[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [xri] My discussion about XRI on IRC
Yes, that's what I was trying to convey... You interpolated correctly! -Gabe > -----Original Message----- > From: Chetan Sabnis [mailto:chetan.sabnis@epok.net] > Sent: Wednesday, April 20, 2005 2:53 PM > To: Wachob, Gabe > Cc: xri@lists.oasis-open.org > Subject: Re: [xri] My discussion about XRI on IRC > > > Good discussion. Glad to see that they were receptive. > > I am wondering about some of your responses in the IRC. > Specifically, > the (authOne) authority does have control over *foo in that whatever > auhtority (authOne) maps to...that authority is the one that gets to > control registration *foo and decide what the descriptor for *foo is. > Assuming that the *foo descriptor has another Authority element, then > that other Authority element has control over any subsequent > registered > subsegments. > > In a nutshell, (authOne) controls what descriptor gets > returned for *foo > but does not necessarilly have any control over what > descriptor gets for > *foo*bar if *foo's authority element is hosted elsewhere. > > Is that what you were trying to convey? > > > Chetan > > 01:47:37 <sbp> so then that raises a further question: > > 01:48:09 <sbp> if you have something along the lines of the > xri://(authOne)...(authTwo)... URI that I'm so keen on, I > presume you'll > first resolve authOne and then authTwo, yes? > > 01:48:17 <GabeW> well > > 01:48:28 <GabeW> XRI resolution treats those as single units > > 01:48:32 <sbp> if so, that would mean authTwo is still > dependent on authOne > > 01:48:35 <GabeW> XRI resolution goes from left to right > > 01:49:03 <GabeW> and therefore, (authTwo) gets resolved, as an opaque > string, in the context of the prefix of subsegments to the left of it > > 01:49:06 <GabeW> ie > > 01:49:18 <GabeW> xri://(authOne)*foo*(authTwo) > > 01:49:36 <GabeW> in that case, (authTwo) is resolved at the authority > identified by xri://(authOne)*foo > > 01:49:49 <sbp> okay, so authOne *explicitly* has to say that > anything to > the right can be extensible by using yet another authority component? > > 01:50:01 <GabeW> well > > 01:50:05 <GabeW> it delegates > > 01:50:12 <GabeW> so foo is the one who decides > > 01:50:18 <GabeW> that is xri://(authOne)*foo > > 01:50:22 <sbp> oh > > 01:50:35 <sbp> I thought foo was just an identifer that > authOne controlled > > 01:50:36 <GabeW> its just left-to-right delegation > > 01:50:39 <GabeW> it is > > 01:50:43 <GabeW> well > > 01:50:49 <GabeW> or it can be an independent authority > > 01:51:04 <GabeW> *'s are the delimiters between authorities > > 01:51:21 <sbp> but authOne gets to choose that? so how is > xri://(authOne)*foo*(authTwo) different from > xri://(authOne)*foo*authTwo? > > 01:52:12 <GabeW> its not different except that the string > (authTwo) is > not a legal cross reference - it would be (=AuthTwo) ... > Xrefs are just > a syntactic way of including a URI - not unlike quotes in the english > language > > 01:52:29 <GabeW> and no, authOne doesn't really have control > over foo in > our case > > 01:52:44 <sbp> this is quite mind-bending! > > Wachob, Gabe wrote: > > As XRI is getting more visibility with W3C folks, I spent > snme time on > > IRC discussing XRI with Sean Palmer, a w3c and semantic > web-interested > > person on #swig - the "Semantic Web Interest Group" IRC > channel. This > > channel is inhabited by a good number of W3C folks, > including several of > > the TAG members. > > > > The logs are below, in case anyone is interested in seeing > my informal > > discussion. It was clear that the introductory document wasn't very > > effective with this crowd as the use cases described in the > document > > didn't seem particularily difficult to meet and it wasn't > obvious how > > XRI was addressing those use cases more directly than one > could address > > them with HTTP URIs... > > > > http://ilrt.org/discovery/chatlogs/swig/2005-04-20.html#T00-33-58 > > > > -Gabe > > __________________________________________________ > > _gwachob@visa.com <mailto:gwachob@visa.com>_ > > Chief Systems Architect > > Technology Strategies and Standards > > Visa International > > Phone: +1.650.432.3696 Fax: +1.650.554.6817 > > >
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]