ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jose Alberto Fernandez" <>
Subject RE: [ANN] Collecting requirements for Ant2
Date Mon, 12 Mar 2001 12:00:40 GMT
> From: Stefan Bodewig []
> Jose Alberto Fernandez <> wrote:
> > The list should remove itself from the bussiness of judging and
> > debating whether a task is worth having or not. There should be a
> > repository where anyone and everyone can submit their "work of art"
> > tasks.
> This is already in the wish list - and one of my biggest wishes. I
> think, we all generally agree that a separate module for contributions
> is needed, but we will need rules for this (every committer would have
> commit access to all tasks in this module - but when/how may he/she
> use it, what are the rules for releasing a version of a task ...).

I do not think we need CVS support for this. The more burocratic (to use
some word) we go the worst it becomes overtime. We just need a repository of
jars witha short description. That is all we need. Maybe we can store it in
one of those big "contribute java code" sites that are all around. Then the
only thing we need is a link from the Jakarta pages.

If people want to post good, crappy, comercial, free, gnu, or whathever kind
of code it should be up to the users to decide whether to pick it up or not.
If people want to publish the sources, or encript the code it should be up
to them.

In other words, make ANT a platform (just like Apache is) and lets people's
creative juices flow!!!

> > The only debate on the list should be whether a contribution is good
> > enough and powerful enough as to make it into the official core jars
> Not sure, whether the notion of "good enough" is a good measure for
> core jars - sounds a little bit (how do you say in the US?,
> "politically challenging") discriminating. Something like "will this
> be useful for the average build environment or is it a special case
> tool" - which is vague enough - seems more appropriate.

Well I was not setting the rules here. What I meant was that we need to have
some criteria which may be based in some "usability", "where do we want ANT
to go", "broad usage", and of course "contributed sources" criteria. I think
we may  enforce voting and real discussion for things to enter here. Since
the developer may always put it in the contributed code. I am not trying to
impose a criteria at this point just panning the space of possibilities.

> > or about changes to the core architecture.
> Agreed - but I think ant-dev should serve as the discussion list for
> the ant-contrib module as well. I wouldn't want to have a separate
> user list for the contrib stuff either (well, we probably wouldn't
> need a user list for the core anyway).

I do not about discussion, may be promotion, understanding, criticism. I
just think the best way to manage such a beast, is to manage it as little as
possible. Just say what is in there and have recommendations or complaints.
If you want your code moved into the official optional tasks, then you will
need to port it to the correct packages and such.

Jose Alberto

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message