I think we need to release the reorganized/repom'd specs... that was
the general plan after cleaning them up was to make a release so we
have a set of consistent non-SNAPSHOT specs to work off of.
After this re-release, we can debate one version, or whatever. I'd
still like to see one version win out, but I have yet to come up with
the smoking gun argument to convince everyone that I'm not on crack.
BUT... for now, we gotta move forward with something. Its easy
enough to change later if we want.
--jason
On Nov 7, 2006, at 9:21 AM, Kevan Miller wrote:
On Nov 7, 2006, at 11:28 AM, Dain Sundstrom wrote:
Are you kidding me? You want to open this up for discussion again
just as we were going to finally make some progress and hopefully
put this to bed. The reorganization we just did was to have one
version number per spec.
Well, as you have probably guessed, I'm not. You aren't putting
this to "bed". You're ignoring it. I'd like to put it to bed.
We agreed to reorg specs. We did not decide whether or not to
release specs in independent or interlocked version numbers. Isn't
that what we discussed on [EMAIL PROTECTED]
As you can see, I am totally against changing what we have right
now, one version per spec.
Fine. And I'm against releasing a bunch of specs, for which we've
gotten the version numbers wrong for twice in this discussion
thread. If we can't keep them straight, now, seems like we're
destined for confusion in the future.
So, you and I cancel each other out. Why don't we let the community
decide and move on? I'm happy to abide by our decision...
P.S. I'd assumed that we'd be passing TCK before releasing the
specs (at least that's how we've operated in the past). There's
some risk (probably pretty small) that we'll uncover a problem with
specs with our testing. Are you planning on releasing prior to tck
passing or afterward?
--kevan
