home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!think.com!barmar
- From: barmar@think.com (Barry Margolin)
- Newsgroups: comp.security.misc
- Subject: Re: Stupid Licenses (YUCK!)
- Date: 23 Dec 1992 00:26:52 GMT
- Organization: Thinking Machines Corporation, Cambridge MA, USA
- Lines: 24
- Message-ID: <1h8bocINN5tl@early-bird.think.com>
- References: <1992Dec22.151201.22766@kakwa.ucs.ualberta.ca>
- NNTP-Posting-Host: telecaster.think.com
-
- In article <1992Dec22.151201.22766@kakwa.ucs.ualberta.ca> DMARTIN@vm.ucs.UAlberta.CA writes:
- >Maybe I'm misunderstanding what you mean by a "bug level", but I'm having
- >difficulty with what you propose. WHAT IS an "acceptable bug level", defined
- >so that I know when I have reached it? You can't simply ask that there be
- >no more than x "minor bugs" present in the software - that gets you into
- >a variation of the issue of proving software correct (How can I possibly KNOW
- >that I am down to x "minor bugs" and 0 "major bugs"?).
-
- As several others have posted, the reasonable goal is not "less than N
- bugs" but "less than N *known* bugs". We all realize that software that's
- powerful enough to be useful is too complex to be perfect. But if a vendor
- knows about a serious bug when they ship the software then they're clearly
- giving higher priority to deadlines than to quality.
-
- The question arises: how do you prove that the vendor knew about the bug?
- And even if the vendor admits to knowing about it, they might not agree
- that it's a serious bug. If it only affects .1% of users, is it serious?
- Maybe not to the vendor, but if you're in the .1% it probably seems so.
-
- --
- Barry Margolin
- System Manager, Thinking Machines Corp.
-
- barmar@think.com {uunet,harvard}!think!barmar
-