IRC log of vmtf on 2005-07-05

Timestamps are in UTC.

12:58:42 [RRSAgent]

RRSAgent has joined #vmtf

12:58:42 [RRSAgent]

logging to http://www.w3.org/2005/07/05-vmtf-irc

12:59:14 [Ralph]

Meeting: SWBPD VM Task Force

12:59:21 [Ralph]

Agenda: http://lists.w3.org/Archives/Public/public-swbp-wg/2005Jul/0003.html

13:00:55 [aliman_scribe]

aliman_scribe has joined #vmtf

13:02:17 [Zakim]

SW_BPD(VMTF)9:00AM has now started

13:02:24 [Zakim]

+??P2

13:02:35 [Zakim]

+Ralph

13:02:48 [Ralph]

zakim, ??p2 is Alistair

13:02:48 [Zakim]

+Alistair; got it

13:03:21 [Zakim]

+Tom_Baker

13:03:28 [Zakim]

+Danbri

13:03:48 [Ralph]

Regrets: Libby

13:06:35 [Ralph]

Topic: Basic Principles for Managing an RDF Vocabulary

13:06:48 [danbri]

new version, http://www.w3.org/2001/sw/BestPractices/VM/principles/20050705

13:10:54 [Ralph]

Chair: Tom

13:11:22 [Ralph]

rrsagent, please make logs world-visible

13:11:54 [cgi-irc]

cgi-irc has joined #vmtf

13:12:15 [danbri]

hi tom

13:12:16 [Ralph]

cgi-irc is TomB

13:12:20 [cgi-irc]

hi!

13:12:42 [tbaker]

hi again

13:13:42 [Ralph]

Tom: an important function of this note would be to explain good practice for putting a document at term URIs

13:13:48 [Ralph]

... i.e. when you click on a term

13:13:57 [Ralph]

... does Dublin Core do this 'right'?

13:14:06 [Ralph]

... am willing to take an action to fix if not

13:14:20 [danbri]

danbri@fireball:~$ HEAD http://purl.org/dc/elements/1.1/title

13:14:20 [danbri]

200 OK

13:14:20 [danbri]

Connection: close

13:14:20 [danbri]

Date: Tue, 05 Jul 2005 13:14:09 GMT

13:14:20 [danbri]

Accept-Ranges: bytes

13:14:21 [danbri]

ETag: "8d809e-3ad8-4210b8e2"

13:14:23 [danbri]

Server: Apache/1.3.28 (Unix) mod_jk/1.1.0

13:14:25 [danbri]

Content-Length: 15064

13:14:27 [danbri]

Content-Type: text/plain

13:14:29 [danbri]

Last-Modified: Mon, 14 Feb 2005 14:42:42 GMT

13:14:33 [danbri]

Client-Date: Tue, 05 Jul 2005 13:14:25 GMT

13:14:35 [danbri]

Client-Peer: 132.174.1.71:80

13:14:37 [danbri]

Client-Response-Num: 1

13:15:51 [aliman]

http://livehttpheaders.mozdev.org/

13:16:27 [Ralph]

[[

13:16:27 [Ralph]

homer% HEAD -S http://purl.org/dc/elements/1.1/title

13:16:27 [Ralph]

HEAD http://purl.org/dc/elements/1.1/title --> 302 Found

13:16:27 [Ralph]

HEAD http://dublincore.org/2003/03/24/dces#title --> 200 OK

13:16:27 [Ralph]

Connection: close

13:16:28 [Ralph]

Date: Tue, 05 Jul 2005 13:15:50 GMT

13:16:30 [Ralph]

Accept-Ranges: bytes

13:16:32 [Ralph]

ETag: "8d809e-3ad8-4210b8e2"

13:16:34 [Ralph]

Server: Apache/1.3.28 (Unix) mod_jk/1.1.0

13:16:36 [Ralph]

Content-Length: 15064

13:16:38 [Ralph]

