> Would /> be better than ?>?
> I keep reading
> as meaning that that the value is in doubt!

<?XML blahblah!> was considered.  that would certainly have removed
doubt.  :-)

> But
> would match the EMPTY tag close delimiter, which would add some
> symmetry without hurting anything, I think.  It would also help understanind,
> in that <?XML....> is self-contained, just like <e>.
> I suppose one could argue that <!-- xxxxx --> should have a similar property,
> <!-- xxxxx --/>.

The general feeling of the ERB when we talked about it was
overwhelmingly "if we were doing it from scratch, we'd set up the
closing delimiters to make symmetries all over the place, e.g.,
pairing up <! and !>, and <? and ?>, and so on.  but since we're
trying to remain as SGML-like as we can, let's just keep to a minimum
the amount of necessary fiddling with the reference concrete syntax".


.---o  Tom Magliery, Research Programmer                         .---o
`-O-.  NCSA, 605 E. Springfield                  (217) 333-3198  `-O-.
o---'  Champaign, IL 61820          O-        mag@ncsa.uiuc.edu  o---'