I'm referring to the fact that, once an SGML DTD is chosen, there
are _hard limits_ on the lengths of attribute specifications,
literals, tags, and just about everything else (ATTSPLEN, LITLEN,
TAGLEN, etc.). This is mainframe thinking at its most annoying...
In the case of a scripting extension to the Web (such as Java),
what you really need is an arbitrary-length URI, in order to specify
both the location of the script and the initialization data that
should be passed to this invocation; e.g.
http://www/java/script.class#x=1;y=2;name=hello-world;...
Because of the artificial limits imposed by SGML, a URI like this
must either be
- represented (wrongly) as element content, or
- broken up (kludgily) into multiple tags for re-concatenation.
All proposals for Java so far have come down on the side of
splitting up the URI over multiple tags, as in
<APPLET><PARAM><PARAM><PARAM><PARAM> ... </APPLET>
> Smart SGML-based applications using VRML notations can
> use semantic linktypes derived from HyTime ilinks and
> notation locations (notlocs).
Great, but wouldn't <!ENTITY> declarations (which is what HyTime
uses to declare URIs, I believe) be limited by some SGML capacity as
well? If they are so constrained, we still haven't escaped the
basic problem.
> SGML systems and applications that process the SGML
> Declaration entity can adjust name lengths, character
> sets, tag levels, attribute capacity, literal length etc.
The mere existence of these limits is what I find troublesome...it
doesn't matter how "adjustable" they are.
--------------------------------------------------------------------
Paul Burchard <[email protected]>
``I'm still learning how to count backwards from infinity...''
--------------------------------------------------------------------