On Thu, 23 Jan 2003, Nicola Ken Barozzi wrote:
> > > Henri Yandell wrote: > > There are a few things here, in decreasing importance: > > > > 1) Code must be maintained to be worth using. > > 2) Code must have a community to be maintained. > > 3) The same code ought not to live in more than one place. > > 4) Reusable code ought to be in Jakarta Commons. > > Well said, apart from 4 which I would say "could" instead of "ought to". I thought you had buy in now that Commons was the place for low level reusable bits if possible? ;) > > Now, if the Ant developers are the only ones doing 1), and they are the > > only 2) for the code, then according to 3) the code should be in one > > place. This place ought to be Jakarta Commons, but if this is not possible > > then it should be in Ant as 4) is the least important of the 4 things. > > This is why it was put in Commons, but the magic didn't happen ;-) There is no magic/spoon [depending on movie/book of your choice]. > > So, +1 to the Ant guys managing the code inside Commons, +0 to the Ant > > guys offering the jars as a separate build. > > I think the main issue with this is that Ant is somewhat at the top of > the Gump dependency graph, and splitting things out is not something > that fast and easy to do correctly. Yep. If Ant people can be happy with it, I'm happy to build the three jars in a Commons project and publish it on the site. Otherwise, I'm just waiting for a nod to remove the source from the Commons repositories :) Hen -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
