Re: tweet2rdf vocabulary convergence

On 28.09.2009 13:39:32, Alexandre Passant wrote:
>For instance, for mentioned people, it should be enough to use  
>sioc:topic or other related property and mention that the linked  
>resource is a foaf:person, e.g.
>
>:mytweet sioc:topic :alex .
>:alex a foaf:Person .
I'd like to keep things simple and close to the source format where
possible. Other tools could still infer sioc:topic from 
tweet:mentionedUser (os somesuch), but I wouldn't enforce this
in the initial source. SPARQLing real-time data will get heavy 
enough even with flat properties, I guess...

>Author-avatar profile can also be addressed with the sioc:User  
>properties (sioc:avatar)
yep. I also like the statusnet approach of attaching the postIcon
to the tweet/dent and not the user. Less JOINs and faster 
processing.

>We also discussed the possibility of adding the recipient list in SIOC  
>(e..g for e-mails - can be also used in microblog) several time, might  
>be time to finally add that sioc:recipient in the spec.
That would be cool. Would the range be the account URL (i.e. sioc:User
or foaf:OnlineAccount)? Twitter doesn't have person URIs, just account
URIs, for example.


Thanks for the suggestions!
Benji


>
>--
>Dr. Alexandre Passant
>Digital Enterprise Research Institute
>National University of Ireland, Galway
>:me owl:sameAs <http://apassant.net/alex> .
>

Received on Monday, 28 September 2009 13:35:42 UTC