As a programmer, I always felt unconvinced by the slogan that "all software should be free". Why should software be free when housing, education, groceries, books, movies, music and utilities are not? What is it about the product of my labor that requires it to be free when the product of other people's labor is not?

Note that

  1. The distinction between "libre" and "free" is not relevant here, because the former implies the latter, in practice -- just ask RedHat about CentOS/AlmaLinux. They really sell tech support, not libre software. It's impossible to sell "libre" software to consumers.

  2. The distinction between "rivalrous" and "non-rivalrous" is not relevant here either: Many of the products I mentioned are also non-rivalrous: books, movies and music are electronic these days. The cost of one extra consumer of TV cable? Negative! (watches ads, does not bother customer support)

This is the only part of the existing answer that's not just calling out these irrelevant distinctions (as of this edit):

because of the unique position software holds in running critical systems in our lives and our world.

But many other things are just as critical. Imagine living without books, housing or food, for example.

Why not demand that authors give books away for free and make money via donation pages, speeches and consulting fees instead?

  • I know your first, second and even third impulse is to downvote this question, but ask yourself if your downvote will change my opinion.
    – MWB
    Nov 30 '21 at 2:39
  • 3
    BTW, there are many countries where education is free, where housing and utilities are subsidized, and low-income households are provided with (vouchers to buy) groceries. Dec 1 '21 at 8:05
  • 1
    @MWB I rolled back your last change as it doesn't seem to me to add anything to the question, it's ad hominem, and above all, it's unanswerable, as it asks us here why some well-known person (who isn't here) hasn't said something. You already have a well-received answer to this question; may I ask that you either accept it, or indicate what (on-topic) questions remain in your mind to be answered?
    – MadHatter
    Dec 3 '21 at 7:49
  • 1
    I have edited my answer to address briefly your concerns about nonrivalrous goods other than software, as well as your concerns about selling support as an economic model. As you can see from the end my answer, I'm not sure how to address your added questions except as combinations of what I've already said -- please let me know if that still doesn't give you sufficient understanding of the relevant arguments. (You do not need to find the arguments personally persuasive to learn what they are from this Q&A site, of course.)
    – apsillers
    Dec 4 '21 at 3:33
  • 1
    @MWB I'm sorry you feel that way, but I disagree; the question you link to is a "what..." question about a single point of law, whereas yours is a "why..." question. Also, four people voted to close yours, and I was only one of them (albeit my vote counted as two). I feel we've tried quite hard to get you to tell us what on-topic question remains unanswered, without success; there comes a point where trying to get an answer that pleases the OP becomes unproductive. However, nothing stops you voting to reopen the question; if enough others agree, the closure will be reversed.
    – MadHatter
    Dec 4 '21 at 7:42

I suspect your question is predicated on the idea that universal software freedom might result in software development labor becoming economically worthless. This conclusion isn't correct. In a hypothetical world with complete software freedom, software is still valuable and money would still flow to developers in exchange for software development labor. (This money is separate from and in addition to any incidental money they might receive for support services or tutorials.)

First, be aware of the adage "free as in free speech, not free beer": free software, as the FLOSS community uses the term, may be both distributed and enhanced for any fee you set.

"Free software" does not pertain to price but indicates the freedom to copy and modify the software. Free software is an ideology that pertains to freedom to control the software that you rely on (rather than the other way around): it includes the freedoms to inspect it, copy it, and improve it

These actions aren't applicable to many of the commodities you list. Software is economically nonrivalrous: one person's use of some code does not naturally prevent another person's simultaneous use of that code. This is in contrast to rivalrous commodities like a car, which can only be driven by one person at a time.

Free software is not an economic ideology: it is completely agnostic to whether software is sold for a fee, or given away free of charge. Of course, allowing unrestricted redistribution of some piece of software renders obsolete most monopoly-based business models, which charge per copy, but this hardly means money cannot be made developing software that respects user freedom.

Your question asks about what economic leverage a software developer may still have over the product of your labor without leveraging a copyright monopoly. Take care not to conflate copyright with natural law. For rivalrous physical goods, the model is quite straightforward: you sell the good to a person so they may have it, instead of someone else. This physical arrangement doesn't apply to nonrivalrous software which may be copied endlessly at no disadvantage to any other holders of the software. Instead, we have legal instruments which invent a kind of legal restriction to imitate the economic reality that the physical world provides for rivalrous goods.

