RE: [bt-spec] URIs and address-as-X (MAJOR)

["WEBBER,JIM (HP-UnitedKingdom,ex1)" <jim_webber@hp.com>]

Peter:

> No, you meant URL.  URNs do not "sit within the context of a 
> protocol". They are unambiguous names for things. They no 
> more tell you where something is than "UK Passport number 
> 020011112" tells you where it or its holder is.

Which can resolve unambiguously in a given situation, e.g. immigration.

> > The fact remains: URIs can be used as a globally unambiguous naming 
> > mechanism, whose form is defined according to the rules of the 
> > protocol that the particular URI is bound to, which is more 
> > conventional (i.e. accessible
> > to the lay person) and loses nothing in expressiveness over 
> the current
> > address+identifier scheme.
> 
> I agree it is better than the current (0.9.1) 
> address+identifier scheme.

Good :-)

> We think it is worse than the proposed 77, 78  
> location-unaware identifier, identification-independent 
> address scheme.

To quote, "To summarise: The "Identifier" will be defined as a URI, and
therefore as globally unambiguous."

Which is pretty much what Mark and I said. Having decided to go that way, we
can then simplify a bunch of stuff, as per previous mail exchanges.

Jim