14 Apr 2009


See also: IRC log


+1.781.643.aaaa, TimBL, +3539149aabb, Alan_Ruttenberg, MichaelHausenblas, JonathanRees




<jar> z

<jar> can you hear me?

<scribe> Scribenick: mhausenblas

<timbl_> BA7059

<timbl_> sunday arriving around 21:00

Voc from Michael


<timbl_> http://www.w3.org/2006/gen/ont.n3


<jar> Tried to open ont.n3 in firefox with Tabulator loaded. Not recognized.


<timbl_> ont.n3

<timbl_> http://rdfs.org/ns/http-sem/rdf


<timbl_> http://rdfs.org/ns/http-sem/n3

is gen:InformationResource owl:sameAs http-sem:TimBLGenericResource ?

<jar> TimBL, if you asked a librarian "are you happy with this ontology" (genont)

the two ontlogies gen: and http-sem should be cross-referenced

<timbl_> F__R



<jar> Maybe they would say it wouldn't work in their world.

<jar> Maybe they wouldn't talk about the Bible at all.

<jar> ... Maybe they would only talk about it as published in a particular way.

<jar> TimBL: What I (TimBL) meant in the generic resources memo is obvious to me.

<timbl_> Yes

<jar> 'generic resource' is the same in the memo and in the ontology.

<timbl_> gen:InformationResource owl:sameAs http-sem:TimBLGenericResource

TRUE: gen:InformationResource owl:sameAs http-sem:TimBLGenericResource


<jar> Thus http://rdfs.org/ns/node/22, 'generic resource', and genont:GenericResource are all the same.

yes, that's how I understood it

<jar> oops... page not found... I mean 'TimBLGenericResource' as seen at http://rdfs.org/ns/http-sem/html

<timbl_> s/temporal part/temporal version/

<jar> michael: Let's say "grounding" not "definition" because some people find "definition" to imply more rigor than we want to have

rather than has temporal part use XXX

<timbl_> I prefer "has temporal version" to "has temporal part"

<jar> TimBL quibbles with "has temporal part"

<timbl_> gen:timeSpecific i sthe relationship

<timbl_> if I understand it

+1 for has temporal version

<jar> ("temporal part" is from BFO I think)

<timbl_> It is clear to me that jar:hasTemporalPart owl:sameAs gen:timeSpecific

http-sem: TimeInvariantResource owl:sameAs onto-gen:TimeSpecificResource

<timbl_> s/it is clear/it seems/

<timbl_> yes

<jar> I'm looking for a relationship R that relates A to B where every rep of A is a rep of B, and at some time t every rep of A is a rep of B (at t), and A doesn't vary through time

<timbl_> agree with http-sem:TimeInvariantResource owl:sameAs onto-gen:TimeSpecificResource

<jar> something like that anyhow.

<jar> OK

<jar> Top 2 boxes of diagram are aligned with terms in genont

<jar> R = genont:timeSpecific

<timbl_> Content-encoding

<timbl_> Comtent-type

<jar> TimBL is bothered because 'encoding' means something else to him

<timbl_> Please though remove the TIMBL "Fixed resource" from the jar diagram.

<jar> TimBL: 'rest representation' is not akin to 'fixed resource'

<jar> jar: But they're in 1-1 correspondence

<jar> TimBL: a fixed resource is something id'd by a URI, and is a snapshot of e.g. the persistence policy as of today

<jar> ... When you do fetch today, you'll get a given set of bits

<jar> ... The bits are not the resource

<jar> ... Fixed res never changes its info content

<jar> ... But the headers can change

<jar> ... Maybe the date changes

<jar> jar: The content-type is part of the FR's identity, yes?

<jar> TimBL: The rep is the http entity, is the same rep

<jar> jar: Are FR and rep in 1-1 correspondence?

<jar> TimBL: OK, yes, let's assume that

<jar> jar: HTTP entity = representation

<jar> ... There are response headers that are not entity headers

<timbl_> There is a 1:1 relationship between fixedresource and response.

<jar> There are natural mappings in both directions

<jar> a FR has only one rep, thus one can map a FR to its rep

<jar> given any rep, one can define a FR that has that rep

other examples for fixed resources are: VCS, Internet Archive, microblogging posts

<jar> TimBL: Beware of FR assertions that might not be true

also: Wiki versions

<jar> TimBL: Style sheet may change, banner ads may change, in which case the thing isn't a FR

<jar> alan: If you fix a spelling mistake it's also not a FR

<timbl_> Specifically source code repositories mught be goo dexample so ffixed resources

<timbl_> data:text/plain;Hello%20World%21

<jar> The resource named by the URI data:text/plain,banana is a FR

<timbl_> great -- we have lots of examples, lets move on


<jar> I will add another box in the diagram for 'fixed resource', Tim

<timbl_> Why?

