RE: [xri] XRI discovery -vs- resolution

["Gabe Wachob" <gabe.wachob@amsoft.net>]

I think we are risking getting into a semantic morass here. 

This is close to the URI v URL permathread. 

This is even close to the "what is a resource" permathread. 

I originally wanted to call this resolution because it was very much modeled
on DNS resolution and because it was essentially the same function - mapping
an identifier to some metadata - metadata that is very constrained in the
DNS world and infinitely extensible in the web/XML world. 

And also, I don't agree that Discovery means only "Service Discovery"
(there's history on this very topic inside the XRI TC), so I don't think the
service-based distinction is useful/accurate/needed either. 

I do think the term Discovery connotes a more flexible approach to
converting an identifier into metadata than the term Resolution. But if you
dig into this at all, the argument goes in circles. 

XRI provides a standard *interface* for getting (not implementation - that's
partially addressed the GRS specs) metadata about an identifier. We've
always said from day one that that interface could be implemented through
any number of metadata storage systems such as file systems, LDAP servers,
RDBMS systems, RDF stores, etc... That's where the flexibility comes in for
us (and with the delegation model we have) - so in some ways the more
flexible connotation of "Discovery" perhaps does help communicate that
message better.

But more than terms, what is going to communicate the flexibility of XRI is
demonstrations of how XRI authorities can be used to expose pre-existing
namespaces in new and interesting ways... and that's why we have to enable
developers to play around - that whole Thrill of the Hack thing ;-)

	-Gabe

> -----Original Message-----
> From: Ajay Madhok [mailto:ajay.madhok@amsoft.net]
> Sent: Wednesday, December 20, 2006 11:03 PM
> To: 'Steven Churchill'; 'Tan, William'; xri@lists.oasis-open.org
> Subject: RE: [xri] XRI discovery -vs- resolution
> 
> +1
> 
> =Ajay
> 
> 
> -----Original Message-----
> From: Steven Churchill [mailto:steven.churchill@xdi.org]
> Sent: Wednesday, December 20, 2006 2:53 AM
> To: 'Tan, William'; xri@lists.oasis-open.org
> Subject: RE: [xri] XRI discovery -vs- resolution
> 
> 
> I agree with Wil's point.
> 
> With regard to Les' point about DNS resolution, I feel that XRI resolution
> is pretty much analogous: if you consider DNS names to be symbolic names
> for
> a given DNS "resource", then you can view DNS resolution as "obtaining the
> metadata for the given named resource". I feel that XRI resolution is the
> same except that it returns an XRDS as resource metadata instead of DNS
> resource metadata.
> 
> ~ Steve
> 
> -----Original Message-----
> From: Tan, William [mailto:William.Tan@neustar.biz]
> Sent: Tuesday, December 19, 2006 11:28 AM
> To: xri@lists.oasis-open.org
> Subject: RE: [xri] XRI discovery -vs- resolution
> 
> Discovery would be more suited to finding out the available services on an
> authority (bare I-name / I-number). Whenever a service selection criterion
> is present, I think the term "resolution" is appropriate.
> 
> So, there are a few operations on any given XRI:
> 
> Discover(authority) => set of services + canonical IDs + provider ID etc.
> Resolve(xri, type, mtype) => set of services (mostly one) that matches the
> criteria.
> 
> 
> =wil (http://xri.net/=wil)
> 
> 
> 
> > -----Original Message-----
> > From: Victor Grey [mailto:victor@idcommons.org]
> > Sent: Tuesday, December 19, 2006 12:00 AM
> > To: xri@lists.oasis-open.org
> > Subject: Re: [xri] XRI discovery -vs- resolution
> >
> > +1
> > =vg
> >
> >
> > Chasen, Les wrote:
> > > I was wondering what everyone thinks about the idea of instead of
> > > using the term XRI resolution we start calling it XRI discovery?