See also: IRC log
<DanC> raman, it's traditional to have the machine draft minutes, then edit them and check them in under /2001/tag/YYYY/MM, and then mail a pointer to www-tag. But practice varies considerably.
so you dont want me to scribe on IRC?
<DanC> sorry, yes... the machine drafts them based on what you write in IRC
I'd prefer to just type in here --- dont have cycles to turn it into a new work activity of its own:-)
last week's minutes approved subject to HT's final edits.
<DanC> (henry, so the minutes will stay at http://www.w3.org/2006/03/28-tagmem-minutes.html ? I want their address in today's minutes)
<ht> DanC, yes.
<Norm> So do we have tentative dates in Oct?
<Norm> Yes, nevermind.
<DanC> I prefer a 2 day meeting. I'm not sure I can muster 3 days of steam.
also issue: Venice is a long way to travel for 2 days
<DanC> ACTION: DanC to explore Venice meeting venue [recorded in http://www.w3.org/2006/04/04-tagmem-minutes.html#action01]
<DanC> June meeting logistics
<Zakim> DanC, you wanted to note a possible conflict with the June meeting
<scribe> ACTION: Norm to send hotel details in a week [recorded in http://www.w3.org/2006/04/04-tagmem-minutes.html#action02]
Vincent: AC Meeting At WWW 2006 Edinburgh
Tag Summary to be reviewed by email
Possible TAG session at AC Meeting
Conclusion: no active TAG interest in adding to AC Meeting agenda
<ht> http://www.w3.org/2006/03/28-tagmem-minutes.html is now cleaned up
<DanC> our recent WD
<inserted> Issue URNsAndRegistries-50
<inserted> Issue URNsAndRegistries-50
HT: Introduces URN discussion
<ht> http://www.w3.org/2001/tag/doc/URNsAndRegistries-50.html#metadata
<DanC> following my nose from the agenda, I get 1.7 2006/04/04 16:32:06 ; can anybody confirm?
<ht> yes, DanC
HT: Document covers a number of
different proposals/patterns of use for new URI schemes etc.
--- identify things we think of as information resources
... reorged to make section2 a simple, short summary of why the
TAG doesn't think that for many purposes such things are
different from http: in interesting ways.
<DanC> (implicit? darn.)
If not, then you shouldn't be forking the Web
HT: would like feedback on the document style --- and also wants input on sections 2.7 and 2.8
<timbl> ?NRI
<Norm> "New Resource Identifiers" I expect
DanC: NRI is jarring
<DanC> "recent proposals ([RFC 3688], [oasis URN], [XRI])"
HT: we're not just responding to XRIs --- we want to cover anyone who wants to mint a new URN subspace for electronic resources e.g. New Zealand Govt
DanC: suggests adding info:
TimBL we can discuss Info:
TimBL: convinced XRIs are a bad idea, not sure if info: is any worse than mailto:
DanC: info: looks exactly like XRI
Some confusion as to how info: thingies are resolved
tvr: recommend not using an acronym like NRI
<timbl> nRI
<timbl> _RI
TimBL: one of the problems with
HTTP is that one cannot get a URI space for ever.
... perhaps pull that out as an issue?
Perpetual Resource Identifiers?
HT: believes that the XRI spec as it stands no longer claims to solve/address the perpetual resource problem, since they also use DNS for resolution
<DanC> [[
<DanC> The http: URI scheme implements a two-part approach to identifying resources. It combines a universal distributed naming scheme for owners of resources with a hierarchical syntax for distinguishing resources which share the same owner. Widely available mechanisms (DNS and web servers, respectively) exist to support the use of http: URIs to not only identify but actually retrieve representations of information resources.
<DanC> ]]
HT: as we've worked on the finding, we have moved towards DanC's position --
DanC pastes it in below
<DanC> right under 3 The value of http: URIs
<Zakim> DanC, you wanted to say yes, timbl, let's hope folks considering DAV: and dix: URIs schemes _will_ look here
<timbl> or "the poverty of"
<Zakim> timbl, you wanted to suggest a title "The dangers of URNs and Registries"
<DanC> (formatting around 2.7 Rich authority is goofy)
HT: 2.6 URI for an object vs URI
to the metadata for that object
... what do people think?
<DanC> (this is another place where I'd find a full survey more useful.)
HT: clarifies that here metadata is not http header like metadata in response to question: if you have a uri to metadata and uri to object, then how do you keep them in sync?
<ht> The kind of metadata people are looking for in the nRI case is things such as dc:creator etc.
<timbl> Link:
TimBL: given foo.html then foo.html,meta might give you a lot of extensible metadata
<Zakim> DanC, you wanted to note that the state-of-the art in metadata is the <link> and to note that the state-of-the art in metadata is the <link> element and Link: HTTP header; see
<ht> HST notes that this only works for HTML
<DanC> the Link: header field works
<ht> HST asks for what value of 'works'
http://www.mozilla.org/projects/netlib/Link_Prefetching_FAQ.html
<timbl> Cool by design (tm)
<DanC> I know somthine bout what trustred resolution _might_ mean, but I was hoping to take advantage of henry's survey work
DO: reviewing Section 4
<DanC> (huh? I thought the point of URIs is that they're context free; i.e. that The Web is _the_ context.)
<DanC> (btw, timbl, on permanent domains, note that example.org is permanently allocated by the IETF, i.e. the DNS technical standardization body. I think it's probably efficient to just do that again whenever necessary.)
<Zakim> ht, you wanted to say that XRIs _are_ meant to be dereferencable
DO: XRIs use XRI namespace for XRI descriptors
<Zakim> DanC, you wanted to note that there are scheme-independent resolution mechanisms (DDS)...
<ht> DO, HST to consider two examples, one using URNs for namespaces and one using XRIs for documents
<Zakim> DanC, you wanted to note that there are scheme-independent resolution mechanisms (DDS)...
DO: Suggest splitting into two examples A) namespaces a la oasis B) documents that are meant to be location independent
DanC idea of a uri scheme that cannot be looked up sounds absurd
TimBL: shall we set up a W3C resolver that does its best to resolve any types of resource?
<DanC> (that was suggested in jest, I'm pretty sure)
<timbl> http://undereferencable.net/ can only be used for dereferencing undereferencable URIs
<Zakim> DanC, you wanted to ask what's the path from here to XRI proponents
<dorchard> I can take another swag at 2 example.s
<DanC> raman, can you start with http://www.w3.org/2006/04/04-tagmem-minutes.html , edit it a bit, and send it to www-tag?