W3C

SWBPD RDF-in-XHTML TF

6 Feb 2006

Agenda

See also: IRC log

Attendees

Present
Ben Adida, Ralph Swick, Steven Pemberton, Mark Birbeck
Regrets
Jeremy Carroll
Chair
Ben
Scribe
Ralph
Previous
2006-01-30

Contents


integration of RDF/A syntax into XHTML specs

Steven: integration is proceeding. We hope it will be done by the end of this week.

Ralph: will this TF be given a URI of an editor's draft for us to glance at?

Steven: yes, absolutely
... to speed up the process we're inclined to say now "comment on the Last Call draft"
... as if we issue another WD now it will delay Last Call
... every time we reissue a draft we get lots of new comments that prevent us from exiting Last Call
... so we'd like to give the TF a final Last Call

Ralph: can we see a "final" editor's draft?

Steven: if you really, really want

Ralph: it would be reasonable to ask the TF to restrict its comments to "show stoppers"
... if integration finished this week, then next week's telecon is probably when you decide readiness for publishing

Steven: yes
... the integration will be done this week, though we might not be in last call next week.

Ralph: how does the XHTML WG share editors' drafts?

Steven: nightly build URL that I'll send along

Ralph: so we can look at the nightly build over the weekend

ACTION: once Steven sends editors' draft of XHTML2, all TF members take a look and comment on showstopper issues only [recorded in http://www.w3.org/2006/02/06-htmltf-minutes.html#action01]

Review of action items

ACTION: [CONTINUES] Jeremy followup on edge case [recorded in http://www.w3.org/2005/12/06-swbp-minutes#action03]

ACTION: [CONTINUES] Jeremy followup with Mark on the question of multiple triples from nested meta and add to issues list [recorded in http://www.w3.org/2005/12/06-swbp-minutes#action01]

ACTION: [CONTINUES] Jeremy propose wording on reification [recorded in http://www.w3.org/2005/12/06-swbp-minutes#action02]

ACTION: [CONTINUES] Ben to draft full response to Bjoern's 2004 email [recorded in http://www.w3.org/2006/01/24-swbp-minutes.html#action03]

Ben: this is mostly done, just need to finish the bit on collections

ACTION: [CONTINUES] Ben start separate mail threads on remaining discussion topics [recorded in http://www.w3.org/2005/12/06-swbp-minutes#action04]

Ben: a couple of issue threads left
... but consider yourselves reprimanded for not responding to the existing threads

ACTION: [DONE] Ben add lack-of-consensus notes to the RDF/A Primer [recorded in http://www.w3.org/2006/01/30-swbp-minutes.html#action12]

Issue 9; <link href='...'>

Ben: see "Discussing Outsanding Issues - Issue #9"

Steven: so the proposal is that <link> should be equivalent to <a>?
... another option is to say that only <a> is intended to be clickable

Ben: note that <link> has the special semantics of 'refers to parent element only'

Steven: <link> currently has the property that the links are often accessible in the browser
... e.g. <link rel='next'> shows up in a menu
... what's new in XHTML2 now is that <link> is permitted in body

Ben: is content permitted in <link>? if so, what does it mean?

Steven: yes, content is permitted. It's a question whether <link> is display:none

Ben: it's a decision for the HTML WG to say whether link is display:none by default and can be overridden by stylesheets. Would it be the case that the value of display can be overridden in a stylesheet?

Steven: the attribute that makes <link> clickable is not specifiable in a stylesheet
... since href is now permitted everywhere, this raises the question of whether every element now becomes clickable
... the HTML WG's current feeling is that the href is accessible to the user in some way but not necessarily rendered in the same way as <a>

Ben: this could be an issue for users if href has both RDF semantics and other semantics

Ralph: could be a user interoperability problem; href seems likely to get some presentation semantics which might differ between browsers

Ben: is the HTML WG leaning towards making href on other elements behave differently than on <a>?

Steven: I think that <meta> and <link> are display:none
... but the question still arises on what happens of the stylesheet does make it displayable
... my feeling is that if <link> or <meta> is made visible it should not appear as a primary clickable link

Mark: as soon as we say that href is a left-clickable link everywhere this makes things consistent
... then if we say these elements are display:none by default [we are compatible with older browsers]
... right-button on a <blockquote>, for example, would be a nice way to show a menu of all links from this quote
... not unlike what browsers do to create a menu bar for <link rel='next'>

Steven: the question is where to draw the line between the XHTML specification and user agent behaviour
... whether [user agent behaviour] should be made normative or not

Ben: the question of default value of display really only applies when the <link> or <meta> is not empty

Mark: I suggest we say that the link content is an rdfs:label that applies to the subject; with inline text [content] a clever browser can use the rdfs:label in its menu

Ben: <li href=''> is clickable, right?

Mark: yes, and with the primary means of navigation

Ben: so the question is whether <link> and <meta> are consistent with this

Mark: I don't think it would be right to make a special case for this except by adding display:none

Ben: so we agree here that <link> and <meta> are clickable but the HTML WG may decide to add display:none

Steven: so the suggestion is that any element with href is clickable [in the primary method] when it is visible?

