W3C home > Mailing lists > Public > www-tag@w3.org > February 2006

Re: Principle of Least Power (was Re: Agenda of 7 February 2006 T AG teleconference

From: Elliotte Harold <elharo@metalab.unc.edu>
Date: Sun, 12 Feb 2006 06:28:52 -0500
Message-ID: <43EF1BF4.3090807@metalab.unc.edu>
To: Dan Connolly <connolly@w3.org>
CC: "Bullard, Claude L (Len)" <len.bullard@intergraph.com>, "'Henry S. Thompson'" <ht@inf.ed.ac.uk>, noah_mendelsohn@us.ibm.com, Vincent.Quint@inrialpes.fr, www-tag@w3.org

Dan Connolly wrote:

> I suspect it's still more expressive than DTDs, though I'm not certain.
> Surely the XPath string and numeric operations are more expressive
> than DTDs. Seems like an interesting research topic.

I would define "more expressive" in this context as meaning it was 
possible to write (in some language, not necessarily Schemtron) a 
DTD-to-schematron compiler. i.e. for every DTD there was a Schematron 
schema that validated the same set of documents. Is that possible?

No, it's not. It runs ashore the same rock that all other attempts to 
replace DTD so: no entity definitions. Schematron also won't do 
notations or attribute types. So in terms of expressivity, Schematron is 
neither clearly more nor clearly less. You can say some things in 
Schematron you can;t say in DTDs, but the reverse is also true.

Perhaps if you could define a more quantitative measure of expressivity, 
the statement that Schematron is more expressive than DTDs might still 
be true. You could say "Schematron can say 23 things and DTD can only 
say 17 things." but I'm not sure how you'd count that.

´╗┐Elliotte Rusty Harold  elharo@metalab.unc.edu
XML in a Nutshell 3rd Edition Just Published!
Received on Sunday, 12 February 2006 11:29:03 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:32:48 UTC