httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From r..@covalent.net
Subject Re: Review then commit.
Date Sun, 28 May 2000 00:50:25 GMT

I must be missing something.  It seems relatively obvious to me that at
some point we must go from alpha to beta.  What better time to go from
alpha to beta than when there are no more known bugs and no more open
showstopper issues or more issues documented as desired in the
release?  This is the case that we are rapidly approaching.  Am I the only
person who sees this as an obvious time to become a beta?

> > But the config stuff doesn't need to be in an alpha because it is
> > invisible to the user.
> 
> Once again, you seem to see some type of milestone event that
> I don't, that causes us to transition from alpha status to beta.
> And from some other comments in this thread, I don't think I'm
> the only one missing what you think causes this transition -- so
> I don't believe we can take it for granted.

When would you see this becoming a beta?  The reason given for not going
straight to beta in January was that not enough people had played with the
code and it wasn't stable enough.  I believe both of those issues are
disappearing now.

> > I really don't know what you mean by non-filesystem
> > resources.
> 
> I think he means being able to map URLs to things like objects in
> a database, or files in a tarball, or the like.

Oh, that makes sense.  Thanks.

> > Here's my position.  We have had three potential 2.0's now
> > (not alpha's, three independant 2.0 repositories), and we are
> > about to have a fourth alpha for this one.
> 
> Um, I'm not sure I understand what you mean.  Are you referring to
> the NSPR, initial 2.0 sandbox, and hybrid server CVS modules that
> have come and gone?

There was a pthreads port a while ago (can't remember who wrote it), NSPR,
the initial apr sandbox and now this one.  So, that's actually four
potential 2.0's, but I combined the last two because the apr sandbox
became this alpha.

> > This one is getting stable, and many of the people who have been
> > working hardest on this would like to see it released and
> > moved forward.
> 
> I think you're the only one I've noticed agitating for this (aside
> from Bill, who has IBM reasons).  Have I missed something?  Who
> are these 'many?'

I have had conversations off-line with at least five more AG members who
would like to see this move forward.  I agitate for this, because the more
somebody agitates for stuff like this, the faster it moves forward.  I
don't really expect to see a beta anytime soon, but yes, if it were up to
me (I know it isn't, I said IF) I would like a beta at the end of this
month.  There is a lot of very useful stuff in this codebase already.  Why
do we feel that EVERYTHING must go into 2.0?  Why can't some of it wait
for 2.1 or 2.2 or maybe even 3.0?

I am not pushing for this for business reasons, I want that made very
clear.  I am pusing for this because I have been working on this code base
with others for over a year, and I think it is almost ready to be
unleashed on the public.

> > I don't want to take another year to get 2.0 released.  I would
> > like to head towards beta relatively soon, because I think this
> > is almost ready for it.
> 
> Well, since it's unclear to me what you think differentiates alpha
> status from beta, I can't comment on your last sentence.  However,
> I am strongly opposed to either rushing or pinching off development
> just because of some artificial and arbitrary release-date pressure.
> When it's ready, and the group has reached consensus to that
> effect, is when it should be released.
> 
> Perhaps if I understood your alpha/beta vision I would agree.
> However, since the notes of a couple of days ago sounded like
> you want to stop innovation in 2.0 and just finish the current
> task list, I'd like a real clarification..

I think stability separates alpha's from beta's.  I think the way we tend
to operate, new features go into the tree throughout our releases, so the
only thing that really can differentiate is stability.  This is the
general feeling I got when I first suggested putting out an alpha/beta in
January.

Ryan

_______________________________________________________________________________
Ryan Bloom                        	rbb@apache.org
406 29th St.
San Francisco, CA 94131
-------------------------------------------------------------------------------



Mime
View raw message