[whatwg] foreign attributes Re: several messages about XML syntax and HTML5

[Elias Torres]

Ian Hickson wrote:
> On Tue, 5 Dec 2006, Karl Dubost wrote:
>> Le 5 déc. 2006 à 05:34, Ian Hickson a écrit :
>>> The other issue, supporting other vocabularies in HTML5, is an open 
>>> issue, but it will be addressed in due course. We need more 
>>> implementation experience first, and there are far more pressing 
>>> problems.
>> slightly related. What about foreign attributes. I haven't found in the 
>> document the handling rules for unknown attributes. 
>> http://www.whatwg.org/specs/web-apps/current-work/
> 
> Bogus attributes end up in the DOM but have no other requirements on 
> processing, so, they are ignored. The spec covers this.
> 
> 
>> A group with a need of specific attribute might want to add new 
>> attributes to Web Apps 1.0.
>> - What are the extensions mechanims for this?
> 
> "class", "rel", and "profile" are the extension mechanism for HTML. For 
> examples of how this is used to great effect, see the Microformats.org 
> work, where entire vocabularies like vCard, iCalendar, etc, are embedded 
> into HTML documents using "class" and "rel". (This has the added advantage 
> of being a highly accessible extension mechanism, unlike most.)
> 
> Cheers,

Hi Ian,

I work at IBM/Lotus and are in the process of shipping a series of
internal components used by IBM employees as Lotus products. We have
tried using microformats as an extension mechanism for HTML but found
several limitations that make us look onto other things for
extensibility. We are hoping HTML5 will address some of the needs we
find to be extremely important in order for us to include structured
information in HTML.

One of the first problems we have with microformats is that even with
small structures like vCards, maintaining uf parsing code is not as
straightforward as writing the implementing the simple cases. Many have
tried to create extensibility mechanism for customers to submit new
microformats, but have ended up with solutions that require JS code to
be submitted in order to properly parse any number of customer-defined
microformats. Microformat work very well on clearly specified cases, but
do breakdown (in my opinion) when everyone wants their own microformat.

I'm currently participating in the RDFa task force but IBM would rather
not have to wait for XHTML2 to have a valid/official solution to this
problem.

-Elias