[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: r21207 (Renaming --no-svndiff1 to --pre-1.4-compatible)

From: Max Bowsher <maxb1_at_ukf.net>
Date: 2006-08-26 18:19:32 CEST

Philip Martin wrote:
> Max Bowsher writes:
>
>> Let rc5 be tossed *if* the conversation concludes with a consensus in
>> that direction. Either way, we'll have had the benefit of additional
>> soak time for the fixes merged since rc4. If the final consensus is that
>> we can release 1.4.0 rc5 as final, then we will have saved time.
>
> Or we could have rolled rc5 without your change, continued the name
> change discussion, and reached a consensus that the original name was
> better and then released rc5 as final. With the current rc5, if we
> decide the original name is better we would need an rc6 and you will
> have delayed the release.

Had rc5 gone out without my change, then unless someone had come up with
an unforeseen argument to convince me otherwise, I would have argued
strenuously for a rc6, to the point of calling for a vote if necessary.

Therefore, I felt that I could only make the time to release shorter by
merging the change so that it got into rc5, since the possibility of
objections from someone else was better than the certainty of objections
from myself.

> As it happens I think the name change was pointless, we cannot expect
> users to understand what this option does just from its name; reading
> the documentation, or the code, will be necessary.

Terse though it is, I feel that the option name and help that we have
now give users a better chance at understanding it without referring to
the documentation.

> What I dislike is
> the way you pushed this through just before a release, it was
> unnecessary to do it like that because if the change was important
> enough then there would have been a consensus to delay the release in
> order to make the change.

How could I reasonably have *not* merged the change when I did? It was
validly +1-ed, and is not suitable for pushing to 1.4.1, and we were
rolling a candidate for a final RC.

If I had deliberately refrained from merging, I would have either been
deliberately assisting in the production of an RC of lower quality that
it was within my power to make it, or I would have been implying that I
felt the change did not deserve the +1 I had given it.

Max.

Received on Sat Aug 26 18:20:11 2006

This is an archived mail posted to the Subversion Dev mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.