lets do a release! (was: Re: [PATCH] Third and Final (hopefully))

[Havoc Pennington]

On Mon, 2005-11-21 at 16:17 -0500, John (J5) Palmieri wrote:
> 
> Why don't we get this stuff in and do a release.  It is obvious we
> aren't going to get feedback here without there being a release.  I can
> always do a 0.60 release if we don't want to commit to a freeze just
> yet.  Letting this linger is just stalling the process.  We can debate
> to worlds end on the theoretical but until it shows up in distros most
> people don't really care much and the ones that do (us ;-) are too busy
> elsewhere.  It is a pain if we change again but everything is going to
> be a pain at this point until we get to 1.0.

I think you're right that 0.60 with all pending changes makes sense.
Maybe people have to go through two binary-incompat releases, but they
will build a bridge and get over it eventually. :-P Modular X will
distract them!

> This all sounds good.  Really I think the argument was that if we felt
> that queuing might not be right it might be safer to just make it act
> like it wasn't a queue but if apps found the queuing useful they could
> turn it on.  If we feel that queuing is right then I have no problem
> leaving the patch as is with the DO_NOT_QUEUE flag.

I'm not convinced I understood fully where Robert was coming from (thus
some of the questions I was asking in my mail) but I would propose
leaving it as-is until it's either further argued out or (better) we get
some actual experience.

> > That probably makes sense, but we do need to be sure to get *all* the
> > planned ABI breaks ... a few more small ones might trickle in but we
> > should try to avoid any blatant oversights.
> 
> So can I check in all my patches? We can then look things over to see if
> we need to do more ABI breaks and only do a release after we are sure
> there are no more big ABI changes.

Yeah why don't we take the plunge.

Havoc