Do not misread this essay as a fixation on money.
Economics deals with the collection of issues surrounding the allocation of scarce resources. There are a whole lot of different sorts of scarce resources, including:
The amount of bandwidth between the Free Software Foundation (where software might reside) and some other host on the Internet
The number of productive hours that Richard Stallman has available to him in a typical week
The amount of time available to the programmers that are intimately familiar with the LibreOffice codebase, and who are thus most capable to maintain it
The amount of disk space free on my hard drive
The amount of time available to people that are capable to write good documentation for GNU software
These resources are of varying nature and importance; all represent examples of scarce resources that can be relevant (to one degree or another) to decisions on how to best allocate scarce resources to development projects.
Money happens to be a useful abstraction of value that allows people to more conveniently trade off one scarce resource for another. By the use of money, we have a common basis for evaluation rather than having to figure off relative values for every commodity that could possibly be traded. As the "common basis," it is not surprising to see a lot of attention paid to methods of valuing things in terms of money.
Nonetheless, in particular situations, evaluations need to be done in terms of whatever resources represent binding constraints for the particular system. If we're trying to figure out whether people should spend time working on GNU Guile versus XFree86 versus the GNOME Desktop Environment, the critical resource is likely to be the time available to would-be developers. (And that set of choices is very 2002. In 2012, the debate might be between LibreOffice, GCC, Git, Ruby, KVM.) Money is not necessarily of great importance in this sort of decision. Having an extra million dollars around isn't likely to increase The number of hours that A Crucial Programmer To A Particular Project can spend awake in a week, for instance.
That being said, people tend to need to have food to eat, a place to live, clothing to wear, and other such things. Money is useful to free software projects if it can be used to free up time that would otherwise be spent on money-seeking pursuits (to satisfy those needs for food/clothing/shelter) that wouldn't involve building free software so that people can spend time building free software.
Can someone of the free software persuation tell me why, in order to do such a nice thing as create free software, it is necessary to completely trash the way I make my living?
|-- Scott Moore|
RMS (in the GNU Manifesto) is only cutting off your chance of becoming wealthy through a royalty model.
Only in the rather tiny sector where programmers write code and sell it would this be considered a severe hardship. Most professions in this world have no equivalent to the royalty model.
In those that do, very few people - who are commonly well-known media superstars - become wealthy.
Most programmers are not paid royalties for the software they write. It is vastly more common for programmers to be compensated via:
Neither of which happen to be royalties.
And it is, furthermore, not a dramatic overgeneralization to suggest that most programming is of this nature, and that, furthermore, most of the money that changes hands does not do so as a result of royalties.
" Will free software make the sale of proprietary software obsolete? Won't that injure my ability to make a living?"
If the use of free software grows to the point of seriously hampering the "spread" of proprietary software, then this obviously "does injury" to those whose livelihoods depend on revenues from the sale of proprietary licenses.
But that represents one very specialized sort of revenue, and certainly not the entirety of economic revenues.
This is not unlike the way that buggy whip makers were "injured" by the advent of automobiles. The growth of rail networks is "injured" every time another airport or highway gets built. Similarly, each time Microsoft releases another version of Microsoft Office, other vendors of "office productivity" software seem to get hurt.
None of these factors establish that the growth of use of free software will systematically injure either the economy in general or the computer industry in particular.
There is really minimal risk to the economy in general; if money doesn't get spent on proprietary software licenses, it presumably will be spent on other things. And the people that sell "other things" will doubtless be pleased at the increase in their sales.
Within the computer industry, it is really only the sales of proprietary software licenses that will suffer. Services, distribution, and hardware sales should not be adversely affected, and may even grow. The availability of functionality free of licensing charges can encourage increased use of computers, which certainly doesn't hurt these other sectors.
There seems to be a boundless ability of people to desire more and more functionality. (Much as they like getting newer and fancier automobiles, entertainment systems, pseudo-gourmet foods, and the like...) People will have to pay the costs required for that to get implemented.
Another common way that the fallacy gets worded goes along the lines of "Won't the government be hurt when it loses the tax revenues from the Microsofts and Oracles of the world?"
Even within that more limited context, it is easy to demonstrate the fallacy of the question. Yes, indeed, the government would lose the tax revenues from the royalty incomes. But this would be offset quite exactly by the fact that companies would no longer be deducting those very same amounts from their incomes as "software expenses." If Microsoft loses $1B in Office 2003 revenues, the would-be "loss of tax base" is offset by the fact that customers no longer spend, and deduct, that same $1B. The loss of royalty deductions may actually represent a net gain for government tax coffers.
It is often viewed that the presence of free software means that people will stop producing software. After all, why should anyone write software compilers if they have GCC to use?
This makes the implicit assumption that that the only alternatives are between:
Replicating the functionality of some existing piece of proprietary software, and
Being out of work.
The specific result of a cost/benefit analysis may change when some costs go away, but the notion of there being costs and benefit and some sort of balance thereof does not. If GCC is free, this may make it unworthwhile to undertake a project to build a proprietary compiler. But by the same token, the lack of cost of GCC makes projects practical that would be too costly with proprietary tools.
Rather than wasting resources replicating GCC, or paying someone else's "extortive rent" for the use of their proprietary alternative, an organization can use GCC, and have money left over to implement additional functionality in their systems.
In effect, when the cost of licensing software tools drops to zero, development projects that were too expensive because the tools were expensive wind up with a positive Net Present Value, and thus become worth doing. There may be no point to building Yet Another Proprietary Compiler; this is obviously going to put people out of work that lack the imagination to consider other options.
Free software is neither capitalist nor communist.
It may be regarded as rejecting that software licenses represent "capital;" that could be interpreted as a rejection of capitalism, but only political extremists seem willing to take this as a wholesale and total rejection of capitalism.
... Linux is proof of the system working. Capitalism (as practised in this country) is very open to other systems working inside it. All other systems can only make it stronger. They force capitalism to improve. Notice the flurry of companies trying to follow the Linux model for some sub-set of their products. Some really care about OpenSource, others are shamelessly trying to get attention. The market will sort it all out.
If a company was successful with another economic model in dealing with its "customers", then other companies would pay attention and react to attempt to keep those customers.
But notice, the State is not trying to shut Linux down. As long as you pay taxes to the State, the State simply does not care how you and your associates decide to have an economy. Other systems fear capitalism and strive to keep it out.
Try to run a large company developing software in a commercial for-profit way in a communist country.
Curiously, North Korea is pretty active in providing a whole variety of kinds of contractors, including building contractors and even software contractors, that are used a lot around Asia. (Notably, it appears, in Russia and China, who aren't nearly as dramatically politically aligned against North Korea as is the case in "The West".) The main merit of this seems to bear more of the marks of slavery than of anything else: North Koreans will work for nearly nothing, under dubious conditions, as they have little alternative. Quitting isn't an option.