Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!mips!wrdis01!gatech!bloom-beacon!eru!hagbard!sunic!liuida!isy!lysator.liu.se!bellman
From: bellman@lysator.liu.se (Thomas Bellman)
Newsgroups: comp.unix.wizards
Subject: Re: Shared libraries (was Re: Window system bashing (was Re: X11 bashing))
Message-ID: <583@lysator.liu.se>
Date: 18 Apr 91 23:14:24 GMT
References: <26550@adm.brl.mil> <97@titccy.cc.titech.ac.jp> <1991Apr15.173955.7909@athena.mit.edu> <1991Apr15.224311.3950@ida.liu.se> <6783@stpstn.UUCP>
Sender: news@isy.liu.se (Lord of the News)
Organization: Lysator Computer Club, Linkoping University, Sweden
Lines: 24

In article <6783@stpstn.UUCP> Ken Lerman (lerman@stpstn.UUCP) writes:

> So I spend weeks trying to track down some bug in one of your
> libraries.  Then, since I don't have the source code, I come up with
> some ghastly workaround so I can ship my product.

> Two weeks after I ship the release to two thousand customers, you
> release a new version of your library.

> And my code stops working.

> A bug fix IS a change in functionality.

Well, don't you know that you should not exploit undocumented
features? :-)  Actually, if you do a workaround for the library bug,
then, IMHO, that is a bug in *your* code if the workaround doesn't
work with the bug fixed.  What happens if I release a new version of
the OS with a bug in a syscall fixed?


--
Thomas Bellman,  Lysator Computer Club   !  "Make Love - Nicht Wahr"
          Linkoping University, Sweden   !  "Too much of a good thing is
e-mail:         Bellman@Lysator.LiU.Se   !   WONDERFUL."     -- Mae West