Re: Locking support

[Greg A. Woods]

[ On , August 28, 2001 at 12:13:59 (-0700), Brian Clark wrote: ]
> Subject: Re: Locking support
>
> PooPoo.  I've used CVS for years and have avoided that "insecure"
> feeling some of us get from not knowing if we are working on a file
> that someone else is working on.  We've wrapped cvs edit in a perl
> script that determines who else is editing the file and warns (ok
> informs, if that makes you feel better) you who is editing the file.

Talk about putting crutches on your crutches!

(I was going to say "training wheels on your crutches", but that's just
gives completely the wrong imagery!  :-)

> CVS has allot more to offer besides the "C" in its name:
> 
> -It's primary interface has remained command line.

That's a pretty meaningless feature -- at least in any Unix-like world
where pretty much anything of any value has, and always will have, a
"command-line" interface.

> -It supports directory structures (unlike rcs and sccs)
> -It is free and open sourced.
> -It has an extremely knowledgable user base.

You've not mentioned the one primary almost unique-to-CVS feature that
seems to drag all kinds of people over and to suck them into using CVS
even if it doesn't really meet their more important criteria (eg. even
if they handle lots of non-mergable files and/or they are extremely
opposed to concurrent editing for mergable files), and that's:

-It has client/server support.

> I don't know of another version control system that meets all of these
> criteria even close to as well.

I'd say you've not looked very far (especially if you ignore as a
requirement the client/server feature I added to your list).  I've
repeatedly mentioned several version control systems which VASTLY
improve on CVS for many purposes and which have all the attributes you
list, but which do not force concurrent editing in the way CVS does.


In any case *ALL* of what you said is pretty pointless -- CVS was
designed to *FORCE* you to get used to the idea of concurrent editing.
That was the primary design goal:  parallelisation of development!  The
other features are in part just gravy to make the concurrent editing
easier to swallow.  Building crutches around or into CVS so that you can
pretend you're eating your broccoli while feeding it to the dog under
the table is really kinda silly.

(The fact that CVS only really handles mergable text files with any
degree of long-term success wasn't really a design goal so much as an
inherent feature of the environment for which it was designed to serve.)

-- 
                                                        Greg A. Woods

+1 416 218-0098      VE3TCP      <address@hidden>     <address@hidden>
Planix, Inc. <address@hidden>;   Secrets of the Weird <address@hidden>