• From: David Carlisle <davidc@n...>
• To: "Henry S. Thompson" <ht@i...>
• Date: Wed, 04 Jan 2012 10:55:52 +0000

On 04/01/2012 10:25, Henry S. Thompson wrote:
> Unless you mean "without DOCTYPEs", isn't this exactly what the
> current HTML 5 spec [1] does?  That is, it effectively provides a
> catalog which conformant user agents are required to use which maps a
> range of PUBLIC identifiers onto a data: URI which contains all the
> HTML and MATHML entity declarations.
>
> ht

It does do that, but bizarrely the list of PUBLIC identifiers for which
this data: URI is used all resolve (if used with any normal catalog) to
legacy DTD which define _different_ definitions for the entities, and
declare a different set of elements.

Thus it isn't at all easy to have a file using entities that doesn't
suffer data corruption or become non well formed if used with both a
normal XML toolchain and a browser XML parser. It is not possible at all
without abusing these existing FPI to have a file using entity
references that is dtd-valid and uses html5 features such as <canvas> .

This is the subject of:

https://www.w3.org/Bugs/Public/show_bug.cgi?id=13409

David

-- 
google plus: https:/profiles.google.com/d.p.carlisle

________________________________________________________________________
The Numerical Algorithms Group Ltd is a company registered in England
and Wales with company number 1249803. The registered office is:
Wilkinson House, Jordan Hill Road, Oxford OX2 8DR, United Kingdom.

This e-mail has been scanned for all viruses by Star. The service is
powered by MessageLabs. 
________________________________________________________________________

• References:
  • Re: what's missing in XML? What's coming?
    • From: Rick Jelliffe <rjelliffe@a...>
  • Re: what's missing in XML? What's coming?
    • From: Henri Sivonen <hsivonen@i...>
  • Re: what's missing in XML? What's coming?
    • From: ht@i... (Henry S. Thompson)