On Fri, Feb 16, 2001 at 08:54:45AM -0800, [EMAIL PROTECTED] wrote:
> On Fri, 16 Feb 2001, Juergen Vigna wrote:
> > On 15-Feb-2001 Michael Schmitt wrote:
> > > However, it is my impression that 1.1.6 has been released a bit too early
> >
> > Well the problem is that we wouldn't have found all the problems with
> > the tabulars so fast if we wouldn't have released it to public, only
> > few people try cvs-versions HARD :)
>
> I'm not criticizing all the wonderful work you all are doing. I really do
> like what I see in 1.1.6 and am looking forward to the new things in 1.2
> and beyond, and I'll still be here to help document them. But 1.1.6 (and
> fix1) should never have been released as a public, "stable" version,
> particularly since it was known that the tabular code wasn't complete.
Both of you have good points. Personally, I think 1.1.6 should have been
released without the tabular code. The math bug and others got fixed pretty
quickly, and things like that may happen with any release. But when you make
a major change in file format (so that changes aren't reversible) I think
you have to be careful about releasing. I mean, especially because there was
actually reduced functionality in the new tables!
I'm not criticizing JÜrgen's work -- I'm actually amazed he got the new
tables put in so quickly! But I don't think a large new piece of code should
go in if it's not finished. And you knew it wasn't finished. It's
unfortunate that we don't have enough developers for an unstable branch, but
we just don't. So the only solution is to make each release rock solid, and
1.1.6 just wasn't.
JMarc, maybe you could try and hurry fix2, with some of JÜrgen's table
fixes? Or haven't they been back-applied yet?
-Amir
