Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!rutgers!tut.cis.ohio-state.edu!gem.mps.ohio-state.edu!ginosko!uunet!mcsun!unido!tub!uwbln!ckl
From: ckl@uwbln.UUCP (Christoph Kuenkel)
Newsgroups: comp.lang.c
Subject: Re: Healthy skepticism towards documentation (was: effect of free)
Message-ID: <1813@uwbull.uwbln.UUCP>
Date: 11 Sep 89 12:37:30 GMT
References: <10817@riks.csl.sony.co.jp> <10990@smoke.BRL.MIL> <14176@bloom-beacon.MIT.EDU>
Distribution: comp
Organization: UniWare GmbH, Berlin
Lines: 22

In article <14176@bloom-beacon.MIT.EDU>, scs@hstbme.mit.edu (Steve Summit) writes:
> Of course, the next question is, if you can't trust the
> documentation, what can you trust?  (It used to be the source
> code, but fewer people have access to it in these days of [...])
i do it the following way:

1. see my local man page.  feature not available (fna) -> forget it (fi)
2. see SVID. fna ? fi
3. see x/open guide. fna ? fi
4. see pure bsd4.2 manual. fna ? fi
5. see sunOS manual. fna ? fi
6. remember, what was in v6. fna ? do i really need it ? fi
7. now i try to find a common subset of 1 to 6 (by stripping the
   result type of sprintf() for example). empty ? fi

at this point, almost all of whats left can be found in the v6 manual pages
or can be implemented in an optional compatibility lib. and i'm porting
to *ix only!
-- 
# include <std/disclaimer.h>
Christoph Kuenkel/UniWare GmbH       Kantstr. 152, 1000 Berlin 12, West Germany
ck@tub.BITNET                ckl@uwbln             {unido,tmpmbx,tub}!uwbln!ckl