One of my commenters reports that he showed my essay on evaluating the harm from closed-source software to Richard Stallman, who became upset by it. It shouldn’t be news to RMS or anyone else that I think he’s a fanatic and this is a problem, but it seems that every few years I have to explain the problem again. I make the effort not because of personal animus but because fanaticism does not serve us well – we’ve made huge progress since 1998 by not repeating RMS’s mistakes, and I think it’s important that we continue not to replicate them.
When I was say that I judge RMS is a fanatic, I mean something very specific by that. I cite Santayana’s definition: “Fanaticism consists in redoubling your effort when you have forgotten your aim”. By his own account of his road-to-Damascus experience, RMS started out attempting to solve a problem; there was this broken printer driver that he couldn’t fix because he couldn’t get the source. RMS correctly identified source secrecy as a damaging practice leading to bad outcomes.
Unfortunately, RMS made an early decision to frame his advocacy as a moral crusade rather than a pragmatic argument about engineering practices and outcomes. While he made consequentialist arguments against closed source (and still does) his rhetoric and his thinking became dominated by terms like “evil”, to the point where he repeatedly alienated potential allies both with his absolutism and his demand that anyone cooperating with him share it.
I think this is precisely the sort of displacement Santayana had in mind – means overwhelming ends, rhetoric taking over and trapping the fanatic in a position where the harm he was originally reacting against is forgotten. Instead the language of revelation, virtue, sin, purity, corruption, and redemption dominates. RMS parodies this aspect of his own propaganda when he presents himself as “St. Ignucious”, but the parody does not banish the fact that he is in fact living the role of ascetic holy man bent on purging sin from the world.
There are some advantages to this strategy. It taps into old, powerful emotional responses in human beings – the same responses that give messianic religions their power. As a way of recruiting a small hard core of dedicated followers it’s tough to beat, and sometimes – if you’re, say, the Gautama Buddha or Jesus or Mahavira – you can make it scale up. But I described it as a trap for a reason – most such attempts do not scale, remaining tiny marginal cults.
By the late 1990s, after having observed RMS’s behavior for more than a decade, I had long since concluded that the Free Software Foundation’s moralistic rhetoric was serving us badly. The problem with it is the same problem with messianic religions in general; for people who are not flipped into true-believer mode by any given one, it will come off as at best creepy and insular, at worst nutty and potentially dangerous (and this remains true even for people attached to a different messianic religion).
I was not the first or only person to diagnose this problem, and note that it was severely damaging our ability to talk people outside the hacker community into giving up code secrecy. I was the first person to devise a solution, an entire discourse that could compete with the FSF’s – the rhetoric of “open source”, and determinedly pragmatic arguments for it centered in engineering and economics.
Fifteen years later I think it is clear from results that teaching the hacker community to stop alienating potential allies with terms like “evil” and the rhetoric of sin and redemption was very effective. Understanding that RMS is a fanatic matters, because it reminds us that we have achieved an unprecedented measure of mainstream success by not replicating his rhetoric and his mistakes, and that we need to continue not to replicate them.
It shouldn’t need saying, but this criticism is not personal. I still try to be a friend to RMS on the rare occasions that he permits it. He’s done amazingly good technical work, and is without doubt one of the heroes of our culture. He has the virtues of his vices; he’s a man of unshakable honesty and integrity. But for the sake of the future – indeed, for the sake of RMS’s own original objectives – I have to call him on his fanaticism. There is too much at stake for me to be diplomatically dishonest about this – it did immense damage to the cause of openness, and I had to spend a good many years remediating that damage.
Its is still theoretically possible that RMS and the FSF could clean up its act. A good first step would be to stop characterizing people who refuse to use the rhetoric of moral evil as unprincipled and traitorous. It would be better to drop the quasi-religious rhetoric entirely. But I don’t expect this to happen; too much history and personal investment locks RMS and the FSF into their position. Thus, I expect to have to keep pointing out periodically that it’s fanaticism and that such fanaticism does the open-source community more harm than good.