- From: William F Hammond <hammond@csc.albany.edu>
- Date: Tue, 09 Jun 2009 17:04:56 -0400
- To: "Robert Miner" <robertm@dessci.com>
- Cc: "Cameron McCormack" <cam@mcc.id.au>, <www-math@w3.org>
"Robert Miner" <robertm@dessci.com> writes:
> . . . Another post recently pointed out that this is such a
> common question/problem that it really deserves to be addressed
> directly in the spec, and I think the consensus is that is right, so
> we will try to add that in the next draft (which we hope will be the
> Last Call draft).
Thanks!
There are some subtle cases. For that reason it would be good if
there were ways with markup to handle distinctions rather than
simply relying on cdata. Our experience with the U-2032 series
(where we have not only the stix glyphs but also, consistent
with that, the rendering in xterms with X11 fonts matching
"-misc-fixed-medium-r-normal--*-*-iso10646-1") shows the peril
of depending only on cdata.
I do think that
(1) <ci>x'</ci>,
(2) <mi>x'</mi> (*),
(3) <mrow><mi>x</mi><mo form="postfix">'</mo></mrow>, and
(4) <msup><mi>x</mi><mo>'</mo></msup>
should, by default, have the same browser rendering,
and modifications from that should be available with
appropriate (but yet to be specified) settings of
attribute values.
(*) To avoid dependence on cdata it would be reasonable to require an
attribute value setting in case (2) for escaping the rule that a
multi-character indicator should be set upright.
-- Bill
Received on Tuesday, 9 June 2009 21:05:40 UTC