Newsgroups: news.software.b
Path: utzoo!henry
From: henry@zoo.toronto.edu (Henry Spencer)
Subject: Re: Time for 8 bit news, isn't it?????.
Message-ID: <1990Jul22.194732.28080@zoo.toronto.edu>
Organization: U of Toronto Zoology
References: <1990Jul13.022224.25441@lth.se> <3119.269d97ea@mccall.com> <777@hades.ausonics.oz.au> <15688@bfmny0.BFM.COM> <1990Jul21.054016.10409@looking.on.ca> <1990Jul21.231352.6234@zoo.toronto.edu> <1990Jul22.062034.20896@looking.on.ca>
Date: Sun, 22 Jul 90 19:47:32 GMT

In article <1990Jul22.062034.20896@looking.on.ca> brad@looking.on.ca (Brad Templeton) writes:
>C news may indeed *handle* arbitrary byte stream messages, but does it
>*support* them?
>In particular, input programs like inews must deal with them, and a new
>header line must be added to classify article body types...

C news "supports" arbitrary byte-stream messages to exactly the same
extent as it "supports", say, poetry:  it doesn't give a damn what's
inside the message or what's in any headers other than the ones it
needs to know about.  With the exception of the inews problem, you're
purely and simply talking about reader issues, not transport issues,
and no changes to C News are necessary.

The inews business is a nuisance, but the blame rests primarily with
the Unix utilities rather than with inews proper.  There is an inews
rewrite already in the works, which might improve things somewhat.

It might also be worthwhile to define a new input interface without
all the goo and dribble that have crept into inews over the years.
Being backward compatible was a real pain there.  A clean interface
could eliminate a lot of messy handling of stuff that inews should not
have to care about.  (For example, the specs say that inews must try
to guess whether the beginning of the input text looks like headers,
in which case it *is* headers.  This could be eliminated by demanding
that the -h flag be used in such cases.)

As for content classification, I believe there has already been some
work done on this for RFC822-X.400 interfacing, although I don't know the
details.
-- 
NFS:  all the nice semantics of MSDOS, | Henry Spencer at U of Toronto Zoology
and its performance and security too.  |  henry@zoo.toronto.edu   utzoo!henry