W3C

- DRAFT -

SW Best Practices and Deployment and HTML WG RDF-in-HTML Task Force telecon

28 Jun 2005

See also: IRC log

Attendees

Present
Ben_Adida, MarkB_, Danbri, Emiller, jjc
Regrets
Steven, Ralph
Chair
Ben Adida
Scribe
danbri_scribe

Contents


 

em: GRDDL status?

ben: dormant in TF currently

admin

IPTC

IPTC upcoming discussion(s)

scribe: meeting under discussion for July (London), re potential adoption of RDF/A and XHTML2 by IPTC

em: re IPTC... july 8 meeting in london

nearby: http://lists.w3.org/Archives/Public/www-rdf-interest/2005Jun/0016.html

em: ...morning of 8th

jjc: i could probably attend quite easily

em: yes pls :)
... biggest issues i see, some modeling Qs that need sorting out. Various IPTC-related discussions across SWIG and DC lists, hard to get a clear sense of requirements

ben: i agree. discussed possibility w/ danc of having a telecon pre-meeting to scope and make better use of f2f time

em: there are some good summaries around (esp mark's). being around a whiteboard will be a big help.
... its v important to identify 'customers' for our tech
... thats my interest in seeing this work happen
... to extent that we can have 2 or 3 of them satisfied by the same underlying tech, those are helpful
... but also helps us make a case for wider deployment (in toolkits, browsers, etc)

danbri: seems v healthy to me

mark: i've been involved w/ both orgs, and would even say that if we don't get this kind of adoption, sw won't happen
... i was also in Dublin for w3c mobile meeting
... presented the xhtml2 story
... enthusiasm for the metadata aspect makes you think that if we could get mobile world + newsvendors, both using xhtml2, some major possibilities

em: working with them to acid test this stuff... and help use them as examples, indicators for others to follow

sorry for noise

ben: completely agree. cusomers are a big deal
... we want this to be real, ...

<Zakim> danbri_scribe, you wanted to ask if we can we fix the agenda and to note that abbreviated URIs for subjects and objects of rdf statements seems to be a req and to ask ericm whether

danbri: allowing qnames instead of uris would be a big win

ben: we talked about that last week

danbri: i saw something (but didn't read detail yet)

<Zakim> danbri_scribe, you wanted to suggest mark and ben collab on agenda prep for next time

jjc: is there an active HP rep on the HTML WG? and active?

mark: melinda(sp?) involved w/ print stuff

jjc: I could investigate the HP/HTML WG connection... get our RDF interest represented

em: ...if we could get not just us from the RDF camp to talk about qnames, human readability etc., but from early adopters (eg., mobile web, news, ...)
... draw attention to issues from outside the rdf 'camp'

mark: generally those on the wg have the interests of HTML authors at heart
... there are other communities beyond the HTML author (news, mobile, ...); good to have backup of use cases from these wider groups

em: we have to focus on 'uses' as well as 'user'

<Zakim> danbri_scribe, you wanted to ask if we should consider this a meeting of the TF/WG, or just a related meeting?

ben: we could...

danbri: ah, i remember now, w3c requires advance notice of the meeting

ben: ok
... please do circulate some results/outcomes of meeting

jjc: are iptc hosting?

em: Misha/Reuters are hosting, or arranging hosting

(fleet street?)

<scribe> ACTION: ericm or danbri circulate results of london f2f meeting at IPTC (assuming it happens) [recorded in http://www.w3.org/2005/06/28-swbp-minutes.html#action01]

ben: 2 main issues?

mark: iptc need to decide whether to use xhtml2 or their own language

(I missed the sketch of the 2 issues; scribe-help welcomed)

<benadida> 2 issues: (1) qnames in subjects and objects (2) attributes defaulting to RDF properties

jjc: presume we wouldn't consider it a failure if they had their own language, so long as was transformable with XSLT into XHTML2/RDFA

em: not sure i'd consider it a failure, but its not the target i'd hope for
... a fallback position
... might be reality of non-tech situation, re migration path etc
... but i hope to be able to bypass that middle step

jjc: personally i wouldn't be committed to that approach
... maybe best way to use is directly, or indirectly, ...

em: easier to make assumptions explicit in f2f setting
... treat them as an xhtml2 metadata customer
... help them model data in a highly flexibile way
... stumbling blocks i see are qname one, and modelling (ie. understanding their goals)

<em> hello?

<em> testing 1,2,3

<benadida> ping em

ben: good to contrib to the mail threads before meeting
... so meeting is a week fri, pls try to contrib mails before then
... good stuff
... some rels like next, prev, defining the rdfs

http://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/2005May/0005.html

[HTML] Draft RDF/XML description of XHTML 2.0 link types

http://www.w3.org/2005/05/hrel/linktypes.rdf

danbri: definition of 'Document' ok?
... shall i ref to TAG InformationResource

mark: they're nto always retrieivable, so emphasising they're abstract

danbri: propose using TAG InformationResource

mark: 2nd'ed

no objections heard

resolved: class Document definition should refer somehow TAG InformationResource

ben: copyright is 1 thing needing discussion

[[

<rdf:Property rdf:about="#copyright">

<rdfs:label>copyright</rdfs:label>

<rdfs:comment>Refers to a copyright statement for the document.</rdfs:comment>

<rdfs:range rdf:resource="#Document"/>

<rdfs:domain rdf:resource="#Document"/>

<rdf:type rdf:resource="http://www.w3.org/2002/07/owl#ObjectProperty"/>

<!-- aside: see extensive debate on Atom list re a similar construct; also

dc: rights, and Creative Commons. Scope for sub-property mappings here? -->

</rdf:Property>

]]

scribe: trying thing thru changes
... copyright, also add 'license'?
... how much stuff to add into this core vocab?

