- From: Mark Nottingham <mnot@mnot.net>
- Date: Sun, 10 Jul 2005 18:49:10 -0700
- To: Karl Dubost <karl@w3.org>
- Cc: www-validator@w3.org
Hi Karl, I had a conversation with Tantek Celik the other day, and he corrected my misconceptions WRT microformats (at least as he preaches them ;). AFAICT, they only use class and id attributes, along with HTML profiles, to add semantics to documents, and thus don't break validity. I.e., they *don't* arbitrarily add attributes and elements. Apologies for the confusion. I still think it's worthwhile to consider HTML's extensibility story in light of the ongoing work on extensibility and versioning happening in the TAG and elsewhere, but it's not as urgent for the requirements of microformats. That's probably not directly of concern here. What the validator *could* do, however, is support validation of HTML profiles; e.g., by validating profiles described in XMDP; http://gmpg.org/xmdp/ Just a thought. Cheers, On 09/06/2005, at 8:54 AM, Karl Dubost wrote: > Mark, > > Le 05-06-06 à 17:36, Mark Nottingham a écrit : > >> Just curious if there's been any discussion about making the >> validator more microformat-friendly; i.e., not calling extension >> attributes/elements invalid, only calling out the first >> occurrence, etc. >> > > Would you like to give an example of such a document, just a small > piece of code and the version of HTML 4.01, you are using. To be > sure if you are talking about > new extensions/attributes > or > predetermined values of attributes. > > > > -- > Karl Dubost - http://www.w3.org/People/karl/ > W3C Conformance Manager > *** Be Strict To Be Cool *** > > > > -- Mark Nottingham http://www.mnot.net/
Received on Monday, 11 July 2005 01:52:24 UTC