rapin@bnrmtv.UUCP (Eric Rapin) (09/28/87)
Hello,
I'm doing a little research to evaluate a
moderately large s/w program (~500,000 lines of code).
I've found some articles on doing quantitative analysis
of programs but the content was a little general and
didn't show much in the way of real application.
Articles that I did find that were "case-studies"
were too simplistic in their approach (although I
have to admit, at this point, that I have not been
very exhaustive in my search).
What I'm looking for are articles, texts, papers, etc.
on the topic of how do you evolve out-of-date s/w
into s/w which uses state-of-the-art techniques
without throwing away all of your existing features.
(If needed, I can be more specific about what I mean by
out-of-date.)
If this sounds vague, that's because it is and currently
from what I can tell in the literature, not much has
progressed in this area of Computer Science. Much has
been written but I don't think a lot of it is very
well understood and I'm trying to set up some kind of model
on which to base this evaluation.
(Good luck, right?)
Any help that anybody could provide would be appreciated.
Real experience in this area is especially desired.
I would imagine that this could turn in to a real
bandwidth eater on the net so mail responses are preferred.
If people are interested, I'll try to summarize (assuming
I understand what's going on myself).
Thanks in advance,
--
Eric B. Rapin UUCP: {3comvax,amdahl,ames,csi,hplabs}!bnrmtv!rapin
Bell-Northern Research, Inc.
Mountain View, California