Content-Type: text/plain

13:16:40 [Ralph]

Last-Modified: Mon, 14 Feb 2005 14:42:42 GMT

13:16:42 [Ralph]

Client-Date: Tue, 05 Jul 2005 13:15:55 GMT

13:16:44 [Ralph]

Client-Peer: 132.174.1.71:80

13:16:46 [Ralph]

Client-Response-Num: 1

13:16:48 [Ralph]

homer%

13:16:50 [Ralph]

]]

13:16:57 [Ralph]

DanBri: TAG would prefer 303 See Also rather than 302 Found

13:17:40 [Ralph]

s/Also/Other/

13:18:14 [danbri]

http://www.w3.org/Protocols/rfc2616/rfc2616-sec10.html#sec10.3

13:18:31 [danbri]

10.3.3 302 Found

13:18:40 [danbri]

ralph: resource resides temp at different uri

13:18:43 [danbri]

...for 302

13:18:53 [danbri]

303 says, can be found under a different URI and SHOULD be retrieved using a GET method on that resource

13:19:50 [tbaker]

Danbri: send note to PURL developer list, ask whether

13:20:03 [tbaker]

they would consider making 303 an option.

13:20:30 [aliman]

I'm just going to restart mozilla after install live http headers

13:20:44 [tbaker]

Ralph: is there a reason to ask them to make 303 an option, or

13:20:51 [tbaker]

maybe just ask them to change it?

13:21:02 [tbaker]

Is there a reason for staying with 302?

13:21:18 [aliman]

aliman has joined #vmtf

13:21:57 [Ralph]

http://www.w3.org/Protocols/rfc2616/rfc2616-sec10.html#sec10.3.3

13:21:58 [Ralph]

[[

13:21:58 [Ralph]

10.3.3 302 Found

13:21:58 [Ralph]

The requested resource resides temporarily under a different URI. Since the redirection might be altered on occasion, the client SHOULD continue to use the Request-URI for future requests.

13:21:59 [Ralph]

]]

13:22:02 [tbaker]

ACTION: Danbri, contact PURL developers

13:23:35 [tbaker]

Al: should we be saying "return rdf/xml"

13:24:01 [tbaker]

Ralph: strictly speaking, the way RDF schema is served up,

13:24:11 [tbaker]

it is a text document with lots of angle brackets.

13:24:17 [Ralph]

Alistair: should dublincore.org return something other than Content-Type: text/plain ?

13:25:02 [danbri]

if we make HTML available too, then what is "http://dublincore.org/2003/03/24/dces#title" the URI of?

13:25:14 [danbri]

q+ to reopen that can of worms

13:26:12 [aliman]

http://www.w3.org/2004/02/skos/core

13:26:23 [tbaker]

Al: does the TAG resolution preclude us from content-negotiating on a hash URI?

13:26:39 [Zakim]

danbri, you wanted to reopen that can of worms

13:26:45 [tbaker]

Given that every term in SKOS Core vocabulary on this base.

13:27:07 [Ralph]

DanBri: because DC is redirecting to ...#title some of the issues go away

13:27:17 [Ralph]

... the meaning of #frag is relative to the media type

13:27:30 [tbaker]

Danbri: still left with ugly business of content negotiation on hash thing

13:27:46 [tbaker]

Danbri: if only RDF, no problem.

13:27:47 [Ralph]

... in an application/rdf+xml document we can say Title means whatever we want

13:28:06 [Ralph]

... but if there is also an HTML document there, then we're stuck with #Title meaning something specific in HTML

13:28:46 [Ralph]

Alistair: I wrote once before that #URIs must _not_ content negotiate

13:28:47 [tbaker]

Al wrote: "hash URIs must not support content negotiation"

13:29:01 [aliman]

http://esw.w3.org/topic/SkosDev

13:29:09 [tbaker]

Was that a tenable position?