mark: things like cc:license... discussion of whether it goes in the cc ns

ben: we wantto move away from that

mark: if you remove the "cc:" prefix, you'd need a new property

jjc: although presumably this xhtml2:copyright is similar to

danbri/ben: dc:rights, dcterms:license are the closest dc terms

ben: re html, arg is that the 2 should be distinct
... then q of whether to subPropertyOf

jjc: my point is that a group like Dublin Core spend more time than the html wg on this sort of thing
... as much as we can suggest legacy migration

mark: could dan add equivalence statements?

jjc: yes, they'd be legacy migration statements
... they'd need endorsement by the html wg

danbri: not our job. html guys get to decide how much to clutter their ns.
... but they seem to have gone down the modularity, external extensions route

ben: ... additioanl terms are up to the html wg
... lets do the following
... dan, this is great, we need to move to a WG Note, and start formatting this in that format right now
... i'll notify the wg that we're working on this

danbri: i'll write an XHTML doc to go with this RDF/XML

mark: you've used # but the HTML WG's ns ends in / currently

<scribe> ACTION: danbri record an issue re # vs / [recorded in http://www.w3.org/2005/06/28-swbp-minutes.html#action02]

jjc: lets take this to SWBPD WG

mark: there's all the role values that need to go in here (header, footer, main content...)

<!--

issue-bookmark-title-etc

Excerpted text:

"The title attribute may be used, for example, to label the bookmark.

Note that several bookmarks may be defined for a document." ... what does this

mean in an RDF context?

-->

mark: sugar for rel=role href=...

ben: jjc how much time do you have for help w/ tests?
... incl xslt

jjc: doesn't sound too hard
... what's timescale that you want it done by?

mark: can i chip in?
... item re looking at hierachy
... didn't do it
... we have an action to do it...
... can we work together on it?
... looking for a good rdf example
... something everyone'll accept is a good test

jjc: my presentation at webconf on this TF
... had 4 examples in html and rdf
... tried to motivate the tf

<scribe> ACTION: jjc to circulate www conf paper [recorded in http://www.w3.org/2005/06/28-swbp-minutes.html#action03]

ben, mark you want to base inheritance rules on this? test them?

mark: testing. everytime you accom a usecase you risk losing another
... we have some stuff for anon nodes

danbri: propose division of labour. jjc focus on the xslt. danbri on examples.

mark: a set of examples, eg a foaf file, rss feed, 2 or 3 docs, ... would be ideal

<scribe> ACTION: danbri make examples [recorded in http://www.w3.org/2005/06/28-swbp-minutes.html#action04]

ben: do we have the inheritance rules written down yet

mark: i think jjc had the most substantive review comments on the rules, lets get some motivating examples first, to avoid wasting xslt hacking time if rules chang

e

jjc: the rules in the latest WD are simple, only 2 in (1) and (2) in metadata attribs module
... not inheritance rules as such

mark: notthing more written down yet, but more to come
... we agreed rdf/a separate doc needs to be brough t into line w/ current thinking
... make sure we happy with that
... we've lost some things, eg statements all about a common object

er subject

mark: only ambiguity i have left, in terms of inheritance, or rather 'chaining', ...object of first is subject of next

then we changed to another method

now different again, can do along subject or object(?)

can't have both, need to pick on e that matches common

danbri: expresssiveness vs limits to graphs?

jjc: xslt 2 is ok?

danbri: yeah for starters, as easier

[exit jeremy]

wiki http://esw.w3.org/topic/FrontPage

scribe: can use for examples?

<scribe> ACTION: ben to pull in action items from prev meetings [recorded in http://www.w3.org/2005/06/28-swbp-minutes.html#action05]

<scribe> ACTION: ben to coordinate with mark for next week's meeting [recorded in http://www.w3.org/2005/06/28-swbp-minutes.html#action06]

Summary of Action Items

[NEW] ACTION: ben to coordinate with mark for next week's meeting [recorded in http://www.w3.org/2005/06/28-swbp-minutes.html#action06]
[NEW] ACTION: ben to pull in action items from prev meetings [recorded in http://www.w3.org/2005/06/28-swbp-minutes.html#action05]
[NEW] ACTION: danbri make examples [recorded in http://www.w3.org/2005/06/28-swbp-minutes.html#action04]
[NEW] ACTION: danbri record an issue re # vs / [recorded in http://www.w3.org/2005/06/28-swbp-minutes.html#action02]
[NEW] ACTION: ericm or danbri circulate results of london f2f meeting at IPTC (assuming it happens) [recorded in http://www.w3.org/2005/06/28-swbp-minutes.html#action01]
[NEW] ACTION: jjc to circulate www conf paper [recorded in http://www.w3.org/2005/06/28-swbp-minutes.html#action03]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.126 (CVS log)
$Date: 2005/06/28 15:26:12 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.126  of Date: 2005/05/16 16:49:48  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

No ScribeNick specified.  Guessing ScribeNick: danbri_scribe
Inferring Scribes: danbri_scribe
Default Present: Ben_Adida, MarkB_, Danbri, Emiller, jjc
Present: Ben_Adida MarkB_ Danbri Emiller jjc
Regrets: Steven Ralph
Got date from IRC log name: 28 Jun 2005
Guessing minutes URL: http://www.w3.org/2005/06/28-swbp-minutes.html
People with action items: ben danbri ericm jjc

[End of scribe.perl diagnostic output]