luck and have been redubbed accordingly. The homeless used to be called bums and tramps; sexually transmitted diseases were formerly known as venereal diseases. Most of the professionals who work with drug addiction insist that it is not a bad choice but a kind of illness.

But that’s not the entirety of the story. It’s not just a matter of society becoming more “permissive” or amoralizing all behaviors. Pinker continues,

for all the behaviors that have been amoralized in recent decades, we are in the midst of a campaign to moralize new ones. The Babbitts and the bluenoses have been replaced by the activists for a nanny state and the college towns with a foreign policy, but the psychology of moralization is the same. Here are some examples of things that have acquired a moral coloring only recently: advertising to children, automobile safety, Barbie dolls, “big box” chain stores, cheesecake photos, clothing from Third World factories, consumer product safety, corporate-owned farms, defense-funded research, disposable diapers, disposable packaging, ethnic jokes, excessive salaries, fast food, flirtation in the workplace, food additives, fur, hydroelectric dams, IQ tests, logging, mining, nuclear power, oil drilling, owning certain stocks, poultry farms, public holidays (Columbus Day, Martin Luther Kind Day), research on AIDS, research on breast cancer, spanking, suburbia (”sprawl”), sugar, tax cuts, toy guns, violence on television, weight of fashion models

I saw on the elevator news the other day a perfect example of the results from one of the recent actions to be moralized: smoking. Pinker writes,

For many years the decision of whether to smoke was treated as a matter of preference or prudence; some people simply didn’t enjoy smoking or avoided it because it was hazardous to their health. But with the discovery of the harmful effects of second-hand smoke, smoking is now treated as an immoral act. Smokers are banished and demonized, and the psychology of disgust and contamination is brought into play. Nonsmokers avoid not just smoke but anything that has ever been in contact with smoke; in hotels, they demand smoke-free rooms or even smoke-free floors. Similarly, the desire for retribution has been awakened; juries have slapped tobacco companies with staggering financial penalties, appropriately called “punative damages.” This is not to say that these decisions are unjustified, only that we should be aware of the emotions that may be driving them.

The tiny tv screen in the elevator revealed to me that the MPAA will take smoking into account when determining movie ratings. In other words, if a movie is on the borderline between PG-13 and R, the presence of smoking that doesn’t “reflect the dangers of the habit or portray a historical figure” may tip the scales to R.

Like Pinker, I don’t think these decisions are necessarily unjustified, but at the same time, I wonder along with Pinker:

The question is whether they are best handled by the psychology of moralization (with its search for villians, elevation of accusers, and mobilization of authority to mete out punishment) or in terms of costs and benefits, prudence and risk, or good and bad taste. Pollution, for example, is often treated as a crime of defiling the sacred, as in the song by the rock group Traffic: “Why don’t we … try to save this land, and make a promise not to hurt again this holy ground.” This can be contrasted with the attitude of economists like Robert Frank, who (alluding to the costs of cleanups) said, “There is an optimal amount of pollution in the environment, just as there is an optimal amount of dirt in your house.”

Is the psychology of moralization the best choice in this case? Whether or not it is, I think this recent MPAA decision certainly shows that it is the choice we as a society have made.

Slashdot has a great article today about HBO’s CTO who thinks that the real reason people don’t like DRM (Digital Rights Management) is because of the name. Remember, DRM is the thing that makes it so that you can’t watch a DVD you bought in Europe here in the US. It’s the thing that makes music you download from iTunes Music Store not work on most mp3 players. It’s the thing that makes some music CDs not work in your car’s CD player. Basically, controls what you can do with content you’ve purchased, typically restricting your “fair use” rights under copyright law.

CTO Bob Zitter instead wants to use the phrase:

Digital Consumer Enablement, [which] would more accurately describe technology that allows consumers “to use content in ways they haven’t before,” such as enjoying TV shows and movies on portable video players like iPods. “I don’t want to use the term DRM any longer,” said Zitter

Of course, those of us who obtain non-DRM content can already enjoy TV shows and movies on portable players like iPods or Palm Pilots. DRM is what stands in the way of easy media transportation, not what enables it. This sort of bizarre language-changing is best seen by this Slashdot response (paraphrased):

I’m not going to call it piracy anymore. I prefer Personal Choice Enablement. PCE allows consumers (not customers, since you won’t be paying for the service) to enjoy content not only in ways they haven’t before, such as on portable video players like the iAudio A2, but at a more reasonable price than they have been offered in the past. This is also a win-win situation for the content creators as it alleviates all packaging and most distribution costs, as well as providing excellent word-of-mouth advertising for FREE!

Brilliant. Please, enjoy your new Personal Choice Enablement.

I am not a copyright abolitionist per se—that is to say, I don’t necessarily think that the optimal state for society is one with no copyright protections at all. However, I do believe that abolishing copyright would be more optimal than the state we are in today, where copyrights last 70 years past the author’s death.

First, let’s make sure we’re in agreement on what copyright is: a monopoly (primarily) on the right to duplicate a work. The US Constitution declares what the purpose of copyright should be: “To promote the Progress of Science and useful Arts”.

There are definitely some benefits to copyright, both to society and individuals. On an individual basis, copyright allows a way for a person to make a living from creative endeavors by charging for copies of their work (which otherwise would have a marginal cost of near zero). On a societal basis, we get more creative works produced because of the individual benefit.

I do not consider the individual benefit to be in any way the goal of copyright—it is only desirable inasmuch as it benefits society as a whole. If we enrich the individual at the expense of society, copyright has failed. If the total number of good creative works produced goes down, copyright is not doing what it’s supposed to be doing.

There are also some costs to copyright. The most obvious is the cost of enforcement—government agencies, private lawyers, groups like the BSA. The other obvious cost is that to purchasers of copyrighted works—the fifteen bucks I spend on a music CD is money I could have spent elsewhere had copyright not been in existence. A less obvious cost is to those who cannot afford to pay for the copyrighted works and must go without (or risk fines and legal action).

Another commonly overlooked cost is the decrease in derivative works—far fewer derivative works are created because of the licensing costs involved. This applied both to individuals on a small scale—such as unlicensed “fan fiction” authors who may write prolifically but can never put their name on their work for fear of lawsuits—as well as to large corporations—Disney makes most of its movies based on stories in the public domain or original works, rather than basing movies on popular copyrighted works, due at least in part to the licensing costs associated with the latter.

I see the tradeoffs of copyright as being most usefully represented as a graph, where the X axis represents the length and scope of copyright protections, and the Y axis represents the number and quality of creative works being published.

Naturally the graph isn’t linear–you get a much larger increase in Y when you go from X=0 to X=1 than you do going from X=99 to X=100. But Y certainly isn’t zero when X=0 (the false idea that copyright is necessary for innovation), but also I don’t think that Y is constant for all values of X (the false idea that copyright doesn’t affect invention and innovation at all).

The “sweet spot”—the length of copyright that I think is optimal—is when you take a look at the cost to society of X, compare it to the benefit to society of Y, and maximize the result: in other words, max(B(Y) - C(X)). Where is that sweet spot? I’m not sure. Some people think it’s as long as twenty or thirty years. I’m pretty sure it’s less than ten in today’s world, and very plausibly zero.

Edited to add: video sound is now fixed. My apologies.

My excuse for such lame posts recently is because I haven’t been feeling well. That’s my story and I’m sticking to it!