home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.software-eng
- Path: sparky!uunet!spool.mu.edu!umn.edu!The-Star.honeywell.com!saifr00.cfsat.honeywell.com!shanks
- From: shanks@saifr00.cfsat.honeywell.com (Mark Shanks)
- Subject: Re: Software Inspections. How many does it take?
- Message-ID: <1992Dec18.145952.26062@saifr00.cfsat.honeywell.com>
- Organization: Honeywell Air Transport Systems Division
- References: <1992Dec14.192008.15480@gallant.apple.com> <BzB9J2.2q2E@austin.ibm.com> <BzEqpI.5yp@NeoSoft.com>
- Date: Fri, 18 Dec 92 14:59:52 GMT
- Lines: 46
-
- In article <BzEqpI.5yp@NeoSoft.com> claird@NeoSoft.com (Cameron Laird) writes:
- .
- >I have no experience with reviews in which the
- >author(s) is (are) *not* present, but I like
- >the idea; is there anyone who can report on how
- >that works in practice?
- >
- >My attraction to this has nothing to do with
- >"ego"; I just want to have the documents stand
- >on their own, without the explanations that the
- >author(s) inevitably carri(es).
-
- I suppose this begs the question, but what is intended to
- be accomplished at the review? If the software is expected to
- be maintained by someone other than the author (and THAT'S a
- good possibility), it makes some sense to me that the software
- ought to be able to "stand on its own". But what are all
- these other characters doing? The "moderator" (read "facilitator"
- in more fashionable circles) is an excrescence; if you can't
- get a group of professionals to conduct an inspection without
- this person, you don't have a group of professionals. Who is
- the "scribe" (a term that because of my upbringing conjures
- up images of cowled monks illuminating gospels) and what
- EXACTLY are they doing, collating defects? Flame me if you
- like, but I think that's what SQE should be doing. I see a
- HELL of a lot of hours spent on inspections, which is not
- bad in itself, but when you multiply how many people are
- involved, there must be a point of diminishing returns.
- IMHO, the duties of moderator and scribe can and should be
- assumed by SQE. As another has posted, the more designated
- inspectors involved, the less each individual is likely to
- feel responsible for actually inspecting, and the review
- turns into a nit-pick ritual.
-
- Let me start some new threads: WHEN should software be inspected?
- Or: How many times should software be inspected prior to
- initial delivery? Should software be inspected and put into
- a configuration management system months before design is
- complete and months before scheduled delivery, i.e., at a
- Critical Design Review stage? What are the differences between
- a formal and informal inspection?
-
- Just asking ;-)
-
- Mark Shanks
- shanks@saifr00.cfsat.honeywell.com
-