[foaf-dev] Re: RDF triple assertions live forever?

[Phillip Rhodes]

Renato Golin wrote:

> It's the same with RDFAuth, you have to trust someone sometime, you need =

> a list of trusted sites, people, documents, beliefs. If your site says =

> "renato is bad" it may "like" better Site A and even automatically add =

> it to the "trusted sites" or even keep a score of things you agree with =

> the site as the "automatic trust level" as opposed to your "hardcoded =

> trust level" when you trust someone even if you don't agree with him/her.

Yes, exactly.  That was pretty much they way I had thought of it all
along, but the statement that was made about RDF triples living
forever had my head spinning for a minute.

Clearly the possibility for contradictions, inconsistencies, etc.
will exist, and I think those will absolutely need to be resolved
using some combination of trust scores, pre-defined trust relationships, =

maybe even (depending on the context of the question) Bayesian =

statistics / fuzzy logic or other ML / AI techniques.


Thanks,


-- =

Phillip Rhodes
Chief Architect - OpenQabal
https://openqabal.dev.java.net
LinkedIn: http://www.linkedin.com/in/philliprhodes
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mindcrime.vcf
Type: text/x-vcard
Size: 224 bytes
Desc: not available
Url : http://lists.usefulinc.com/pipermail/foaf-dev/attachments/20080328/08=
bfc03a/mindcrime.vcf