Steve Holden
Posts: 42
Nickname: holdenweb
Registered: Apr, 2003
|
|
Re: Desktop Linux Just Died
|
Posted: Mar 30, 2004 7:46 AM
|
|
Interestingly similar to a point of view I tried to promote in my own blog at:
http://www.artima.com/weblogs/viewpost.jsp?thread=7351
Most of the responses to that post appeared to focus on the fact that open source is anarchic (which I happen to believe is a good thing, up to a point), and that it deliberately avoids the commercial management style (which, up to a point, is also a good thing). But since nobody who responded seemed to believe that any change in the direction of more management was good, I just let the thread die. I've been in way too many discussions with other open-sourcers where the discussion effectively seems to be the point as far as some participants are concerned, and that gets old quickly. If I can make a difference to something I will contribute, but if I can't then I'd rather not waste my time or other people's.
I don't claim to be the world's best programmer, though my thirty-odd years' experience have certainly taught me to code with economy - what I describe as "programming with a scalpel" contrasted with "programming with a shovel". Since others are better-qualified to write the gcc's and other high-flown software, I tend to focus on contributing to the social milieu, by writing blogs, organising conferences, mentoring beginners, and so on. I like to think that these make an overall difference, even if there's no single epiphany to point to where some vast new concept was created.
One feature of an interaction that rings loud alarm bells in my mind is when conversations extend past the point when conversation is useful. I have steeled myself, when things get to that point, either to make an autocratic decision (if I'm in charge) or to walk away (if I'm not). Life's too short to waste time on endless navel-gazing.
If you look at the really successful open source projects (Gnu emacs, the Linux kernel, Python, Perl, ...) the interesting characteristic they all seem to have in common is the presence of what Python calls the BDFL, Benevolent Dictator For Life - someone to whom ultimately the whole project will defer if a decision has to be made and things have to move on. This isn't actually that much different from the Microsoft model that Rick mentions in his original post, with the distinction that in open source the goal is usually excellent software rather than profitable software.
Many people who profess "belief" in the open source movement appear to like it just the way it is. This seems wrong-headed to me. Sure, open source is producing good software now, but how much more, and better, software could it produce if we learned to manage our groups more effectively, and interact in a more controlled manner. The problem may be that there's a reaction against the tactics and management styles employed by big software producers, and the reaction is not well-enough focused so it throws the baby (effective project management) out with the bathwater (breadhead behaviors). If we aren't prepared to admit that things could be better we are doomed to repeat our past mistakes.
If the open source movement wants to produce better software more effectively, it might have to accept that there's something to learn from Microsoft, Oracle and other large software producers. If it wants to bury its head in the sand and ignore this obvious truism, all open source users will suffer as a result. I agree that the GNOME project (which, from my limited perspective, I already perceive as the winner of an internecine battle for the Gnu/Linux desktop) needs to retain focus amd move forward, and that open source does have the advantage that the codebase is available for forking by those for whom the chosen direction is unattractive or worse.
The acid test of community will come when the discussion gets to the point where all alternatives have been presented and discussed. If the navel-gazing continues, you may be right, and desktop Linux may well be dead. If we can't agree what the desired goal is, or if we can't agree to work together towards that goal, we will always trail behind a well-motivated (think "money") single-minded (think "money" again) commercial organization. Overall a focus on excellence is the way to keep moving forward. We need to be as obsessed with quality as the commercial producers are with profitability, and we also need to achieve an understanding that, although open source software isn't primarily produced for money, money is still needed to make it happen. Hence the various Software Foundations.
When I wrote the post I referred to in my introduction, I bemoaned the fact that there are far too many open source projects that get stalled at release 0.68 and never get finished. I'm less convinced now that is a Bad Thing (tm), but I still would like to see more effective project leaders in the open source world. There will be a lot more "release 0.68" projects that fall by the wayside. If they can contribute positive "genetic material" to other open source projects that follow, however, then they will have served a useful purpose.
|
|