home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!psinntp!curly.appmag.com!pa
- From: pa@curly.appmag.com (Pierre Asselin)
- Newsgroups: comp.unix.aix
- Subject: reporting bugs (was: ptys and getpwnam())
- Message-ID: <1253@curly.appmag.com>
- Date: 12 Aug 92 18:42:38 GMT
- References: <1992Jul30.141015.27287@news.ysu.edu> <1992Aug9.185135.21404@awdprime.austin.ibm.com>
- Organization: Applied Magnetics, Goleta, CA
- Lines: 59
-
- In <1992Aug9.185135.21404@awdprime.austin.ibm.com>
- jfh@greenber.austin.ibm.com (John F Haugh II) writes:
-
- >In article <1251@curly.appmag.com> pa@curly.appmag.com (Pierre Asselin) writes:
- >>As things stand now, not fixing bugs because they haven't been
- >>"officially reported" is a useless charade.
-
- >No, it is NOT a useless charade[...]
-
- What! Is, too!
-
- >Trying to fix random bugs off of USENET is often pointless because the
- >location of the actual problem is hard to determine.
-
- Then toss the article in the trash can. I have no problem with USENET
- articles being treated as hearsay. What I want is for the *convincing*
- hearsay to get injected into the bug reporting system without anyone
- having to go through a charade (--isn't! --is, too!).
-
- >[...]because the system is designed to insure
- >that a reported problem EXISTS and has been FIXED. The "exists" part means
- >that there is a reproducible test case that can be used to demonstrate the
- >behavior. Just saying "such-and-such function call is broken" does nothing
- >to reveal the location of the problem. The "fixed" part means that there
- >is some way to determine that any bug which is found actually is the same
- >bug that the customer found and that the fix which the developer produces
- >will fix the problem that the customer has.
-
- I can pick up the phone right now, get a problem number, and tell
- Software Defects that such-and-such function call is broken. Now you
- have to work on it. You don't have a reproducible test case because
- there's no way I can give you one verbally over the phone. That's why
- I highly recommend email filing with aixserv. The aixserv status reports
- are extremely instructive and helpful, too.
-
- The hard ones are the suspected bugs that aren't ready from prime
- time. IBM wants us to call our SE's when we need that kind of help.
- If you think that IBM support in the field actually works, you'll have
- to visit my planet some day. The quality of field personnel varies a
- lot. THAT is the real problem. It is a BIG problem. Too big for me.
- Is there a band-aid that will work now?
-
- So I'm on my own. I read the net to see the punches coming and to
- trade individual help with other netters. Now then: every once
- in a while, I read an article about some bizarre glitch that I had seen
- before but didn't have time to chase. Now I know I'm not crazy. If I
- see a bunch of "me too" followups, I know there is a real problem.
- We all know. So why do I have to read yet another article that says
- "report it, guys and gals, or it will never get fixed"?
-
- Once again, my humble proposal for a band-aid:
-
- 3) IBM:
- Allow your employees to file bug reports at their discretion
- for problems reported on Usenet.
- --
-
- --Pierre Asselin, Magnetoresistive Head Engineering, Applied Magnetics.
- pa@appmag.com the usual disclaimers apply.
-