The essential problems of moving from CVS

[Curt Sampson <cjs%netbsd.org@localhost>]

On 2010-01-12 09:33 +0100 (Tue), S.P.Zeidler wrote:

> Now could we please skip this non-issue and consider the actual -problems-?

Well, they differ quite greatly depending on to where we want to move.
My thought on the matter is that the key decision is one of these three:

    1. We wish to move, in the next year or two, to a distributed system
    such as Git, Mercurial, or whatever.

    2. We wish to move, in the next year or two, to Subversion.

    3. We wish to stay with CVS for at least the next two years, and
    perhaps do a bit of work to make the repo available to others in a
    non-CVS format.

So, have I missed anything here? For example, did I miss any other
centralized systems that we should be considering? (I don't think that
there are any other than Subversion that are both open source and offer
significant improvements over CVS.) Is there another major group of
VCSes I've missed?

The questions behind this list seem to be, "do we want to change our
development style to gain the advantages that moving to a DVCS would
enable," and "if not, does CVS cause us enough pain that Subversion
would mitigate to make the switch worthwhile."

Incidently, FreeBSD's answer in 2007/2008 appears to have been 2.

cjs
-- 
Curt Sampson         <cjs%cynic.net@localhost>         +81 90 7737 2974
             http://www.starling-software.com
The power of accurate observation is commonly called cynicism
by those who have not got it.    --George Bernard Shaw