On Feb 23, 2006, at 7:33 AM, Mark Galbreath wrote:
> I've read all the literature on the svn website, and the online
> book at red-bean, and thought I understood the differences between
> CVS (a locking reporsitory) and SVN (a merging repository). Then I
> got into a discussion with a guy at the office I just met and he
> tells me I have it all backwards. Then tells me that developers
> can walk on each other's code with CVS, but not SVN because SVN
> locks write access. This is 180 degrees off from my
> understanding. He maintained that SVN is way better than CVS, but
> I got the impression he didn't know what he was talking about, and
> I am still too green to stand my ground in such a debate.
>
> Thoughts?
>
> mark
I'm somewhat new, but here's my understanding and it may help you
figure out where this guy is coming from.
Both CVS and SVN are primarily merging repositories (the "good" type,
locking repositories are the "bad" type if you want to make the
distinction). SVN can be set up to implement a locking system if this
is desired (it may be desired for things like binary files where
merging really doesn't make much sense). So he is sort of correct
(SVN can implement a locking system), but that isn't the preferred
way to run SVN and that isn't why SVN is better than CVS. As far as
walking all over other people's code, that is a feature of every
merging repository and the locking repository's alternative (having
other people lock you out of code you need to update) is worse.
Likely this guy just doesn't like to do the messy work involved with
a merge. I can't blame him, but it's better than the alternatives.
Aaron
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Thu Feb 23 17:25:30 2006