Re: My no-longer pseudo code, the way I understand it:-)

Yes, you are probably right. I must admit that I have never ever used
the <base> element before, and I totally forgot about its existence:-)
But the semantics in HTML clearly suggests that we should interpret it
in RDFa, too, as setting the base for the whole document. And we should
probably bury @xml:base for now:-(

Ivan

Niklas Lindström wrote:
> I agree, a pity. It is correct though, XHTML 1.1 has no reference to
> xml:base at all, but a very clear definition of "/html/head/base":
> <http://www.w3.org/TR/xhtml-modularization/abstract_modules.html#s_basemodule>.
> So while @xml:base is off limits for RDFa in XHTML 1.1, it seems
> reasonable to treat the value of base/@href in the same way (as if an
> @xml:base with that value was present in the root html element).
> 
> Does this seem correct?
> 
> (In XHTML 2 though, @xml:base is available everywhere, so resolving
> how to treat that is still relevant in the future.)
> 
> Best regards,
> Niklas
> 
> 
> On 8/11/07, Ivan Herman <ivan@w3.org> wrote:
>> Hm. Do you mean to say that XHTML does not understand xml:base at all? I
>> was wrong there if that is true...
>>
>> I am not sure what this means for us. Does it mean that xml:base is once
>> and for all banned from RDFa, ie, we should use the top URI as a base
>> and that is it?
>>
>> I see your point on the discrepancy of HTML vs. RDFa in that aspect, I
>> must admit, though it is a real pity...
>>
>> Ivan
>>
>> Ben Adida wrote:
>>> Ivan,
>>>
>>> Great stuff with the pyRDFa implementation. More implementations and
>>> more feedback on tests is just great.
>>>
>>> I have one problem: I'm not sure we agreed that xml:base would be taken
>>> into account here. I can't seem to find any resolution or spec that says
>>> that it should be taken into account.
>>>
>>> The main issue with using xml:base is that it obviously does not affect
>>> @href, since the HTML specification doesn't take xml:base into account.
>>> In other words, HEAD/BASE/@href in HTML *should* affect @href, @about,
>>> and other relative URIs, but I don't see the argument for supporting
>>> xml:base without causing all sorts of havoc.
>>>
>>> -Ben
>>>
>> --
>>
>> Ivan Herman, W3C Semantic Web Activity Lead
>> Home: http://www.w3.org/People/Ivan/
>> PGP Key: http://www.ivan-herman.net/pgpkey.html
>> FOAF: http://www.ivan-herman.net/foaf.rdf
>>
>>

-- 

Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf