I recently explored the FAQ -> WWW converter and server being run by Thomas
Fine at Ohio State University. I was impressed and want to share what I
learned.
To my surprise and delight, the server converted my FAQ smoothly even though I
did not conform to "official" digest format and had not followed any imposed
format. It turns out I did one thing right, and that was all that was
required: I began each section with a Subject: line preceded by a blank line.
The server took my Subject:s and built an index file from them as the top
level html file.
The OSU people discovered that very few FAQs were conforming rigorously to
digest format, so they had to adopt a practical approach. Thus the much more
lenient detection of sections initiated by Subject: lines. They actually have
a more sophisticated way of detecting sections, but if you need those
specifics you are probably already using digest format anyway. They also
recognize a special format from which they can derive additional html
information. The details are available in
http://www.cis.ohio-state.edu/hypertext/faq/usenet/technical-notes/top.html
or is it /faq.html?
Sorry, I'm not in a place where I can easily verify which is correct. You can
just look at the directory, or at the usenet directory. This only works for
WWW -- apparently ftp access to these directories is not authorized.
I've made two changes to my FAQ as a result. The first is to put Subject:
lines before the table of contents and before my signature so that they get
indexed and don't got lumped with other things (the TOC with the article
headers and my sig with the last section). The second is to include a pointer
to the OSU server in my "Where's the FAQ" section and also in a new section:
Subject: Online resources: World Wide Web
=========================================
If you are using the World Wide Web (aka WWW, W3, lynx, Mosaic), you can reach
a WWW-formatted version of the FAQ via the URL
http://www.cis.ohio-state.edu/hypertext/faq/usenet/diabetes/top.html
This is an automatic conversion of the plain text FAQ and does not contain
WWW links to other information.
As far as I know, the OSU server operators welcome publicity about, and
pointers to, their server. I've had several requests recently from people who
want to reformat my FAQ and make it part of another information resource.
Naturally I resist such requests, since there is no mechanism for keeping
such copies up to date. Thus I welcome the ability to tell them just to
implement a WWW link to the OSU server. This way neither they nor I need
worry about their copy going out of date.
The only problem I encountered was an old version of my FAQ cohabiting with
the current version, a result of my changing the archive-name a few months
ago. This is due to the server simply processing posted news. There is a
mention in the wish list in their technical notes that they would like to
mirror rtfm.mit.edu for greater accuracy, but haven't done it yet. For now,
presumably the only way to delete out-of-date materials is to make a manual
request by email or to do a one-time posting under the old archive-name of a
small file directing the reader to the new file.
Oh, and I think the server needs a name so that I don't have to keep
referring to it as "the OSU FAQ -> html WWW server".
My thanks to Thomas Fine and his helpers at OSU!
-- Edward Reid ed@titipu.resun.com (normal) PO Box 378 Edward_Reid@acm.org (forwarding) Greensboro FL 32330 reide@freenet.tlh.fl.us
[
Usenet Hypertext FAQ Archive |
Search Mail Archive |
Authors |
Usenet
]
[
1993 |
1994 |
1995 |
1996 |
1997
]
© Copyright The Landfield Group, 1997
All rights reserved