* Kate Ebneter <[EMAIL PROTECTED]> [000216 15:35] wrote:
>
>
> Alfred Perlstein wrote:
> > Forking a _working_ version of CVS would just be too resource
> > intensive for most companies.
>
> Really? We forked a working version of gcc. We haven't forked CVS
> because we haven't needed to, but we would if we did.
Why should it be nessesary to fork a version of CVS just to get real
fixes in?
> > > > Basically, if this is the official CVS list, then a real developer
> > > > presense is nessesary, maybe not to answer questions, but to make
> > > > sure that people have the right kind of idea about your project.
> > >
> > > There's nothing "official" about CVS. It's open source, get it?
> > > Moreover, it's not like Perl, where Larry Wall has the final say about
> > > what happens to it; only the end users, ultimately, have the final say.
> >
> > Apparantly you're completely confused about this issue, numerous patches
> > have been brought forward to address things that are broken in CVS
> > which don't seem to get a second look once Greg gets a chance to point
> > out that it fixes something he's opposed to actually working.
>
> I'm not confused; on the contrary, some of these patches are misguided.
> Some might be useful. Some are useful to some people, some are useful
> for everyone. Your definition of what's "broken" in CVS is obviously not
> the same as mine. Or Greg's.
My definition of broken is 'not working'. I'm unsure of your interpretation
of the word but feel free to explain.
> > There's at least five people a week asking for locking.
>
> ...and an equal number of people saying that the locking they're asking
> for is INCOMPATIBLE with the design goals of CVS!
Yes, but it's the SAME FIVE objecting everytime it comes up.
> > > Minor rant: I'm tired of hearing, on this list, about how CVS is broken
> > > because it doesn't support locks, or because it doesn't support symlinks
> > > properly, or... SYMLINKS??? Symlinks have no place in a serious build
> > > environment unless they're created on the fly -- they're not portable,
> > > for one thing, and, well, I could go on, but I won't. And as for
> > > locking, I'm sorry, but what part of CONCURRENT Versioning System don't
> > > people understand??? If y'all wanna turn it into Control Via
> > > Synchronization, well, could you please do it to another source code
> > > control system? Please? There are TONS of very good source code
> > > control/versioning systems using synchronization via locking. There are
> > > very few that permit or properly support CONCURRENT development, let
> > > alone make it the center of the development model. I _NEED_ CVS to be a
> > > CONCURRENT versioning system. I don't need locks, I don't need symlinks,
> > > and I certainly don't need 100+ emails a day telling me that Greg Woods
> > > is a jerk...
> >
> > blah blah blah, look, you don't really get it do you?
> >
> > If symlinks and locking is so terrible then for christ's sake
> > RIP IT OUT OF THE CODE!!!
> >
> > I personally don't give a hoot what you use CVS for...
>
> That's very clear. It's also very clear that you don't understand why
> symlinks in a version control system are a really bad idea, and that you
> don't understand the model of development that CVS is intended to support.
>
> <rant snipped for space>
pfft, I thought it was somewhat amusing.
Look, I understand how 'cvs is supposed to work', I'm not bringing
up these arguments to get locking in CVS nor symlink support, I'm
bringing them up because I'd hate to see the same thing that happened
in my 'rant' to happen to anyone else.
> > Now that's to the best of my recollection, it may not have been
> > exactly airplane glue he proposed, but I don't have my email archived
> > from that far back.
> >
> > At this point I really don't care all that much, I unsubscribed
> > sometime after sending that intial email, CVS is going to continue
> > to be broken and I'm tired of watching this depressing show.
>
> No. CVS is _not_ broken. You're trying to use a hammer to drive screws,
> and you're complaining when the maintainers of the hammer point out that
> instead of changind the hammer into a screwdriver, perhaps you should
> use a screwdriver. And, oh, by the way, that tack you're trying to
> drive, too, should probably not be there at all (the symlinks). Look, I
> don't know the specific problem that you're trying to deal with in
> detail. If I did, perhaps I could help you find a source code control
> system that was suited to your needs.
>
I still think you don't get it, I have no need for locking nor
symlinks.
All I'm asking for the is that the code be axed entirely... or
heaven forbid patches accepted to actually fix the troublesome
crap.
Allowing this garbage to stagnate for over a year really boggles
the mind. In fact it'd be really easy to get rid of it, all the
code is wrapped with '#ifdef PRESERVE_PERMISSIONS_SUPPORT' it'd be
trivial to take it out.
It doesn't work, it never worked, it's not been working for over
a year and patches aren't be accepted to fix it... get rid of it.
-Alfred
