RDF-in-XHTML Task Force -- 29 Jan 2007

<RalphS> Previous: 2007-01-08 http://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/2007Jan/0009.html

<RalphS> F2F minutes: http://www.w3.org/2007/01/23-swd-minutes#item01

FTF Results

http://www.w3.org/2007/01/23-swd-minutes#item01

Ben: We agreed to potentially take all three of our docs to Rec track

Michael: Please remind people reviewing doc to CC our mailing list

Ben: Agreed

XHTML 1.1 Module

Ben: April 1 is the decision day for deciding if this module should go to Rec
... let's in any case not be bogged down by the problems around HTML/XHTML

Mark: The module is fine
... I've been doing work on it, been working with Shane a lot
... I've been updating it to match the new Modularization schema's (v1.1)
... but it seems like I have used some incorrect schemas
... so Shane is trying to sort it out
... it's not a lot of work, more coordination
... I'll try to do it Wednesday, during the XHTML call

<RalphS> Steven: there are two "1.1"s:

<RalphS> ... Modularization 1.1 and XHTML 1.1

<RalphS> ... both will be rev'd; Modularization first, then XHTML

<RalphS> ... Modularization 1.1 is about to go to Proposed Rec; it's completed Last Call, needs no CR

<RalphS> ... Modularization 1.1 includes schemas

<RalphS> XHTML Modularization 1.1 Last Call WD

<RalphS> Steven: XHTML Modularization 1.0 only does DTDs, so it's not of much use to RDFa

Mark: There is nothing to stop us doing a draft of RDFa anyway, but it is easier if they walk in step
... but there's no timing problem

Ben: Can we make mid-end February?

Mark: Should be fine

Ben: Can it be put up somewhere just so we can look at it?

Mark: Yes, I can put it in our source control; it's just some schemas

Ben: Excellent

<scribe> ACTION: Mark to put M12N schema up somewhere, and tell us where [recorded in http://www.w3.org/2007/01/29-htmltf-minutes.html#action01]

<mhausenblas> http://esw.w3.org/topic/RDFa

Michael: One other point
... we need to clarify the FAQ
... particularly the one about using RDFa with any version of HTML

Ben: We've answered that question many times

<RalphS> Is Michael talking about 2.1.1. Can I use RDFa on any HTML version?

<RalphS> Steven: differentiate _valid_ HTML and _conformant_ HTML

<RalphS> ... since the HTML rules say to ignore elements and attributes you don't recognize

<RalphS> ... validation is a question

<RalphS> ... it's important to distinguish between _valid_ and _conformant; when people use new things it's important that they know this is not an error

<RalphS> Ben: people also want to know how the new things will affect browser rendering

Michael: This point comes up regularly when people compare RDFa with Microformats

<MarkB_> Also should incorporate the 'gentle slope' idea: http://rdfa.info/2006/06/08/rdfa-the-gentle-road-to-rdf/

<RalphS> Ben: but there is no W3C branded validator that will accept new elements and attributes

Action items

<RalphS> ACTION: Ben to update uc2, image src, uc4, no innerHTML, add link to GetN3 bookmarklet [recorded in http://www.w3.org/2007/01/08-htmltf-minutes.html#action15] [DONE]

<RalphS> ACTION: Michael continue work on the FAQ template that Elias started [recorded in http://www.w3.org/2007/01/08-htmltf-minutes.html#action10] [CONTINUES]

<RalphS> Michael: I did some work, would like to invite others to contribute

<RalphS> ... ok to leave "ToDo's" for me

<RalphS> ACTION: Michael transfer issue data to SWD tracker [recorded in http://www.w3.org/2007/01/08-htmltf-minutes.html#action03] [CONTINUES]

<RalphS> Michael: the mail generated by tracker currently goes [only] to the SWD WG list

Ralph: Two things came out of the FtF: 1) That SWD are aware and feel ownership of issues
... I will look into tracker sending mails to both lists

<RalphS> SWD issues list

Ralph: I think when the new XHTML WG is underway, we should reevaluate the division of mail to WG lists

