[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: cvswrappers - any better suggestions ?

From: Paul Sander
Subject: RE: cvswrappers - any better suggestions ?
Date: Mon, 2 Apr 2001 01:02:26 -0700

Unfortunately, there are enough members of the CVS community who just don't
seem to understand the necessity of many proposed features, and who are
influential enough to defer or even actively discourage their introduction.
Examples of such needed features include full support of non-ASCII file types,
directory versioning, more complete and robust refinement triggers, and even a
generalized merge tool registrar.  These things have been discussed for years
in this forum, and numerous implementations have been discussed.  Nevertheless,
the preference to apply 1988's technology to 2001's programming problems
remains compelling for some.

Unfortunately, there does not seem to be sufficient interest to either
accept such new features into the mainstream, or to splinter off a new
development effort to produce a more capable tool.  RCVS was one attempt to
open up and encourage development of new things, but it didn't work out as
was hoped.

So, you have the following choices:  Design and build CVS v2.0 from the
ground up and make it a useful tool.  Cripple your software development
effort enough that the existing CVS can handle it.  Use something else.

Using something else is the fastest and easiest way to get what you want
(and the cheapest in the long run), but crippling your current process may
be the fastest and easiest way to overcome some immediate hurdle and limp
along until the next crisis arises.

--- Forwarded mail from address@hidden

With all due respect, Greg, I think Gianni made some very telling points which 
CVS will need to address if it is to survive.  I'm an independent consultant, 
and I try to bring CVS into each organization I work with.  I'm sometimes 
successful, but I fail at times for exactly the reasons Gianni articulates.

With the proliferation of MSWord, IDE project files, etc, no reasonable person 
can argue that non-text files are not a necessary part of most projects these 
days.  If CVS does not 'grow up' and attempt to support the new development 
environments, it will slowly be replaced by something that does, and CVS will 
ultimately become a bit player, promoted by a few fanatic users who try to make 
its case at every opportunity (whether appropriate or not), and ignored or 
laughed at by the majority of the population.

I've heard your same arguments applied to Forth, OS/2, Geos, Clarion, Lotus 
1-2-3, Wordstar, UCSD P-System... the list goes on.  I'd hate to add CVS to the 

--- End of forwarded message from address@hidden

reply via email to

[Prev in Thread] Current Thread [Next in Thread]