On Sun, 19 Oct 2008, Martijn Kruithof <[EMAIL PROTECTED]> wrote:

> I know i have not been really active for some time now

don't worry, we all have times where we can contribute more and times
when we can't do as much as we wanted to.

> The choice should not only be based on what new language features
> are offered by a java version, but also the number of java version
> we have to test against.

True.

> Maybe we should consider dropping java versions even somewhat
> faster, I don't see the point in supporting java versions that SUN
> does not support anymore for the general public.

I'm afraid that would lead us to support more branches of Ant than we
currently do.  For example I've been fixing a few bugs only in trunk
lately because I somehow expect us to release 1.8.0 next and not have
any 1.7.x releases anymore (I may be wrong, that's my gut feeling).
Given that some of the bugs are serious, silly or embarassing and the
number of people stuck with Java 1.4 is still big I'd feel pressed to
merge them into the 1.7 branch and push for a 1.7.2 release (which may
be just the right thing to do).

> We in turn we could consider to prolong the life of the point
> releases for critical updates to have at lease one ant version
> operating with the java version upto the state EOSL from SUN.
> 
> Bugs eligible to be solved on the point releases after a new major
> release has been done would have to come from customers running a
> java version that is not supported by a more recent ant release,
> running a java version that is still supported by SUN, these should
> also be real bugs, not feature requests, and the company shall have
> to test the ant release towards the old java version and provide us
> the logs, especially if we do not have the possibility to run the
> old java version anymore.

This sounds good, in theory.  Do you expect us to get enough votes to
do an Ant release that no committer has tested him/herself, though?
Wouldn't that imply we'd still be doing 1.5 point releases for people
using Java 1.1?

I like your approach, I'm just not sure we'll be able to live up to
the implicit promise of supporting older release branches after we've
made our next major release.

> (This because I am quite sure that in 10 years java 1.4 won't be
> running anymore on the "current" hardware and operating systems.)

Agreed, again.

Stefan

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to