[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: Triplets on the Internet

• To: "Kal Ahmed" <kal@t...>,"Bullard, Claude L \(Len\)" <len.bullard@i...>
• Subject: RE: Triplets on the Internet
• From: "Joshua Allen" <joshuaa@m...>
• Date: Mon, 14 Jun 2004 15:00:34 -0700
• Cc: "Rich Salz" <rsalz@d...>,"Didier PH Martin" <martind@n...>,"Jonathan Borden" <jonathan@o...>,"XML Developers List" <xml-dev@l...>
• Thread-index: AcRSU9yJ5T/89cUCQPauhhnEbJ9prwAADtlQ
• Thread-topic: Triplets on the Internet

> Absolutely true. But if that centralizer is simply a node in the P2P
> network, what happens when it propagates that inference to other
nodes.

> to more complex models (I ended up with the possibility of having
> multiple levels of reification : A says 'B says "C says 'foo'"').

Yes, this is the kind of problem I imagine; which is why I suggest
staying away from information which needs to be more trustworthy than
the average usenet post.  

NNTP network is an example of this sort of P2P network of
intermediaries, and you have some of the same issues.  Vax10 asserts
that Unix22 asserts that allenjs@h... said the following...  No
protection against spoofing or alteration.

Maybe certificates are an answer, but certs get tricky with aggregation
and non-trivial inferences.  Certs are tricky, period -- how many people
use certs for e-mail, even?  So I'm not anticipating a trust system that
functions any better than usenet anytime soon.

• Follow-Ups:
  • RE: Triplets on the Internet
    • From: "Paul Sumner Downey" <Paul.Downey@w...>

• Prev by Date: RE: Meta-somethingorother (was the semantic web mega-pe rmathread thing)
• Next by Date: Re: RDF == the "binary relational model"? was Re: Meta-somethingorother
• Previous by thread: RE: Triplets on the Internet
• Next by thread: RE: Triplets on the Internet
• Index(es):
  • Date
  • Thread