Slashdot Mirror
The Economics of Open Source
Jason Kau writes " is a working paper on the economics of open source software from the Nation Bureau of Economic Research entitled "The Simple Economics of Open Source". Focuses primarily on Apache, Perl, and Sendmail but mentions Linux, Debian, VA Linux, etc. It's a 40 page PDF document. Some background in Economics would probably be helpful."
My first impression of this paper is that it is pretty well written for the audience it was intended for. As an Economics major (who is also a full time mainframe system's programmer, wierd huh?) I have run into some strange attempts by my professors and others to try to explain open source in economic terms.
:), the general gist of the paper captures the economic implications of open source pretty well. The section on corporate reactions to open source is very interesting, presents it pretty well.
Most of them get it wrong, assuming that there has to be some form of tangible gain from contributing code. The very science of economics tells us that basically there is no free lunch, and that people will not give up resources (in this case free time) when they do not recieve something of value in return. The recent rash of IPO wealth has sort of supported this belief, despite the fact that it was unexpected (unless this whole thing has been a clever ploy of Linus and ESR's to get rich slow)
While yes, they did get some aspects wrong (what do you expect, they are economists
Finkployd
Because of the current economic disinformation campaign (also known as the US Presidential Primaries), I decided to hit some of my old econ texts. I was quickly reminded that all of economics can be distilled down into a few simple premises (with extremely complex interactions).
One of the basic premises is that a voluntary transaction will not occur unless both sides benefit. Apache itself is not sold but can be obtained for gratuis because it is worth more for the Apache developers not to charge for it. If it were worth more to them to sell it for $50 a copy, they would do so. Since they don't, it isn't.
The reason that it is more valuable for the Apache developers to release Apache at zero monetary cost is because they receive other values for it instead. Another economic principle is that costs and benefits are not limited to money. The various benefits that ESR lists for open sourcing a project, ego stroking, good will, sense of community, etc., are just as economically valuable as monetary payments.
The last point is very important, and one that some people completely over look. If money is less valuable than "good will" or any other ephemeral payment, then neither are these ephemeral payments any more valuable than money. We cannot economically judge one to be more important than another. Thus, to deny a corporation the ability to sell its software products, which is what it wants to do, is to create an economic loss for society. Note that when I say "sell software", I mean in the copyright sense of selling rights to it or selling undistributable copies.
Some folks in this community hold that it is morally wrong to sell software (see above note). Economics cannot judge whether something is moral or not, that is left to other professions. But it can say whether some policy creates losses in value to individuals or society. A mandatory open sourcing of software (whether by government decree or societal pressure) will create net economic losses. Economically, it will become more valuable to a developer to forgo opportunities to create new software in favor of waiting on tables instead.
A Government Is a Body of People, Usually Notably Ungoverned
Actually the 'common wisdom' seems to be correct here. You have to remember that GPL operates under *copyright* law. Thus if you are not copying ... there is no restriction at all
That's the thing about common wisdom, it *ALWAYS* seems correct. It this case, however, it is not even a close question:
If you use computer software, you are copying. Three Circuit Courts cases have held unequivocally (MAI, Southeastern and Apple) that the loading of a computer program from any media into RAM, and the subsequent execution of that program from constitutes reproduction under 35 U.S.C. s. 106. There exists no cases holding to the contrary.
Until the Congress changes the law, or the Supreme Court opines otherwise, unlicensed use of software constitutes Copyright infringement.
Copyright has nothing to say about what you do with your copies
You can't imagine how badly mistaken is this view. Copyright law provides specific exclusive rights in Section 106. Unless you are granted consent, or can find an exception in sections 107 through 120, you are infringing. This is true even if you are the owner of a copy. Ownership of a copy (which is distinct from possession of a copy) does grant certain rights set forth in Section 109 and in the case of software section 117 of the Act. Neither provides a general right to reproduce, and hence, to use, the software.
The common wisdom, as you have stated it, is clearly in error. See a lawyer before you rely on it.
Try
http://www.people.hbs.edu/jlerner/publications.
for the free download of the working paper.