Re: [mmox] Global Object Identification

[Mike Dickson <[e-mail]>]

I mentioned it in a separate message, but have you had a look at the
NCRI handle service for this. It really seems spot on.  There is the
need to address how digital assets map into it and what metadata is
useful to carry around but the handle service was designed specifically
to handle the secure resolution of URI's to "providers".

Mike

On Fri, 2009-06-19 at 21:08 +0000, Infinity Linden wrote:
> Historically, systems like Second Life and OpenSim have used UUIDs to
> identify objects. As a number of us started thinking about
> interoperability, the idea of the "naked UUID" (i.e. - a UUID that
> exists without context) would pose a problem. I think there's general
> agreement that moving to a system where you use URIs to identify
> objects is "a good idea." The idea being that if you have a URL,
> you'll always know where to go to get meta-information about the
> object (if not the digital representation of the object itself.)
> 
> I think there's been discussion about "rebooting" the digital asset
> sub-system to make it a better fit for a distributed virtual world
> (and more scalable) to include the use of URIs to address objects,
> what classes of objects should be defined in the specification, and
> how one extends the list of defined object classes. In previous drafts
> of the charter, there was an objective for this listed as "OGP Virtual
> World Primitive Object Format." A title like "Digital Asset Addressing
> and Definition" might be a better title.
> 
> But the take-away is, there's been some thought about this. John
> Hurliman has been working on a full-on asset service. OpenSim and
> Second Life  (and probably realXtend and others) would benefit from a
> well defined asset service that uses URIs for references, perhaps
> using the existing asset classes defined by SL/OpenSim and maybe
> realXtend.
> 
> -cheers
> -meadhbh
> 
> On Fri, Jun 19, 2009 at 12:18 AM, Tommi
> Laukkanen<[e-mail]> wrote:
> > Dear All
> >
> > Anyone have suggestion how we can achieve indisputable universal
> > identification of objects. It would be a good start for
> > interoperation.
> >
> > Respectfully,
> > Tommi of bubblecloud.org
> > _______________________________________________
> > mmox mailing list
> > [e-mail]
> > https://www.ietf.org/mailman/listinfo/mmox
> >
> _______________________________________________
> mmox mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/mmox