Path: utzoo!utgpu!watmath!looking!brad
From: brad@looking.UUCP (Brad Templeton)
Newsgroups: news.admin
Subject: Article Classification is the key to solving USENET problems.
Message-ID: <3233@looking.UUCP>
Date: 11 May 89 07:49:07 GMT
References: <3222@looking.UUCP>
Organization: Looking Glass Software, Waterloo Ont.
Lines: 46

Why am I so interested in the References: line?

I believe that the final key to solving most of the problems on USENET is
better article classification.  Article headers should contain as much
information about the poster as possible, in a machine understandable
form.

Once we get posting programs including more and more information, it's up
to the reader to configure his or her reading based on that info.  Not
everybody will do it, but the burden now falls on the reader.

Then, if your reader doesn't present USENET to you the way you want, then
that's your problem, and you shouldn't complain to the net about it.

Some say undergrads and novices should not post.  I say that sites should
simply put flags on articles indicating the poster's usenet experience
and let the reader decide.   Some say you shouldn't crosspost between
soc.women and talk.abortion -- I say let the newsreader filter that out.

To this end, I am having written, at my own expense, a more powerful
postnews/Pnews program.  This will include a simple language to control
posting.  Tests can be made, menus can be popped up and information can
be added to articles, both on a global or newsgroup specific basis.  I
will give this program away free.

In the end, the only thing people should have to complain about is whether
or not people classify their articles well.  Those who are novices can
learn, and those who deliberately lie can be put in kill files.

But we need the header information, and we need it to work.  The References:
line is something that was put in several years ago and it still doesn't
provide what it was intended to.  Once the information is there, the various
and sundry news reading and filtering tools (I'm working on some of these,
too) can be used to let each reader see the usenet they want to see -- to
the extent that they're willing to program or buy programs for.

In fact, the only problem this doesn't solve is the disk space and transmission
cost of wasteful postings.   But the biggest cost of USENET is actually
the human cost -- time spent reading postings you don't want to and
complaining about them.  If we can deal with that problem we are well
on the way.

If there is interest, I will list some of the ways I think articles can
be classified, and how they might work.
-- 
Brad Templeton, Looking Glass Software Ltd.  --  Waterloo, Ontario 519/884-7473