Revs for windows alone.

[Ben Hyde <bh...@pobox.com>]

It appears that the window's 1.3.0 has issues what make
a 1.3.1 desirable. Meanwhile the unix side seems pretty
healthy.  To save the unix side folks from a mess of work
(testing, updating, etc.)  I'd suggest we be willing to rev
just the windows side of things.

My intuition is that this isn't the last time this will
happen. 

So does anybody feel strongly that 1.3.1 that is window's
only would be a BAD thing?

 - ben hyde

Re: Revs for windows alone.

[Marc Slemko <ma...@znep.com>]

On Mon, 29 Jun 1998, Alexei Kosut wrote:

> On Mon, 29 Jun 1998, Dean Gaudet wrote:
> 
> > 
> > 
> > On Mon, 29 Jun 1998, Ben Hyde wrote:
> > 
> > > So does anybody feel strongly that 1.3.1 that is window's
> > > only would be a BAD thing?
> > 
> > I thought we agreed ages ago that we shouldn't delay unix just to wait for
> > windows.  It only seems fair to go the other way as well.  1.3.1 should be
> > released for unix as well though. 
> 
> Agreed. If there's a good reason, we shouldn't be hesitant to do
> Windows-only releases. Here, though, I think it's advisable to release
> both. I can think of three reasons:

Exactly.

Windows only releases are ok if necessary, although it is a lot less
likely for that to be likely because the code is normally in better shape
on Unix than Win32.  In this case, however, it is not necessary or
desirable.  Remember, we planned a 1.3.1 for Unix compliation problems all
along.

Re: Revs for windows alone.

[Alexei Kosut <ak...@leland.Stanford.EDU>]

On Mon, 29 Jun 1998, Dean Gaudet wrote:

> 
> 
> On Mon, 29 Jun 1998, Ben Hyde wrote:
> 
> > So does anybody feel strongly that 1.3.1 that is window's
> > only would be a BAD thing?
> 
> I thought we agreed ages ago that we shouldn't delay unix just to wait for
> windows.  It only seems fair to go the other way as well.  1.3.1 should be
> released for unix as well though. 

Agreed. If there's a good reason, we shouldn't be hesitant to do
Windows-only releases. Here, though, I think it's advisable to release
both. I can think of three reasons:

1. There are some non-Windows things in the code. Many, in fact. There are
51 entries in CHANGES for 1.3.1; only 6 are Win32-specific.

2. It's easy to make a Unix source release. If we don't, people will get
annoyed and email us. A lot of folks scour our web/ftp site every day,
just looking for new releases (these were the ones who found 1.3.0 a week
before we released it); no matter whether we tell them to use 1.3.1 or
not, they'll want it. I say we release it.

3. Even if it was Win32-only, a non-InstallShield, source-only release
would be nice, for people who don't want compile their own without IS
getting involved.

-- Alexei Kosut <ak...@stanford.edu> <http://www.stanford.edu/~akosut/>
   Stanford University, Class of 2001 * Apache <http://www.apache.org> *

Re: Revs for windows alone.

[Dean Gaudet <dg...@arctic.org>]

On Mon, 29 Jun 1998, Ben Hyde wrote:

> So does anybody feel strongly that 1.3.1 that is window's
> only would be a BAD thing?

I thought we agreed ages ago that we shouldn't delay unix just to wait for
windows.  It only seems fair to go the other way as well.  1.3.1 should be
released for unix as well though. 

Dean