[rdfweb-dev] The Scutter Volume problem

Morten Frederiksen mof-rdf at mfd-consult.dk
Wed Mar 17 19:36:01 UTC 2004


On Wednesday 17 March 2004 19:02, Julian Bond wrote:
> The alternative is to store it all locally. If I don't smush it first,
> there's no gain over fetching the data right then apart from some speed
> and bandwidth. If I do smush it, I've got a big pile of tricky queries
> and the problem of the data being out of date.
I'm not sure why your queries would be tricky, do as you do when displaying a 
profile, list it all...
If you don't want to display it all, just display what you want, I don't see 
how that's more tricky than your current profile view.

The out-of-date problem, sure, it exists, but with a semi-intelligent scutter 
that shouldn't really be a problem. People don't go to Google with the 
expectation that it'll be completely up to date...

Also, you really won't leverage any of the advantages of RDF, the fact that 
you can merge, and to some extent "understand", unknown terms. Even if you 
don't understand what e.g. foaf:interest means, you can still deduce "a link" 
between to people that share a common interest. Think "related"...

It's not that I don't see a value in displaying only the contents of a 
specific file, it's just that I really don't find it that interesting 
(anymore).


Regards,
Morten



More information about the foaf-dev mailing list