Fwd: [tei-council] DRAFT Agenda for the TEI Council teleconference on June 15, 2007 at 1200 UTC

Conal Tuohy conal.tuohy at vuw.ac.nz
Thu Jun 14 20:43:47 EDT 2007


On Thu, 2007-06-14 at 13:43 +0100, Sebastian Rahtz wrote:
> so we all agree that the TEI says nothing about output rendition? should we
> concretely suggest using html:class / html:style?

It's probably worth pointing out that in HTML, the class and style
attributes are not in a namespace, so we can't just "borrow" them in
this way.

> I find @rend and @rendref rather hard to swallow.  I know the
> TEI is famously indecisive, but this seems going too far.  I suppose
> if it was a choice of @rend *or* @rendRef it would be slightly nicer.

I tend to agree. I know if I had "@rendref" (as a pointer to a
tei:rendition), I would probably use it in place of rend everywhere. In
certain circumstances it might be considered easier for encoders to be
able to encode rendition information inline (using @rend), but I think
this would tend to be where there are a large number of distinct
renditions being encoded. For most uses of @rend which *I* have seen, it
would be easily replaced with a reference.

> What scares me is adding a new attribute to att.global, because
> it becomes so visible to everyone. What you _could_ have is
> @rendRef supplied by an optional module, but then what else
> would that contain?

That seems reasonable to me. Does it matter if the optional module
doesn't include anything else?

Would it be enough for the optional module to redefine @rend as a
pointer? Would that be a "clean" modification of tei_all?

Otherwise, if a new attribute name is required (distinct from @rend),
what about calling it "rendition" - to match the name of the element to
which it refers?

Con




More information about the tei-council mailing list