Ben: yes

RESOLVED: This Task Force concurs that href on any element that is visible makes that element clickable using the primary user agent mechanism but that <link> and <meta> may be display:none by default at the HTML WG's choice

RSS in RDF/A

Ben:> see "RSS in RDF/A, for discussion at Telecon"
plus a correction, noted by Ian Davis:

<benadida> <ul id="#items"

<benadida> should be

<benadida> <ul id="items" about="#items">

Ralph: please add the namespace declaration

Ben: the primary question here is about the rdf:Seq

(the WiKi link is on the -> HTML TF home page)

-> LanguageRdfAExampleRSS

Ben: I wrote 0063 without using any syntactic sugar; i.e. rel="ref:_1"
... can we use rdf:li ?

Mark: unless you use [XML] Schema, you don't get the html <li> semantics if you use rdf:li

Ben: can we use rdf:li to avoid needing _1, _2, etc. Having to use XML Schema is fine.

-> 2.15 Container Membership Property Elements: rdf:li and rdf:_n

Mark: alternatively, we could say that html:li has rdf:li semantics. Hasn't rdf:Seq been replaced by lists?.

Ralph: rdf:Seq is not deprecated, but lists have important semantics that indicate that the list is complete

Mark: rdf:Seq has semantics of saying _ordered_ lists
... in your 0063 example you've used <ul> -- unordered list
... we could use the HTML semantics to advantage; create rdf:Seq or rdf:Bag

Ben: the HTML semantics imply the list is complete, don't they?

Ralph: not clear

Mark: do we expect anything to make use of the ordered/unordered semantics?

Ralph: I would think so, since HTML has those semantics

Ben: I'm in favor of looking at ordered vs. unordered semantics

Ralph: so an important question from the HTML point of view is whether authors using <ol> and <ul> intend to say "... and this is the complete list"?

Steven: given the start attribute, I would not want to make a commitment to what authors intend

ACTION: Ben write out a proposal for how OL and UL turn into rdf:Seq and rdf:Bag [recorded in http://www.w3.org/2006/02/06-htmltf-minutes.html#action08]

Mark: there's a new type list in XHTML 2; an <nl> -- Navigation List
... so <ol> might be an rdf:Seq and every other type of list is rdf:Bag

Ralph: perhaps <nl> is like rdf:Alt, since you choose one navigation path :)
... but I don't think we really want to overload things that much

Section 2 wording

-> see "augmenting section 2 of primer"

Ralph: it's clear that Section 2 does not follow the TAG's instructions
... what is less clear is whether the ambiguity is critical
... how about "do not follow" rather than "incorrect"

Mark: I don't mind "incorrect" if they really don't follow the TAG's instructions

Ralph: propose to add the sentence "pecifically, the

examples in this section use the same URI to refer to both a Person

(or a group of people) and a document.

"

Ben: OK with me

Mark: OK with me
... there are two levels at which the URI confusion is happening; one at the foaf level and the other at the URI level

Ralph: it feels a mistake to me to omit the subject of the triples, particularly in section 2.2.3
... as this is explicitly the nature of this URI debate
... let's not sweep this under the rug as we do know what the current RDF/A syntax spec says

Mark: that's fine with me

Ralph: we can note that there is a philosophical debate going on but let's not dodge that; the Primer should show what the Syntax spec says must be generated

RESOLVED: we agree that the rest of the Section 2 triples can be added before publication

ACTION: Ben update the editor's draft to ass to section 2 [recorded in http://www.w3.org/2006/02/06-htmltf-minutes.html#action09]

Summary of Action Items

[NEW] ACTION: Ben update the editor's draft to add to section 2 [recorded in http://www.w3.org/2006/02/06-htmltf-minutes.html#action09]
[NEW] ACTION: Ben write out a proposal for how OL and UL turn into rdf:Seq and rdf:Bag [recorded in http://www.w3.org/2006/02/06-htmltf-minutes.html#action08]
[NEW] ACTION: once Steven sends editors' draft of XHTML2, all TF members take a look and comment on showstopper issues only [recorded in http://www.w3.org/2006/02/06-htmltf-minutes.html#action01]
 
[PENDING] ACTION: Ben start separate mail threads on remaining discussion topics [recorded in http://www.w3.org/2005/12/06-swbp-minutes#action04]
[PENDING] ACTION: Ben to draft full response to Bjoern's 2004 email [recorded in http://www.w3.org/2006/01/24-swbp-minutes.html#action03]
[PENDING] ACTION: Jeremy followup on edge case [recorded in http://www.w3.org/2005/12/06-swbp-minutes#action03]
[PENDING] ACTION: Jeremy followup with Mark on the question of multiple triples from nested meta and add to issues list [recorded in http://www.w3.org/2005/12/06-swbp-minutes#action01]
[PENDING] ACTION: Jeremy propose wording on reification [recorded in http://www.w3.org/2005/12/06-swbp-minutes#action02]
 
[DONE] ACTION: Ben add lack-of-consensus notes to the RDF/A Primer [recorded in http://www.w3.org/2006/01/30-swbp-minutes.html#action12]
 
[End of minutes]

$Date: 2006/02/06 21:12:45 $