"Ronald Cannes" <roncannes@hotmail.com> writes:
> >But rather than debate this -- look at the locking-plan.txt. It's
> >modeled after the DAV lock specification, which allows both kinds of
> >locks.
>
> Since you discourage further discussion I assume there has already
> been made a decision on how this is going to be solved?
I was just discouraging people from arguing about whether locks should
be breakable or not, that's all. :-)
>
> That's very, very sad, since the locking-plan document doesn't address
> the problems with locking we've discussed. I urge you guys to rethink
> this issue before you start the implementation of such an important
> feature (at least to some of us.)
The lease-based model sounds like it may be very useful. Nobody is
closing the door on that idea. When the community sits down someday
to actually start implementing locking features (post-1.0), we'll
definitely re-discuss everything, including your idea. That
locking-plan.txt isn't the gospel; it's just a "capture" of previous
discussion.
But at the moment, we're all gunning very hard toward 1.0, and aren't
discussing or working on new features. That's the only reason you're
not getting a huge groundswell of reaction at the moment. It's not
that anyone is against lease-based locking, we're just not thinking
about locking at all right now. We're closing 1.0 bugs.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Wed Sep 3 22:11:26 2003