13:30:40 [tbaker]

Ralph: could take less extreme position:

13:31:03 [tbaker]

that if you serve an HTML document, it must not use any of the RDF concepts as IDs

13:31:19 [tbaker]

Al: it must not overload any of the secondary []

13:31:43 [danbri]

q+ to suggest purl for DC might redirect to a URI without a #, maybe?

13:31:49 [tbaker]

Ralph: All IDs using the HTML document should refer to HTML fragments

13:31:57 [tbaker]

and should not overlap with RDF terms

13:32:09 [Ralph]

perhaps the first 'should' is a 'must'

13:32:12 [Ralph]

(to HTML validate)

13:32:13 [tbaker]

s/should/must/

13:32:56 [tbaker]

Al: peeling a banana with tweezers

13:33:13 [danbri]

ack danbri

13:33:13 [Zakim]

danbri, you wanted to suggest purl for DC might redirect to a URI without a #, maybe?

13:34:42 [Ralph]

DanBri: e.g. what if all DC terms redirect to http://dublincore.org/2003/03/24/dces

13:35:13 [Ralph]

... in practice, currently dublincore.org returns the same document for all DC terms anyway

13:35:20 [danbri]

http://rdfweb.org/viewcvs/viewcvs.cgi/xmlns.com/htdocs/foaf/0.1/.htaccess

13:35:28 [danbri]

RedirectTemp /foaf/0.1/Person http://xmlns.com/foaf/0.1/

13:35:28 [danbri]

RedirectTemp /foaf/0.1/Agent http://xmlns.com/foaf/0.1/

13:35:28 [danbri]

RedirectTemp /foaf/0.1/Project http://xmlns.com/foaf/0.1/

13:35:28 [danbri]

RedirectTemp /foaf/0.1/Image http://xmlns.com/foaf/0.1/

13:35:28 [danbri]

RedirectTemp /foaf/0.1/Document http://xmlns.com/foaf/0.1/

13:35:32 [danbri]

etc

13:37:01 [tbaker]

Ralph: we need to find an answer to Alistair's content negotiation question

13:37:03 [danbri]

ralph: we should address this q of conneg, since users would

13:37:14 [tbaker]

When users (as opposed to tools) use URIs, different representations

13:37:15 [danbri]

...benefit from being able to see something useful in their browsrs

13:37:27 [tbaker]

That is a very common question users have.

13:38:26 [tbaker]

Al: everyone still treats HTTP servers as file servers

13:38:30 [berva]

berva has joined #vmtf

13:38:43 [Zakim]

Ralph, you wanted to ask the semantics of http://dublincore.org/2003/03/24/dces#title

13:42:59 [Ralph]

Ralph: http://dublincore.org/2003/03/24/dces is something like a namespace document or an RDF schema but it's not necessarily either one as it doesn't appear at the namespace URI

13:43:07 [aliman]

tbaker: title of http://dublincore.org/2003/03/24/dces is ...

13:43:09 [Ralph]

... but we may find it useful to have a role name for this document

13:43:18 [aliman]

<dc:title xml:lang="en-US">The Dublin Core Element Set v1.1 namespace providing access to its content by means of an RDF Schema</dc:title>

13:43:26 [aliman]

I don't like that at all

13:44:21 [danbri]

"Dublin Core Element Set vocabulary description"?

13:44:45 [aliman]

actual RDF statement is (<http://purl.org/dc/elements/1.1/> dc:title 'The Dublin Core Element Set v1.1 namespace providing access to its content by means of an RDF Schema')

13:45:25 [danbri]

q+ to suggest less is more "Dublin Core Element Set" or similar

13:45:37 [aliman]

+1 on danbri

13:47:39 [Zakim]

danbri, you wanted to suggest less is more "Dublin Core Element Set" or similar

13:47:41 [aliman]

'An RDF Vocabulary Description Document?'

