Re: RDFa and Web Directions North 2009

On 18/2/09 13:40, Mark Birbeck wrote:
> Hi Dan,
>
>>> There are two aspects to this debate. Some people want RDFa to be
>>> added to HTML5, and that browsers do something with it. I'd love to
>>> see that too, but then I'd also like to solve world hunger, and teach
>>> the world to sing.
>>>
>>> The second aspect of this debate is that, provided the HTML5 spec
>>> doesn't do anything that breaks backwards-compatibility with current
>>> browsers, then we can *already* do clever stuff with RDFa in HTML5. We
>>> don't need anything extra in the specification -- thanks for asking.
>> I think you're missing a case here:
>>
>> Many of us are working in environments where (at least looking to the
>> future) documents are expected to be (in some sense) valid. And for better
>> or worse, HTML5 looks like it will be the (or at least "a") dominant
>> browser-supported document format, which many people and organizations will
>> want to use (eg. for the new APIs and fun extras). Those of us who hope to
>> have such documents carry RDFa will be out of luck, if we're in
>> organizations who have some commitment to validating their HTML. Sure, those
>> orgs are in a minority, ... but they are a minority with a lot of
>> interesting and important data.
>>
>> So, I think you miss an important constituency here.
>>
>> Those who:
>>
>>   * don't need browsers to do anything with the RDFa other than tolerate it
>> and expose it in low-level APIs
>>   * expect HTML5 to be widely used
>>   * expect to have RDFa embedded in HTML5
>>   * expect those HTML5 + RDFa documents to validate
>
> Not at all...I'm very much aware of that constituency. In fact,
> despite the enormous amount of work that it has involved, I have
> always pursued my ideas through the W3C, rather than outside of it,
> for that very reason.
>
> But when the W3C itself doesn't seem to know where it stands on the
> question of standards, it leaves a lot of people in an uneasy
> situation. So my point is that *first and foremost*, provided that the
> HTML5 spec doesn't mess up the use of getAttribute() to obtain
> attributes that have a colon in their name, then we can parse RDFa in
> HTML5 browsers using JavaScript, just as we do today.
>
> Beyond that, everything else is a bonus. However, I think it would be
> a mistake to let this first point pass, in the hope that it is
> resolved as part of the bigger discussion -- that's too risky.

OK, I think we're in broad agreement then. Though I won't sleep easy 
until there is an agreed validator service...

cheers,

Dan