<RalphS> ACTION: Ben add a forwarding pointer to http://www.w3.org/2001/sw/BestPractices/HTML/2005-current-issues when the new wiki issues list page is created [recorded in http://www.w3.org/2006/12/18-htmltf-minutes.html#action04] [CONTINUES] [DONE]

<RalphS> ACTION: Ben to look into Science Commons use case [recorded in http://www.w3.org/2006/12/11-htmltf-minutes.html#action04] [CONTINUES]

<RalphS> ACTION: Elias to send email to list with use case from IBM [recorded in http://www.w3.org/2006/12/04-htmltf-minutes.html#action10] [CONTINUES]

Ben: Not sure when Elias is back

<RalphS> ACTION: Mark produce RDFa XHTML 1.1 module and spec [recorded in http://www.w3.org/2006/12/18-htmltf-minutes.html#action06] [CONTINUES]

<RalphS> ACTION: Mark produce more examples of applicability of n-ary relations from IPTC documents [recorded in http://www.w3.org/2006/12/04-htmltf-minutes.html#action08] [CONTINUES]

<RalphS> Ben: SWD WG resolved that RDFa is not required to implement all of RDF/XML

<RalphS> ACTION: Mark write examples/tests of striping support [recorded in http://www.w3.org/2006/12/04-htmltf-minutes.html#action07]

Ben: if there's no use case for it

<RalphS> -- withdrawn

<RalphS> ACTION: Steven to put together sample XHTML2 doc with all mime type, etc.. [recorded in http://www.w3.org/2006/12/04-htmltf-minutes.html#action09]

Ben: Someone volunteered to help with generation of test cases

<RalphS> -- continues

<benadida> "someone" is Wing from IBM

<RalphS> ACTION: Ben start a list of RDF/XML features that are not supported by RDFa [recorded in http://www.w3.org/2007/01/23-swd-minutes.html#action01] [CONTINUES]

<RalphS> ACTION: Ben to get the docs in good shape for next week [recorded in http://www.w3.org/2007/01/23-swd-minutes.html#action04] [CONTINUES]

<RalphS> ACTION: Ben to update issues list with the @CLASS overload problem [recorded in http://www.w3.org/2007/01/23-swd-minutes.html#action08] [CONTINUES]

<RalphS> ACTION: Ben to update RDFa schedule on wiki to aim for last call on June 1 [recorded in http://www.w3.org/2007/01/23-swd-minutes.html#action09] [CONTINUES]

<RalphS> ACTION: Ben to write down the relation between GRDDL and RDFa [recorded in http://www.w3.org/2007/01/23-swd-minutes.html#action02] [CONTINUES]

Ben: On the docs, I will update this week

FtF Slight return

Ralph: Monday evening (this is point 2 from above) we had an informal get together. Guus did a demo of his prize-winning museum system
... Tim asked if the HTML pages contained RDFa, and Guus said "of course they do"

Ralph: Ben went into a corner and tried it with his tools, and then stepped up and demoed cutting and pasting from the pages
... and got a round of applause!

<benadida> RDFa clipboard: http://labs.adida.net/rdfa-clipboard/

Cambridge SemWeb get-together demo

<benadida> http://e-culture.multimedian.nl/demo/local_view?uri=http%3a%2f%2ft-d-b.org%3fhttp%3a%2f%2fe-culture.multimedian.nl%2fdemo%2fartchive%2ftansey%2ftansey_picasso.jpg

<MarkB_> more applause!!

<benadida> http://e-culture.multimedian.nl/demo/local_view?uri=http%3a%2f%2ft-d-b.org%3fhttp%3a%2f%2fe-culture.multimedian.nl%2fdemo%2fartchive%2fgris%2fpicasso.jpg

<RalphS> Ben: only tested in FireFox

Steven: Doesn't yet work in Opera

Ben: Maybe submit to Devtrack at WWW2007

<RalphS> Ben: click 'copy', click the scissors item, the use the normal OS 'copy' item

<RalphS> ... then go to the scissors on rdfa-clipboard, click there, and 'paste'

<mhausenblas> I am overwhelmed

Ben: I need to write this up
... and put it somewhere safe so that my machine doesn't get overwhelmed

<RalphS> Ralph: it would be fine to put the bookmarklets in SWD WG space on www.w3.org

Steven: There is an issue of an icon for RDFa, just while I remember it

Overloading @CLASS

<MarkB_> more applause ;)

<RalphS> Steven: word reached me that TimBL is unhappy about RDFa using @CLASS

Ben: Yes, your warnings were right
... he added RDFa to a page with lots of classes

<MarkB_> prescient ;)

Ben: and tons of spurious triples were generated

<benadida> ...

