Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!crdgw1!crdos1!davidsen
From: davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr)
Newsgroups: news.software.b
Subject: Re: Time for 8 bit news, isn't it?????.
Message-ID: <2365@crdos1.crd.ge.COM>
Date: 24 Jul 90 17:46:31 GMT
References: <3119.269d97ea@mccall.com> <777@hades.ausonics.oz.au> <15688@bfmny0.BFM.COM> <==H&NB&@b-tech.uucp> <1990Jul22.195243.28379@zoo.toronto.edu>
Reply-To: davidsen@crdos1.crd.ge.com (bill davidsen)
Organization: GE Corp R&D Center, Schenectady NY
Lines: 29

In article <1990Jul22.195243.28379@zoo.toronto.edu> henry@zoo.toronto.edu (Henry Spencer) writes:

| Geoff and I thought hard about this during C News development.  The trouble
| with checksums is that most people would prefer a slightly mangled copy of
| an article to no copy of the article.  There are all too many transmission
| channels that do in fact slightly mangle articles (expanding tabs, fiddling
| with the definition of newline, etc.).

  This is a good point, but there are some groung rules which could help
eliminate this. When I do a CRC on text, I ignore all whitespace, and
put a delimiter (@@start and @@stop) around the text. This makes it work
even when fairly heavily munged in the usual ways.

  The problems of line ending, conversion of blanks to tabs and back,
adding or deleting blanks at end of line, can all be ignored this way.
Even if lines are folded you can get a CRC if you ignore whitespace.

  This is not to say you're wrong, just that a partial solution is
available. I have used this for some time, and it seems to be critical
enough to be useful, and forgiving enough to avoid dropping things which
are still readable.

  I use brik for error checking on c.b.i.p postings, for historical
reasons, and I haven't had a complaint in six months. A header field for
CRC would be great, even if all the reader did was output a message
indicating that the data was damaged.
-- 
bill davidsen	(davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen)
            "Stupidity, like virtue, is its own reward" -me