[uf-discuss] VIA or VIA SELF to indicate authoritative hCard[was:UID URL to indicate (relatively) more authoritativehCard(Was:Vote on this: rel="me self" to indicate anauthoritative hCard)]

[Ryan King]

On Feb 11, 2007, at 6:16 PM, Ryan Cannon wrote:
> On Feb 10, 2007, at 2:56 PM, Joe Andrieu wrote:
>
>> Forcing the religion of "uids
>> should be urls" on the rest of the world is not why we are here.   
>> Making it easy for authors to connect their web content and web
>> apps with the semantic web is.  If someone else likes uids that  
>> aren't urls, and the spec supports that, then why should we keep
>> them from establishing authoritative hCards?
>>
>> ...
>>
>> How is via+via self more constraining?  You can do everything you  
>> can with uid+url, but you don't have to use URLs for your UIDs.
>
> +1
>
> UID+URL *is* more constraining. Like rel-tag, you're forcing a lot  
> of assumptions about the documents *surrounding* these URLs--links  
> have to point somewhere after all.
>
> My UID, if you will, is http://ryancannon.com/. I've established it  
> across many different sites as the definitive link for "me". By  
> forcing UID+URL to be used to establish an hCard's source, you're  
> also forcing my most robust hCard must exist at that URL.

No, we're not. Once again, we're just talking about related-ness, not  
authority.

Also, I don't see why authoritativeness has to also mean  
completeness. Wouldn't authority be applied on a field-by-field basis?

> However, when I redesign my home page, if I want to move my contact  
> information to ryancannon.com/contact my UID shouldn't change--it's  
> still just the domain. However, with URL+UID, I'm screwed. With  
> @rel=via, I can still have the option to point to my full hCard.

No, you're not screwed. All you have to do is put a minimal hcard on  
your home page with UID+URL=ryancannon.com/contact. That's it.

> Also, Ryan, you have yet to address the fact that URL+UID changes  
> the parsing rules of the hCard--not that I believe it is currently  
> ideal. Is it responsible for the community to change the rules of a  
> deployed specification willy-nilly? While the implementation in X2V  
> may be trivial, it may not be in other applications in the wild-- 
> it's also not a good precedence to set for uFs in general.
>
> For backwards-compatibility alone, @rel=via seems to me an optimal  
> solution.

Yes, this would mean a change in how UIDs are parsed, however the  
only UID I can find in the wild are URLs. A change wouldn't break  
anything.

-ryan
--
Ryan King
ryan at technorati.com