often exerts quite a bit of leadership over Cocoon because the committers
respect him. Creating formal owners is a great way to fork a community, but
it fails as a social experiment. Certain members in a community through merit,
skill and tact (I obviously am generally difficient in the latter) will be heard more
loudly than others.
Creating ownership or annointing "code leaders" is frowned upon by even some of
today's more formal software methodologies. Everything is everyone's responsibility.
I think guarding and guiding the door to committership on a project until the person
demonstrates cohesion to the group is a more effective method.
As evidence that the "control" idea doesn't work, please look at the state of modern software
developent within most Corporations. It sucks.
Linus, etc exhert a lot less control than you think.
-Andy
Ceki Gülcü wrote:
I keep wondering why you keep bringing up Duncan's Whoa Bessie... mail. I mean this one:
http://marc.theaimsgroup.com/?l=ant-dev&m=97712718421034&w=2
Is it just for historical purposes? Is it because Duncan expresses interesting ideas with eloquence? Sure, Duncan may have been wrong in the Ant context but that should not discredit his ideas altogether.
The liberal ideas expressed by Stefano, Sam and to some extent Costin are very inspiring and definitely please a wider audience than Duncan's ideas defending the actions of a selfish pig as he puts hit. (No, I don't think that Duncan is a selfish pig and you shouldn't either.)
However, liberal ideologies are just that, ideologies. While Duncan's theory of benevolent dictators might not find favor in the eyes of this public, we should not discard it as being contrary to the Apache way. We should instead recognize it as being a legitimate way of development. It may even be the dominant way of development at Apache under disguise.
In addition, it is much easier to stand up and talk about the interest of the community than the interests of individuals less you come off as supporting selfish pigs or being a selfish pig yourself.
On a wider scale, it was very hard for the West to fight Communism because the communist ideology sells much better to the unprivileged. Yet 75 years later, the West won, not because of its persuasiveness but because it had much more to show on the store shelves than the communists. Communism is a great idea but it doesn't work. Capitalism is hard to sell but it ends up having better results on the long run.
Coming back to Jakarta, I am not suggesting that anyone is at fault. All I am suggesting is that we to stop trashing the work achieved by individuals acting as clear leaders. Leadership is not bad per se.
I may be stating the obvious here. So be it.
At 11:01 07.11.2002 -0500, Sam Ruby wrote:
I differ with that rendition, and believe that it is harmful to the community for it to be propogated.
Duncan rejoined Ant and was immediately accepted as a committer. He started work on an internal fork named "AntEater". This went on for a short while, until another fork came along named "AntFarm". At that point, Duncan said "Whoa Bessie" and started to put forward a case that he had a unique right to determine what codebase bore the Ant name.
This lead up to a PMC meeeting with Brian and Roy in attendance where it was affirmed that the name of a project went with the expressed wishes of a majority of commmitters to that project. This has been the policy that we have followed in Jakarta ever since.
References:
http://marc.theaimsgroup.com/?l=ant-dev&m=97712718421034&w=2 http://marc.theaimsgroup.com/?t=97712745500023&r=1&w=2 http://jakarta.apache.org/site/pmc/01-01-17-meeting-minutes.html
- Sam Ruby
P.S. It is my understanding that what is now Apache HTTPD 2.0 is also the result of a number of forks, one of which ultimately emerged as being the one accepted by the community.
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
-- Ceki
TCP implementations will follow a general principle of robustness: be conservative in what you do, be liberal in what you accept from others. -- Jon Postel, RFC 793
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