The ideology of free software does not say you may not make money from your software. Indeed, the truly rivalrous quantity in software is development effort: a developer can only do so much in a day, and there is untold demand for new software functionality to exist. Development effort is naturally, rivalrously valuable, and the principles of economics demand that developers get paid for what they do.

Those who say "all software should be free" mean that all software ought to respect users' freedom to view, modify, and share the systems that power their lives. In a prospective world where all software does so, software will still be valuable, even if it can be copied freely, and those who create and improve valuable software will still get paid for their labor.

Finally, I'd like to highlight that your question is one of ideology (i.e., you object to the notion that "all software should be free"), but this not an ideology shared even among all people who write free software. If you are simply looking for arguments from those who advocate the specific ideology "all software should be free", then the GNU Manifesto is a reasonable place to start, which also contains a modern footnote about the economics of free software development.

I should also highlight that you mention a few nonrivalrous things in your list: music and movies. For rivalrous goods, the reason they cannot be "free" in the same sense as freedom-respecting software is clear: physical goods cannot be modified and copied in the same way software can. But this doesn't hold for music and other media, which is similarly intangible and nonrivalrous.

I think some advocates for universal free software would also say other nonrivalrous digital goods should likewise be free to copy and modify. Others might not, choosing to draw the line at software only, because of the unique position software holds in running critical, life-or-death systems in our lives and our world: it can control how accessible and accountable our government is, the security and accessibility of our financial assets, and the ability to communicate with one another without interference or eavesdropping, to name a few examples.

Obviously, art is also a critical part of our lives, so some are in favor of making art vastly more free as well, and do not draw such a strong distinction. See Lawrence Lessig's Free Culture for an argument in favor of vastly reducing the scope and duration of copyright for art.

You say

Why not demand that authors give books away for free and make money via donation pages, speeches and consulting fees instead?

But free software -- even according to proponents universal software freedom -- does not demand such a standard of software developers. Selling support, or tutorials, or speaking engagements is one economic situation that may naturally arise by meeting the moral imperative to make software free. However, the unchanged economic reality is that the labor put into making software is valuable, and people and companies will pay developers to make software, even if no one can feasibly charge a fee per copy.

Consider the federation of developers who work on the Apache web server, who are well paid by their respective companies to make a software product that is ultimately given away free of charge. The companies all need the software to work well and meet evolving demands, so they all pay to have software that meets those demands. The status of Apache as freely-available, free-of-charge software does not harm the well-paid developers who continue to create it daily.

This argument does not rely at all on the idea of selling support or software-adjacent goods. I am speaking directly about the value of the labor of writing software, as you identify in the first paragraph of your question. Advocates of support-based economic models make a different argument than that one that I make here in this answer.

You also say

But many other things are just as critical. Imagine living without books, housing or food, for example.

Indeed, those things are also critical, but are either rivalrous (so the principles of software freedom categorically cannot apply to them), or else they are a nonrivalrous good addressed in the above section about art.

  • RMS's argument is basically "you can make money by e-begging and tech support", but wow, I noticed that he actually suggests banning stuff: "Low-paying organizations do poorly in competition with high-paying ones, but they do not have to do badly if the high-paying ones are banned."
    – MWB
    Nov 30 '21 at 17:23
  • While "libre" and "free" are different, the former implies the latter. Just ask RedHat about CentOS/AlmaLinux. RedHat really charges for technical support, not software per se.
    – MWB
    Nov 30 '21 at 17:26
  • 2
    @MWB While I think was RMS's view in the 1980s when he wrote the GNU Manifesto, his current view is closer to "If some economic benefit would occur from a particular piece of software existing, and it doesn't currently exist, then the future beneficiaries of that software will pay to make it exist, even without the ability to change rents for it." And: "While people will always continue to write software for fun, the odds that they will coincidentally write for fun the exact software you need is quite small; companies will pay for necessary software." See footnote 7 of the GNU Manifesto.
    – apsillers
    Nov 30 '21 at 17:53
  • 1
    @MWB Also take care not to conflate the price at which software is made available versus the money paid to the developers who made it. The Apache web server, for instance, can be acquired for zero cost, but the federation of developers who work on it are paid considerably by their respective companies to do so. Zero-cost software can be quite expensive.
    – apsillers
    Dec 1 '21 at 15:40

Not the answer you're looking for? Browse other questions tagged or ask your own question.