httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Chuck Murcko <>
Subject Re: use of CHANGES files for research
Date Thu, 02 Mar 2000 22:21:10 GMT
Marc Slemko wrote:
> On Thu, 2 Mar 2000, Chuck Murcko at work wrote:
> > Someone at UPenn's Wharton Business School has approached the group
> > about using the CHANGES files from past versions of Apache in their
> > research. Since this is public data already, my inclination is to think
> > that there are no problems with them doing things like
> >
> > ...
> Sounds fine to me.  As always, I'm sure people would appreciate a pointer
> to a copy of anything they produce.
> They are completely responsible for understanding the dynamics of the
> CHANGES file though (or the CVS commit logs, if they use those instead),
> and the fact that you will have some people that don't show up much but do
> lots, some people that show up a lot but are showing up in the
> committing-other-people's-small-bugfixes mode, etc.  Not understanding
> this could lead to some very biased results without the proper
> qualifications.

Definitely. Statistics and conclusions drawn from them are meaningless
without an understanding of the raw data. They've been hosting open
source conferences at Penn for about a year now, I think. I know they
were at the open source show in NYC a few months back, so they are
beginning to get a clue.

My sense is to get both cvs logs and CHANGES to them, along with an
explanation of the meaning of both.
Chuck Murcko

View raw message