13:48:06 [aliman]

q+ to ask bout historical URIs cf redirect URIs

13:48:43 [Ralph]

DanBri: shortening the title to "The Dublin Core Element Set" might reduce some confusion

13:49:08 [aliman]

cf. {<http://www.w3.org/2004/02/skos/core> dc:title 'SKOS Core Vocabulary'}

13:49:24 [Ralph]

Tom: we thought about that, as the title looks confusing when displayed by browsers, but we worried that making the change might lose some information

13:49:51 [danbri]

FOAF has: <owl:Ontology rdf:about="http://xmlns.com/foaf/0.1/" dc:title="Friend of a Friend (FOAF) vocabulary" dc:description="The Friend of a Friend (FOAF) RDF vocabulary, described using W3C RDF Schema and the Web Ontology Language." dc:date="$Date: 2005/06/03 07:01:49 $">

13:50:09 [tbaker]

Ralph: suggests holding off on changing title

13:50:18 [danbri]

q+ to note the dc:description can carry more info

13:50:39 [tbaker]

First we need to understand the model and convey that somewhere (not in the title)

13:50:55 [aliman]

http://dublincore.org/2003/03/24/dces#title

13:51:00 [Zakim]

aliman, you wanted to ask bout historical URIs cf redirect URIs

13:52:11 [Ralph]

Tom: there's a URI that is an anchor in an HTML document ...

13:52:12 [tbaker]

http://dublincore.org/usage/terms/history

13:52:55 [tbaker]

http://dublincore.org/usage/terms/history/#contributor-003

13:53:16 [Ralph]

Tom: this usage is not a high-profile one and could be changed

13:53:55 [aliman]

tom: these URIs are anchors in an html document

13:54:38 [Ralph]

Tom: the history document is currently missing a change note

13:55:14 [Ralph]

... the URI functions as a URI for a description of a term at a specific point in time

13:55:41 [aliman]

q+ to peel a banana

13:56:46 [Ralph]

q+ to pull on the peel re: history document

13:57:22 [Ralph]

Tom: we don't give http://dublincore.org/usage/terms/history/ a very high profile as we think it might confuse users

13:57:38 [Ralph]

... we thought about putting this in RDF but we were not sure what the requirements were

13:58:06 [Ralph]

... we were aware that good practice might evolve but we wanted to capture the information somewhere

13:58:14 [danbri]

FOAF index.rdf is under CVS control, http://rdfweb.org/viewcvs/viewcvs.cgi/xmlns.com/htdocs/foaf/0.1/index.rdf

13:58:18 [Ralph]

DanBri: do you have history elsewhere, e.g. in CVS?

13:58:24 [danbri]

...but not exposed very easily to RDF processors

13:58:27 [danbri]

eg. sparql

13:58:29 [Ralph]

Tom: all of the RDF schemas are still in the Web

13:59:07 [Ralph]

q+ to note that this 'how to version vocabularies' question is an open issue for VM and/or PORT

14:00:14 [danbri]

ack danbri

14:00:14 [Zakim]

danbri, you wanted to note the dc:description can carry more info

14:00:36 [Ralph]

DanBri: looking at how the FOAF namespace document describes itself ...

14:01:11 [tbaker]

Danbri: in dc, push detail into dc:description?

14:01:11 [Ralph]

... the dc:description has some of the things that the Dublin Core document puts into its dc:title

14:01:26 [Ralph]

... e.g. the phrase "providing access via ..."

14:01:43 [tbaker]

danbri: short, snappy title (DC Namespace), then put versioning

14:01:50 [tbaker]

information into dc:description

14:01:57 [aliman]

SKOS Core <dc:description xml:lang="en">An RDF vocabulary for encoding simple concept schemes such as thesauri and subject heading lists.</dc:description>

14:02:41 [Zakim]

aliman, you wanted to peel a banana

14:02:55 [danbri]

(ok seems i misunderstood ralph; agreement it could go into dc:description)

14:03:04 [tbaker]

al: DC has the beginning of tight management system for RDF vocabulary

14:03:05 [Ralph]

Alistair: DC has the beginnings of what looks like a configuration management system

14:03:10 [tbaker]

we should build on that.

