[Top][All Lists]

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

RE: List of files modified between two tags

From: Arthur Barrett
Subject: RE: List of files modified between two tags
Date: Thu, 6 Nov 2008 21:22:06 +1100


> But I would like to add that I am still not very clear 
> whether the decision
> to use CVSNT for production purpose was right or not. I have 
> also read in an
> article (by a guy working in Cruise Control) that SVN is a 
> better tool than
> CVSNT/CVS and that the code in either of them can be moved to 
> SVN. 

I've written many an article here about how to go about selecting a SCM

An organisation installs SCM to achieve a buesiness deliverable.  The
suitability of any tool should be measured against those deliverables.
Anyone who tells you 'use tool xyz' or 'tool xyz is better than tool
abc' is just a zealot and not interested in helping you or your
organisation.  The right tool for the job can only be determined until
the business drivers are determined.  

Secondly companies that research the effectiveness of SCM (like CMI)
find that once you know your business objective you then should find a
SCM process that supports that objective - and then find a tool that
supports the process.  Skipping the process and going from objective to
tool misses the fact that SCM is a process - not  a piece of software.

Finally - there is some ongoing research (look for my previous posts on
this issue) about the cost to business of retiring legacy software or
'upgrading'.  The evidence I've seen in the reports shows that it is
very high and very rarely does an organisation perform cost / benefit
analysis before embarking on it, and when an independent orgnanisation
comes in and does the cost/benefit after the fact they find that the
costs far outweigh the benefits in most cases.

So in summary - if it's worth switching from CVS to SVN there will be
good business drivers and you will have done a cost/benefit analysis to
show that your organisation or team will directly benefit form the
change.  In fact if it was true the internet would be full of these
reports.  But when I google I can't find any - all I can find are boring
lists of 'svn has feature xyz' and no indication of how that is supposed
to justify the cost of moving or how that is supposed to help my
business or anybody elses business.

March Hare Software who are the primary contributors to the CVSNT
project and who I work for are just in the process of releasing CM
Server (based on EVSCM open source project which was previously known as
CVSNT 3.1).  This is a CVSNT server that can accept SVN clients and
CVSNT clients.  This is designed to address these issues:
* where organisations have several SCM systems and can find cost/benefit
in consolidation
* where allowing a developer to use a SVN client will help team morale
or recruitment
* where the organisation requires failsafe audit, centrally managed
access control, centrally managed policy but do not want the cost of
retraining staff already familiar with CVS/CVSNT/SVN and/or GUI tools
like TortoiseCVS/WinCVS/TortoiseSVN
* where organisations want/require a repository stored in a corporate
SQL databaser
* where organisations want to make it easy for tech wirters to version
documents by using 'web folders'.

Release 1 of CM Server (CM Server 2008) will be avilable early next week
from here:

BTW: a quick list of things that CVSNT has that SVN does not:
* failsafe audit (generally seen as required for SOX compliance)
* user defined changesets (generally seen as required for effective SCM
so that changes to code can be tied together with changes to documents
or test sets or jobs or projects or bugs)
* reserved and unreserved support (ie: both CVS and VSS/PVCS models)
* reliable/stable - built on 20+ years of coding in CVS


Arthur Barrett

reply via email to

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