>Seriously though, I guess I was also thinking of the suggestion in P4 that
>one create an URL attribute for pointers in order to handle, well, URLs. I
>thought it was an elegant solution.
It is not inelegant. It provides TEI users with a way of asserting a tight
binding to a nearly universal application semantic, which they expect. We
are still free to implement more robust solutions using various kinds of
indirection, if we want to maintain the overhead (unparsed entities,
anyone?). So nothing is lost but our innocence, even if it does violate the
strict principle I was trying to assert.
>The question of whether a hybrid might be an answer came up today as I was
>writing an article in TEI (because it handles notes and pretty much
>everything better) that needs to go out in HTML. When I came to a
>blockquote within a paragraph followed in the next by a figure and began
>to ask myself if I should just be writing the thing in xhtml.
>I have lots of quick an dirty documents I work on where I would like the
>benefits of both worlds. I realise, of course, that it is a simple matter
>to build a basic tei>xhtml stylesheet or modify an existing one, but I
>find I'm always having to do that and this time wanted something in
>between to check my work as I was composing. While Firefox was working
>fine for most of the time, a hybrid would be useful for the complicated bits.
I *really* sympathize with this: pondering where the best break is between
hacking/fudging/making-do, and engineering something stronger, is how I
spend my days.
FWIW, I can report (nor am I the only one) that investing the time in
developing your own authoring tag set (profile of TEI or not) is well worth
the effort, if you are not unhappy about spending your days with XSLT for a
while. These days, my own authoring tag set needs only very rare
interventions to extend it or to implement new features, and by now I have
transforms to convert it into HTML, DocBook, the Extreme conference DTD,
and TEI ... about the only major target I haven't gotten around to is
XSL-FO (think of that: and yes, I do have a stylesheet that draws an SVG
"plot" of the document). This is after several years of development, but
during no period has it demanded anything close to full time.
>I suspect the namespace or extensions root might be the way to go. If I
>called it dirty, I'd not be tempted to use it in real projects.
And in the process, you might discover something really clean and sparkly,
But we have come an enormously long way. Stories like James's are
inspiring. When students can look at the toolset, see *through* it at what
it's doing, see the XML piece, the XSLT piece, the CSS piece, make a tweak,
see it work, say "oh cool, it works" -- this is very very far away from the
old days of "some day, we'll have tools that do X and Y and Z".
Wendell Piez mailto:[log in to unmask]
Mulberry Technologies, Inc. http://www.mulberrytech.com
17 West Jefferson Street Direct Phone: 301/315-9635
Suite 207 Phone: 301/315-9631
Rockville, MD 20850 Fax: 301/315-8285
Mulberry Technologies: A Consultancy Specializing in SGML and XML