Bill Janssen wrote:
[...]
I'd like to add some referential markup to the mainly
descriptive/presentational system, though, and thought I'd solicit
comments from this list.
I think that you should
(1) use something like the SGML IDREF/ID mechanism, which is easy to type
and is more likely to be compatible with future software and with the
enriched text thingy :-)
(2) use SGML-style tags unless there is a clear reason not to do so -- there is
really no need to re-invent the wheel here
Alternatively, why not just add these facilities to the MIME enriched text
format? Either you have SimpleText or you don't; IntermediateGradeMail seems
to be stretching things a bit!
[ref: /tmp/foo]
And on non-unix systems?
[ref: vol2.[users.local.tmp].foo;2]
or
[ref C:\tmp\foo]
are filenames too...
and a message-part reference would look like
[ref: <a3fd590083.foo.parc.xerox.com>]
[...]
[ref: access-type=anon-ftp, name="/pub/foo/b.tar.Z",
site="parcftp.parc.xerox.com", mode=image]
I suggest instead
<ref type="file" id="/etc/passwd">An interesting binary file</ref>
and
<ref type="message" id="<a3fd590083.foo.parc.xerox.com>"></ref>
and
<ref access-type="anon-ftp" name="/pub/foo/boys.gif"
site="parcftp.parc.xerox.com" mode="binary" action="display">
A picture of James and his friends in the cathedral</ref>
Here you have an established method for representing the attribute-value pairs
such as access and name that you give in your examples, and also a placeholder
for a piece of text that might help the reader decide whether to follow the
reference or not.
Another thing I've been thinking of is to re-cast the footnote notion as
referential markup, e.g. [note: a footnote might look like this]. This
seems both clear in intent, easy to parse, and easy to understand upon
seeing the direct presentation.
Comments?
I'd prefer either:
<fn-text tag="1">a footnote might look like this</fn> Another thing
I've been thinking of is to re-cast the footnote notion as referential
markup, e.g. <fn tag="1">. This seems both clear in intent, easy to
parse, and easy to understand upon seeing the direct presentation.
or, better,
Another thing I've been thinking of is to re-cast the footnote notion as
referential markup, e.g. <fn>a footnote might look like this</fn>. This
seems both clear in intent, easy to parse, and easy to understand upon
seeing the direct presentation.
I am not suggesting that you use SGML, just that you use its syntax in a way
that minimises new invention.
But I am not sure that there is much point in xrefs and feetnote if you don't
have bold, italic, paragraphs, special squiggles (&pilcrow;) and so forth.
Perhaps there is, but I am not sure.
Lee
--
lee(_at_)sq(_dot_)com (Liam Quin) the barefoot programmer; SoftQuad Inc +1 416
239 4801
OPEN LOOK UI FAQ; Metafont list; HexSweeper NeWS game; lq-text text retrieval
`VAX... 3 Times better than your average vacuum cleaner'
(VAX Appliances Ltd. vacuum cleaner advertisement)