<timbl_> You already have one.

what is the relation between http:Message and rfc2616:REsource

<jar> because {rest,awww}representation != fixed resource

http: as of in http://www.w3.org/TR/HTTP-in-RDF/

<jar> rfc2616 representation is by definition an rfc2616 entity

<jar> according to roy, every entity is a rep, but not nec. of the resource that was requested

<jar> e.g. entity in a 404 is not a rep of requested resource

ok, seems sensible

<timbl_> Yes

<timbl_> If the response if 200, then ...

<timbl_> the entity is a rep of the resource.

<jar> if response is 200, then the entity is a rep of the (state of) the resource


timbl_: carries -> is ?

<jar> TimBL: Not all representations are HTTP. So HTTP entity is a *subclass* of awww:representation

<jar> UML diagram: Arcs are relations that hold between members of the classes (boxes)

<timbl_> No, we change carries to a subclass relationship

<timbl_> a class relationship rather than an instance relationship.

'Representation of information thing' rdfs:subClassOf HTTP entity

<jar> JAR: I will change the arrow from HTTP entity to representation/encoding to be a subclass relation

actually, other way round HTTP Entity is a kind of enconding ....

<timbl_> http:Entity subClassOf awww:Representation.

<jar> i.e. an HTTP is a kind of awww:representation

<timbl_> 'represnattion"

<timbl_> instead of encoding

<jar> alan: What's the relationship between?

'representation' -> 'representing'

<jar> hmm... that shouldn't edit the irc log...

<timbl_> s/is encoded as/has representation/

<timbl_> if you have to

<jar> JAR doesn't like 'awww:representation'... but may use it to respect tim's dislike of 'encoding'

<scribe> ACTION: Michael to update http://rdfs.org/ns/http-sem with the findings and agreements from today's meeting [recorded in http://www.w3.org/2009/04/14-awwsw-minutes.html#action01]

<trackbot> Created ACTION-18 - Update http://rdfs.org/ns/http-sem with the findings and agreements from today's meeting [on Michael Hausenblas - due 2009-04-21].

<jar> TimBL: the intent of info res and generic resource is the same. same concept

<jar> ... But you can talk about an HTTP resource, a somethingorother endpoint, but that's different

<timbl_> If an HTTP Resource supports POST then it is a HTTPPOstResource which is a subclass o f ServiceEndpoint, maybe.

<jar> michael: Do we have a connection between http-in-rdf and our diagram(s)?

<jar> jar: Is a 'response' the event, or is it the set of events?

<jar> timbl: Dan did a lot of interesting modeling of protocols, just by using links

TimBL: DanC did interesting modelling with HTTP

<jar> ... if one message references another, then it was sent after it

<scribe> ACTION: Query DanC re pointers for HTTP modelling [recorded in http://www.w3.org/2009/04/14-awwsw-minutes.html#action02]

<trackbot> Sorry, couldn't find user - Query

<jar> ... Ask Dan about it (but he may have forgotten about it)

<jar> jar: So a coincidentally identical sequence of bits would be a different 'response', yes? (by that notion of 'response')

<jar> michael: Upcoming issues with media fragments working group, also HTTPbis re sub-resources. Should we look into the future?

<jar> ... I know this is coming up

<jar> jar: the ideal of ontology design is that you can deal with new things as they come along

<jar> michael: jquery is more powerful than prototype - powerful javascript library

<jar> ... it's a javascript ajax library, does more

<jar> TimBL: What is prototype?

<jar> michael: I assumed tabulator was using prototype...


next meeting 28 April

(after WWW ;)


Summary of Action Items

[NEW] ACTION: Michael to update http://rdfs.org/ns/http-sem with the findings and agreements from today's meeting [recorded in http://www.w3.org/2009/04/14-awwsw-minutes.html#action01]
[NEW] ACTION: Query DanC re pointers for HTTP modelling [recorded in http://www.w3.org/2009/04/14-awwsw-minutes.html#action02]
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2009/04/14 14:22:18 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
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)

FAILED: s/temporal part/temporal version/
FAILED: s/it is clear/it seems/
Succeeded: s/A (at t)/A/
Succeeded: s/entity/HTTP entity/
Succeeded: s/Encoding/Representation/
Succeeded: s/is encoded as/representation/
FAILED: s/is encoded as/has representation/
Found ScribeNick: mhausenblas
Inferring Scribes: mhausenblas
Default Present: +1.781.643.aaaa, TimBL, +3539149aabb, Alan_Ruttenberg
Present: +1.781.643.aaaa TimBL +3539149aabb Alan_Ruttenberg MichaelHausenblas JonathanRees
Agenda: http://lists.w3.org/Archives/Public/public-awwsw/2009Apr/0002.html
Got date from IRC log name: 14 Apr 2009
Guessing minutes URL: http://www.w3.org/2009/04/14-awwsw-minutes.html
People with action items: michael query

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.

[End of scribe.perl diagnostic output]