On 8/10/06, David Crossley <[EMAIL PROTECTED]> wrote:
Ralph Goers wrote:
> Now we are getting nit-picky. This page lists three categories;
> procedural, code modification and package releases. Quite frankly, I
> don't think this vote has much to do with any of these because:
> a) procedural to me is a process - such as switching from ant to maven.
> One could argue from a warped point of view that changing a dependency
> fits in this category.
> b) code modification - no code is actually being modified by changing a
> dependency. (At least yet).
This "b" is the closest category, i reckon.
I've been hesitating to jump in on this but I'd have to agree; clearly
a change to the JDK implies that changes to the code base that are
backwards incompatible can now be made. We could of course ignore
this and just let any change that is not backwards compatible be
vetoed, but what's the point in that?
<snip/>
I reckon that Joerg did the perfect thing. He felt strongly
enough to vote -1, and then provided an alternative.
It takes a lot of guts to vote -1 ... thanks.
If we think he is wrong then challenge his veto.
Otherwise go back to the drawing board and come up with
an alternative proposal to vote on.
Here's the real reason for my hesitation: I have to wonder, is the
ability to run on older platforms really a valid reason for a veto? I
realize this could be controversial, but personally, I'd have to say
no. If you need to run Cocoon on something that won't support Java 5
then use Cocoon 2.1, it still works. At some point the universe has
to move forward and you can't continue to have support for all older
platforms indefinitely. I'd say for the veto to be valid we'd also
need a _compelling_ reason why 2.2 (and not 2.1) needs to run on the
platforms that don't support Java 5.
--
Peter Hunsberger
