Re: SD5 - Namespaces (Implementation questions)

In message <1.5.4.32.19970519084336.006cc350@mail.u-net.com> Martin Bryan writes:
> At 10:03 18/5/97 GMT, Peter Murray-Rust wrote:
[...]
> >Example
> ><![CDATA[
> >
> ><!DOCTYPE cml SYSTEM "cml.dtd" [ <!-- unqualified, just as an example -->
> ><!-- an additional version the math dtd *using* FQNames; i.e. long-winded -->
> ><!entity % mathml SYSTEM "http://www.w3.org/some/where/fqmathml.dtd">
                            ^^^^^^^^^^^^^^^  A ^^^^^^^^^^^^^^^^^^^^^^^
> >%mathml;
> >]>
> ><CML>
> ><MOL Title="benzene">
> >...
> ><XLIST TITLE="symmetry operators">
> ><?XML-DOCTYPE DOCTYPE="org.w3.mathml"?>
                          ^^^^^ B ^^^^^^
[...]

> Nice idea - one question though. Should it be required that the doctype be
> declared as part of the relevant document type definition? At present there
> seems to be no link between the DTD declared as %mathml in the internal
> subset and that used in the XML-DOCTYPE PI. I can see a good case for a

You are right, of course.  I used the first because I assumed that a 
SYSTEM had to use a 'filename', but I much prefer the entity approach.

> formal definition of the relevant doctype as an entity and then using
> XML-DOCTYPE to point to this, giving it a statement along the lines
> <?XML-DOCTYLE DTD="mathml">. This has one ancilliary benefit. Instead of
> qualifying each embedded DTD's element name with a long URL it needs only be
> qualified by a short (parameter) entity name. This would also match subdoc
> (and concur) much more neatly as here it is the DTD name that is used to
  ^^^^^^^^^^^^

> "qualify" the element name.

Anything that is 8879-style would be preferred  - my reading doesn't 
extend to these - even after your very generous present :-) <A HREF="#fn1">

<PROPOSAL>
We should identify a character (Tim has suggested ':') for indicating
hierarchy in qualified names for GIs.  Even if we can't resolve whether we
should use it *now*, it may help in internal representations and temporary
files.  '.' had the advantage of being DNS and Java-friendly, but maybe
it's a good idea to choose a different one for that reason.
</PROPOSAL>

	P.

<FOOTNOTE ID="fn1">SGML An Author's Guide</FOOTNOTE>

-- 
Peter Murray-Rust, domestic net connection
Virtual School of Molecular Sciences
http://www.vsms.nottingham.ac.uk/

Received on Monday, 19 May 1997 17:53:26 UTC