Ben: this says that a person is of type 'notice'
... based on that we talked about two possible fixes
... 1) not @class but @type
... which I don't like
... 2) Only consider classes that have a prefix
... which I like more

<mhausenblas> what is about class=":Person" then ?

Ben: there is also a similar ul/ol problem

Ralph: Unfortunately we didn't make a copy of the whiteboard
... Tim was distinguishing between real DOM nodes and 'virtual' RDF nodes
... and trying to understand which crossed over between the two
... so there were two questions, one the many triples problem, and Ben tried to argue that the triples were really there
... but the resolution was that they may not be incorrect, but they are unexpected, and therefore noise
... The other question was in the context of lists (I hope Ben will write this up) that lists will only generate triples when there is an explicit subject

Ben: uls/ols must have an explicit subject, not bnodes linking to bnodes
... this is a garbage collection problem

Ralph: Tim added a little bit of RDFa to a page, and ran it through the bookmarklet, and was looking for what he expected, and it was there, but hard to find

Mark: I'm wary about changing the syntax to cope with these things
... there is a difference between authoring and consuming the triples
... and having extra rules to prevent triples needs to be solved at the consuming level. I'm worried we will be blocking potential use cases
... If I use rel='stylesheet', that is a real triple, even if the author hadn't intended it
... we must distinguish between authoring and consuming
... and not block potentially useful information
... I don't like the concept of 'reducing noise'.

Ben: The only syntax change proposed was related to @class

<RalphS> Mark: we may want to add switches to select whether unprefixed class values generate triples

<Steven_> Scribenick: Steven_

<Zakim> RalphS, you wanted to note that the only syntax change that came out of the breakout concerns @class

Mark: The key thing is that some people use class as a presentation thing, and some as semantic

<mhausenblas> May I comment, please ...

Ralph: It is consistent to say that the triples using @class are really there, but many people don't mean them as such
... so do all semantics present in an HTML doc deserve to be output as triples?

<MarkB_> But my point is that even talking of 'being output' is wrong. It's a 'pull' not a 'push' approach.

Ralph: there's also the issue of a node in the DOM tree representing a person

Mark: we need to resurrect that issue

<MarkB_> In brief, there is no way to refer to a DOM node. :)

Michael: I support Mark. The triples are really there, whether the author thought of them as such or not
... the triples are for machines, not humans

Mark: So we are saying the opposite of microformats? :-) Put the machine first :-)

Ralph: We want to give a uniform semantics to HTML attributes, and we think that styling semantics are not against regular semantics
... microformats does this after all.
... But Tim says that microformats does not have consistent semantics for class
... you have to interpret it in context
... so it is different there, and that is one of the problems with microformats.

Mark: There are people who use class properly
... But great work guys in getting Tim's attention, and writing RDFa

Wrap up

Ben: Michael, I wonder if you would coordinate the creation of test cases?

Michael: Sure!

Ben: I will champion the use case doc, and primer
... Steven would you be willing to help with the syntax doc?

Steven: Sure

Ben: Next week we will attack some urgent issues

<scribe> ACTION: Ben to put Michael and Wing in contact for test case production [recorded in http://www.w3.org/2007/01/29-htmltf-minutes.html#action17]

Ralph: Is the Primer ready for tomorrow?

Ben: That's the goal

<RalphS> next meeting: 5 Feb

Ben: next meeting same place, same time

[Adjourn]

RDF-in-XHTML Task Force

29 Jan 2007

Attendees

Contents