Re: When are RDF statements asserted?

* Richard Cyganiak <richard@cyganiak.de> [2006-03-28 19:19+0200]
> 
> Hi all,
> 
> Just checking: Is it correct that publishing an RDF file on the Web  
> does not assert the statements therein?
> 
> Is it correct that out-of-band information (e.g. a web page stating  
> "All these files are up-to-date", or some nonstandard extension of  
> RDF) is necessary before an agent can safely act upon any RDF statement?

You're confusing at least 2 things, I think.

If I publish an RDF document that says, more or less, 'Saddam
has weapons of mass destrction [in Jan 2003]', we can talk about what info you 
need before you can be sure some human (me) is (or was...) making 
the claim. For example, it might be PGP signed, claim itself to have
me as its dc:creator / foaf:maker, be dated today, and the PGP stuff
checks out ok, and the PGP keyservers don't report that my key has been
retracted / compromised.

All that is a world apart from whether an RDF can "safely act" on 
any RDF statement. Firstly, you can act on reported claims, regardless 
who made 'em. You can stash them in databases and directories, index 
them, follow links from them. Classic Web stuff. Secondly, just cos 
we know who asserted them, we don't necessarily believe them to be true. 
And of course, even if we believe their claims to be true, that has 
little to do with safety of acting on them. It's an unsafe world....

> (Hypothetical example: a FOAF agent that fills my address book with  
> contact data for the people I foaf:know.)

Good example. You might have come across some geneological data 
for folk who are long-since uncontactable. I think 'safety' in your
sense is likely to be scenario- and application- specific.

cheers,

Dan

> 
> Cheers,
> Richard

Received on Tuesday, 28 March 2006 21:34:36 UTC