Curt Hibbs <curt / hibbs.com> writes:

> Christian Neukirchen wrote:
>> Nikolai Weibull <mailing-lists.ruby-core / rawuncut.elitemail.org> writes:
>> 
>>>My point is this:  Switching to Subversion is a poor choice, as it
>>>doesn't go far away enough from the CVS way of version control.  There
>>>are better alternatives on the way, and there's no need to rush to the
>>>other choice to CVS right now, before the market has stabilized a bit.
>>>Lately it seems that everyone is writing a version control system, and
>>>hopefully someone will manage to come up with a really good system that
>>>will use many brilliant ideas, like the web service you hint at, and
>>>great portability.
>> 
>>>Anyway, to sum it up, I think that now is a bad time to switch to
>>>another version control system, as there's a lot going on at the moment,
>> Full ACK.  Thank you for the first reasonable post in this thread.
>
> That assumes that you *want* to go "far away" from the CVS model of
> version control. My personal opinion is that it is a good model with a
> flawed (or incomplete) implementation. The aim of Subversion is to
> remedy those flaws, and goes along way towards doing so.

Yes, I want to "go far away" from the CVS model of version
control---in fact, I did for my own projects.  However, I do realize
that darcs still is suboptimal for projects like Ruby.  (I had my
local Ruby in a darcs repository.)

Therefore, I propose to stick to CVS until the real alternatives
(bazaar-ng looks rather promising to me, for example, despite being
written in Python.  It will run on win32, too, but still has a long
way to go.) get better---then will be the right time to afford the
possible troubles of a switch.

A switch from CVS to SVN (or even worse, non-free version control
systems like Perforce) is likely to be more problematic than it's
worth, IMHO.  Of course, YMMV and I'm not a committer, either.

> Curt
-- 
Christian Neukirchen  <chneukirchen / gmail.com>  http://chneukirchen.org