<mhausenblas> heya

TAG Meeting prep

<mhausenblas> sorry, bit late (editing RDB2RDF Use Cases for today's meeting ATM)

<mhausenblas> will dial in in a sec

dbooth: An update at the TAG F2F would be a good idea
... What we've done so far
... what the current issues are

jar: Larry M's confusion about what we're doing
... Dan C has prepared an RDF model theory primer

dbooth: Need motivating examples

jar: LM has always pushed against use of http: for anything other than HTTP

mh: People *are* using http: URIs
... e.g. linked data cloud

dbooth: RDF semantics *permits*

<mhausenblas> http://www.w3.org/TR/rdf-mt/

jar: But the LOD cloud is mostly # URIs, and mostly not "metadata subjects"

mh: Let's list the usages in the wild
... Web services?

<dbooth2> dbooth: we should simply stipulate that http uris are used for non-web things.

<mhausenblas> <soap:binding style="document" transport="http://example.com/smtp"/>

<dbooth2> jar: Also xml namespaces

db: Can't convince anyone about http: URIs...

jar: but we could at least maybe convince someone that it's consistent / coherent ?

<dbooth2> dbooth: some outstanding issues: 1. how do URIs become bound to resources as names? 2. What do redirects mean in RDF semantics?

how might one treat redirects formally

<dbooth2> dbooth: 3. What kinds of resources should return 200 responses?

<dbooth2> dbooth: 4. Authority in HTTPbis: rfc2616 says 203 response is non-authoritative, but it doesn't define "non-authoritative"

<dbooth2> dbooth: 5. How does an RDF consumer determine what resource is intended to be bound to each URI?

residesAt(R,U,t)

<dbooth2> jar: http spec talks about a resource residing at a location. Maybe that can change over time.

R = resident(U,t)

X = resident(U,t) the thing resident at U at time t

to formulate a GET asking for W(X, ...) I need a URI

so we want a URI U such that X = resident(U,t)

<dbooth2> dbooth: 6. For a given resource, how does one know what URI can be used to access a representation of it?

6 is the same as 1

<dbooth2> dbooth: no, #6 is for things that can have representations.

6'. For a given resource, how does one choose a URI that can be used to refer to the resource?

<dbooth2> dbooth: In TAG mtg, I think you should point to the draft doc you gave to show what we've done so far.

for any advice around (3), it would be nice if the advice were justified (as the solution to some real problem)

dbooth2: for certain things, such as journal articles or books, what kind of URI should be used to name them?

For those who want to use the 303 hack for a non-IR, does it work to put the 303 URI as the target of a 302 (e.g. from purl.org)?

<mhausenblas> http://lab.linkeddata.deri.ie/alodin/agent/

GET U1 -> 302 Location: U2, GET U2 -> 303 Location: U3, GET U3 -> 200 description of <U1>

<mhausenblas> http://purl.org/NET/dady

<dbooth2> ACTION: jar to draft TAG overview of where the AWWSW group is and what issues we're still facing [recorded in http://www.w3.org/2010/05/25-awwsw-minutes.html#action01]

<trackbot> Created ACTION-37 - Draft TAG overview of where the AWWSW group is and what issues we're still facing [on Jonathan Rees - due 2010-06-01].

<mhausenblas> ACTION: mhausenb send memo re LD clients to the list/jar [recorded in http://www.w3.org/2010/05/25-awwsw-minutes.html#action02]

<trackbot> Created ACTION-38 - Send memo re LD clients to the list/jar [on Michael Hausenblas - due 2010-06-01].

This is scribe.perl Revision: 1.135  of Date: 2009/03/02 03:52:20  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: i/sorry, bit late/Topic: TAG Meeting prep
No ScribeNick specified.  Guessing ScribeNick: jar
Inferring Scribes: jar
Default Present: +1.216.445.aaaa, dbooth2, jar, mhausenblas
Present: Jonathan_Rees David_Booth Michael_Hausenblas
Got date from IRC log name: 25 May 2010
Guessing minutes URL: http://www.w3.org/2010/05/25-awwsw-minutes.html
People with action items: jar mhausenb