I never argued that F/OSS produces less buggy code, though I wouldn't mind betting a decent part of my knowledge of the field on it, as long as we are comparing apples to apples. Back to the matter at hand: I think dating the idea from Linus forwards is misleading because that is exactly when the idea became, to use a mild term, diluted, and distorted.
> it is why people likely to smugly point at the numbers on apache
> servers in the corporate world. See! They like us! They like us!
> They rilly 'n' trooly like us!~
I don't know if that's what they say, but certainly the ESR and O'Reilly (as well as Torvalds) types exhibit these tendencies. Which is my point as well.
> "great points, and i will re-emphasize that open source is not just
> about downloading and using the software, but participating in the
> community to whatever extent you can. add code. fix bugs. file bugs.
> write documentation. help other users. etc."
> http://mailman.lbo-talk.org/2004/2004-October/024518.html
That's me. And I stand by every word of it. To paraphrase: from each according to his abilities, to each according to his needs. And a liberal dose of liberty and community to go with it. No claims about release cycles, code integrity, corporate adoption, etc.
I really have been going way over quota. I will bunch up future responses.
--ravi
P.S:
The message from me (linked to above) includes this quoted section from Miles:
> Look at it this way: Open source software is a major kick in the
> intellectual groin of economists who bloviate about the inherent
> superiority and efficiency of capitalist production motivated by
> personal profit. Open source software demonstrates that you
> don't need to rely on capitalism to effectively accomplish
> social goals (e.g., serving up web pages to the Internet using
> Apache).
I do tend to agree with the above (with the caveat that I am no expert on the opinions of economists). In particular the part that suggests that Free Software (which is what I always mean even when I follow the thread and use Open source or FOSS, etc) goes against the reductionist model of deriving or basing all [human activity] on/from self- interest / personal profit. This does not make a claim that Free Software is inherently less buggy or more successful, but only that it negates the converse: "inherent superiority and efficiency of capitalist production".
Also, Wojtek writes: "everything else being equal, a managed organization can produce an overall better code and do it faster than a group of independent programmers working as they please.". But this is not the case. On the one hand, "managed organization" is really more mismanaged organisation, with layers of middle managers playing petty politics when not masturbating to Microsoft Project. On the other, few Free Software projects are (and more importantly, need be) a case of "independent programmers working as they please". Most importantly, there is nothing fundamental about Free Software that makes it impossible to carry out in a managed organisation (in particular: a managed organisation whose aim is not to maximise the profits of a few).