Re: [xsl] Re: XSL to ODF/OOXML

Subject: Re: [xsl] Re: XSL to ODF/OOXML
From: David Carlisle <davidc@xxxxxxxxx>
Date: Fri, 26 Oct 2007 11:49:32 +0100
> Are there any browsers that display <br /> as if it were <br> / ?

No. But there are many validators (eg the w3c one or any other code
based on nsgml that is a real SGML parser that does follow the spec)
that will declare <link/> as invalid for exactly this reason. Validators
don't usually complain about <br/> because in any context where <br> is
allowed character data is also allowed so the spurious extra > is valid
(even if undesired). But <link/> (and <meta/>) generate character data
in teh head so invalidate the document (if parsed with an HTML DTD).

> We do a lot of HTML > XML and XML > HTML conversion and this helps us
> enormously so not conformant it may in the most precise terms but for
> getting the job done it is a real time saver.

It is also of course explictly what the XHTML Recommendation recommends
in its non normative "compatibility" appendix, but the xhtml spec
doesn't actually redefine html, so this advice is somehat strange, it's
not surprising that people in general should recommend "something that
works" but it is rather more surprising that the W3C should advise
people to use a document convention that is incompatible with one of
(and arguably their most important) recommendation, the HTNL 4 spec.

But you have to remember that (current job notwithstanding) I am at
heart a pure mathematician and so may give slightly different relative
importance to the principles of "works in all real cases" and "not allowed
by the formal grammar of the language" than some people with a
background more grounded in the real word....


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. 

Current Thread