- From: Ian Hickson <ian@hixie.ch>
- Date: Tue, 3 Sep 2002 02:33:15 +0000 (GMT)
- To: Elliotte Rusty Harold <elharo@metalab.unc.edu>
- Cc: "www-style@w3.org" <www-style@w3.org>
On Mon, 2 Sep 2002, Elliotte Rusty Harold wrote: > > Even in HTML, when the UA sees a P element it is only guessing that > the user meant for that to be paragraph. Authors can and do use P > elements (and pretty much all the other elements in HTML) for very > different purposes. Authors writing invalid documents (e.g. using <p> elements for uses other than paragraghs) are out of scope of UA handling standard compliant documents. > Your approach seems to be to limit everyone to just a few dozen tags > so that browsers can guess more reliably. A few dozen? Have you _seen_ the currently well defined vocabularies? Between Dublin Core, XHTML, MathML and DocBook, to name but a few, we have _hundreds_ of well defined elements with precise semantics. > I understand your point, but this is completely contrary to the > spirit of XML, which is to let thousands of tags bloom Hahaha no. The point of XML is to allow well-defined languages to be created and then shared between UAs who are aware of the semantics of the languages used. e.g. SOAP, SVG, XUL, XBL, etc. All four of those languages, by the way, would fail spectacularily if a UA _guessed_ as to how they should be handled. > You seem to be arguing for dictatorial control of both tags and > their meanings, and that is indeed the HTML way. Not necessarily dictatorial. Merely agreed upon by all parties. Read http://www.w3.org/TR/xag.html I don't mean skim it. I mean read it. The whole thing. -- Ian Hickson )\._.,--....,'``. fL "meow" /, _.. \ _\ ;`._ ,. http://index.hixie.ch/ `._.-(,_..'--(,_..'`-.;.'
Received on Monday, 2 September 2002 22:33:16 UTC