home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.software-eng
- Path: sparky!uunet!usc!zaphod.mps.ohio-state.edu!news.acns.nwu.edu!nucsrl!travis
- From: travis@eecs.nwu.edu (Travis Marlatte)
- Subject: Re: Software Inspections. How many does it take?
- Message-ID: <1992Dec21.190532.25352@eecs.nwu.edu>
- Summary: Having authors present at inspection
- Organization: Rauland-Borg Corporation, Skokie IL
- References: <BzB9J2.2q2E@austin.ibm.com> <1992Dec17.232054.8942@eecs.nwu.edu> <1992Dec18.190221.16549@menlo.cadre.com>
- Date: Mon, 21 Dec 1992 19:05:32 GMT
- Lines: 41
-
-
- The only reason not to have the author present at an inspection is
- due to ego. This is a poor reason. Inspection teams must get past
- that point to be productive whether the author is present or not.
-
- Yes, the code should stand on it own. That's why the inspection is being
- held. But having an inspection does not imply that the code is ready to
- go. Changes will be made to code and in-line documentation as a result
- of the inspection.
-
- Having the authors present simply prevents the reviewers from mentally
- wandering off the track. It is easier for the author to add
- information to the discussion, make a note that the information
- should be added to the documentation and allow the inspection to continue.
-
- My perspective on reviews and inspections is that it is a team effort
- to improve the quality of the software. The author is a part of that
- team and must be present.
-
- The post questioning the value of "scribes" and "moderators" is missing
- the point. Eventually, the author has responsibility for incorporating
- any changes that should be made as a result of the inspection. Having
- a scribe or recorder seperate from the author simply allows the author
- to concentrate on the discussion and points of confusion. The notes
- taken will end up with the author anyway. It is not just a way of
- creating metrics.
-
- Professionals or not, any group of people involved in a discussion with
- time limits needs to have someone in control. Clearly, professionals
- should be able to calmly come to a consensus. But people involved
- in a discussion often fail to recognize the passage of time. It is better
- to have a moderator who can say, "Well, obviously this is a point of
- confusion. Made Bill and Frank can get together with Jane to iron out
- the details later. Lets move on with page 1573."
-
-
-
- --
- Travis Marlatte
- travis@eecs.nwu.edu
- 708-297-0055
-