Re: longdesc URLs and RDFa

I too agree that it is important to have requirements, use cases, and 
examples when proposing the addition of any feature.

However, in this case what we have are some well defined features of the 
languages we support (HTML4 and XHTML 1.1) that we are not currently 
using to reap triples.  So any requirements and use cases should really 
be about why triples from these features are relevant to the semantic 
web (who would consume them, what would the consumers learn if they 
did), not about justifying the features themselves.

This isn't the forum in which we could attempt to justify the features 
anyway - that's the HTML5 working group.  Our job is only to define 
processing rules for extracting triples from the languages that we 
support, and rules for extending that processing into additional host 
languages.

On 8/16/2010 12:01 PM, Leif Halvard Silli wrote:
> Martin McEvoy, Mon, 16 Aug 2010 17:50:27 +0100:
>    
>>   Hello Lief,
>>
>> On 15/08/2010 12:17, Leif Halvard Silli wrote:
>>
>>      
>>> .... The micro format I presented initially - roughly this:
>>> 	<a href="http://example.com/l.htm" rel="longdesc">	
>>> 		<img src="img" alt="text" />
>>>       </a>
>>>
>>> leads to the triple:
>>> http://example.com/doc<longdesc>   <http://example.com/l.htm>   ....
>>>        
>> You know you would find it a lot easier to define a longdesc rel
>> patter if you first proposed it as an actual microformat[1], you
>> would find it a lot easier then to convince people that longdesc is
>> desirable because you would have examples, a use case and a format to
>> persuade people with.
>>      
> That's a good idea. I may do that.
>
>    
>> @longdesc isnt much thought about by the web as a whole (hence why
>> the longdesc attribute was recently removed in HTML5[2])
>>      
> It was long ago removed.  Recently, it was _not added_. I know all
> about that:
>
> http://www.malform.no/messages/appeal-issue-30.html
>
>    
>> so its
>> difficult for people to get a grip of why the value may be useful, Im
>> not saying @longdesc isn't but use cases and examples *should* come
>> first before you can begin a technical  discussion about format, It
>> gives people a solid Idea of what you are proposing and how it may be
>> implemented.
>>      
> Again - a good tip. It could happen.
>
>    
>> Best wishes ;)
>>
>> [1]
>>
>>      
> http://microformats.org/wiki/process#So_you_wanna_develop_a_new_microformat.3F
>    
>> [2]
>>
>>      
> http://lists.w3.org/Archives/Public/public-html/2010Aug/att-0112/issue-30-decision.html
>    

-- 
Shane P. McCarron                          Phone: +1 763 786-8160 x120
Managing Director                            Fax: +1 763 786-8180
ApTest Minnesota                            Inet: shane@aptest.com

Received on Monday, 16 August 2010 17:09:59 UTC