14:03:22 [tbaker]

But questions: those URIs that point to anchors in versioning history document:

14:03:38 [tbaker]

If each identifies the description of a term at a given point in time,

14:03:57 [tbaker]

then fine to use. But if they identify version, then could not do

14:04:04 [tbaker]

content negotiation.

14:04:19 [tbaker]

Patrick S things bad to URIs to versions of a term, but good to

14:04:26 [tbaker]

give URIs to descriptions of a term.

14:04:57 [tbaker]

Al: not "historical version of term".

14:04:59 [Ralph]

Tom: we are giving URIs for versions of a description of a term

14:05:21 [Ralph]

Ralph agrees this is a subtle and important distinction

14:05:22 [tbaker]

Danbri: we are telling stories here...

14:05:37 [tbaker]

-- one term and we are saying things about it at different times

14:05:56 [tbaker]

In early RDF schema spec, I encouraged: if you changed it,

14:06:07 [tbaker]

you have something new... new schema.... etc

14:06:21 [tbaker]

Ralph: the threashold is very community-sensitive

14:06:35 [tbaker]

Tom: That is the point of the DCMI namespace policy

14:07:11 [tbaker]

Ralph: interpret patrick's caution as being against implicitly relying on some sort of URI similarity to convey semantics

14:07:59 [tbaker]

Ralph: even in RDF, original 1999 namespace, and current one, differ in subtle ways, but chose not to change for practical reasons

14:09:00 [Ralph]

s/not to change/not to change the names (i.e. URIs) of the terms/

14:09:02 [tbaker]

Danbri: "we altered the semantics of the terms"... -- to correspond to "what the semantics actually are" (authorial intent, deployment)

14:09:38 [Ralph]

DanBri: different interpretation is 'we changed the description to better describe the real semantics of the property'

14:10:06 [berva]

berva has left #vmtf

14:10:40 [danbri]

ralph: i trust that the dc community has the right body of expertise to get these things right

14:11:02 [danbri]

...when a description can be clarified to match some notion of the 'true semantics' vs when there is enough difference to merit a new name

14:11:35 [tbaker]

Danbri: fine for DC-length vocabularies. But in DL-type vocabularies...

14:12:02 [tbaker]

Ralph: Some of that perspective may be driven by application behavior.

14:12:38 [tbaker]

Danbri: DC community focuses on prose definitions. OTher communities focus on inferencing structures.

14:13:09 [tbaker]

Ralph: But RDF core (and others) had practical problem...

14:13:42 [Ralph]

[acking myself just to preseve my thoughts]

14:13:44 [Ralph]

ack me

14:13:44 [Zakim]

Ralph, you wanted to pull on the peel re: history document and to note that this 'how to version vocabularies' question is an open issue for VM and/or PORT

14:20:03 [Ralph]

adjourned

14:20:05 [Zakim]

-Tom_Baker

14:20:07 [Zakim]

-Alistair

14:20:09 [Zakim]

-Ralph

14:20:14 [Zakim]

-Danbri

14:20:15 [Zakim]

SW_BPD(VMTF)9:00AM has ended

14:20:16 [Zakim]

Attendees were Ralph, Alistair, Tom_Baker, Danbri

15:36:20 [Ralph]

zakim, bye

15:36:20 [Zakim]

Zakim has left #vmtf

15:36:22 [Ralph]

rrsagent, bye

15:36:22 [RRSAgent]

I see 1 open action item:

15:36:22 [RRSAgent]

ACTION: Danbri, contact PURL developers [1]

15:36:22 [RRSAgent]

recorded in http://www.w3.org/2005/07/05-vmtf-irc#T13-22-02