- From: David Carlisle <davidc@nag.co.uk>
- Date: Mon, 27 Nov 2006 15:25:14 GMT
- To: juanrgonzaleza@canonicalscience.com
- Cc: www-math@w3.org
> I am not interested in technical details or if the validators are strictly > following the XML spec or are not. Maybe you are not, but anyone hosting a validation service clearly should be concerned about that, and it was the behaviour of validation services that you were questioning. > Whereas the MathML validator A will say it "The input is valid MathML" and > the B will say it "This Page Is Valid MathML 2.0!". > Then the guy got confused and read the MathML specification and there > explicitely it is said that their code is _prohibited_. This isn't really any different from using an <a> element nested inside another in XHTML. Any DTD validation service will tell you it's valid but the prose text of the XHTML will tell you that it is prohibited. http://www.w3.org/TR/xhtml1/#prohibitions It's always going to be possible to define more constrains in the prose text of a specifcation than in any formal grammar, and it's not at all unusual for language specifications to do that. David
Received on Monday, 27 November 2006 15:25:30 UTC