See also: IRC log, previous 2006-10-09
ACTION: [DONE] All to send Ben email giving times you are available for TF telecons [recorded in http://www.w3.org/2006/10/09-htmltf-minutes.html#action01]
ACTION: [DONE] Ben add striping support to bookmarklet to test new bnode support proposal [recorded in http://www.w3.org/2006/10/09-htmltf-minutes.html#action08]
ACTION: [DONE] Ben move his live editor's draft to http://www.w3.org/2006/07/SWD/RDFa/syntax/Overview.xml [recorded in http://www.w3.org/2006/10/09-htmltf-minutes.html#action10]
ACTION: Ben announce new URIs for editors' drafts to TF [recorded in http://www.w3.org/2006/10/16-htmltf-minutes.html#action04]
ACTION: [DONE] Ben to draft message to the community describing the class proposal and explaining why we think it doesn't break existing uses [recorded in http://www.w3.org/2006/10/09-htmltf-minutes.html#action09]
ACTION: Steven to put together sample XHTML2 doc with all mime type, etc.. [recorded in http://www.w3.org/2006/09/19-htmltf-minutes.html#action01] [IN PROGRESS]
<Steven> http://www.w3.org/MarkUp/2006/xhtml2.xml
Steven: a little more progress
Ben: wasn't Mark going to produce some more tests for the striping proposal?
ACTION: Mark write examples/tests of striping support [recorded in http://www.w3.org/2006/10/16-htmltf-minutes.html#action10]
ACTION: Ben start separate mail threads on remaining discussion topics [recorded in http://www.w3.org/2006/09/19-htmltf-minutes.html#action07] [CONTINUES]
ACTION: Ben update the issues list [recorded in http://www.w3.org/2006/09/19-htmltf-minutes.html#action08] [CONTINUES]
ACTION: Elias start an FAQ [recorded in http://www.w3.org/2006/10/03-htmltf-minutes.html#action10] [CONTINUES]
Ben: implemented in latest bookmarklets
... implements @rel= w/o @href=
Ben: current element with @rel becomes the
subject for the contained statements
... there's an edge case if you put @about= and @rel on the same element w/o
an @href=
... you currently end up with subject and object being the same node
Steven: I haven't come across any counter-examples yet where this doesn't work
Ben: we should decide soon whether to include this in the next document drafts
Ben: issues
list was confused; I had written there that we were "thinking" of not
including reification
... but we did not resolve this question
Mark: my initial examples were the IPTC work
where they wanted to be able to say _who_ wrote something
... more recently, the i18n case came up
... link @rel=x @href=y @title= requires putting the title in a child
element
... the title property must be a property of the link, not of the document
... in RDF terms this seems to be a title of an RDF statement
Elias: reification is a double-headed
monster
... two ways to interpret reification: 1. as quoting, 2. as identification;
"this is the statement"
... I think the intention was more the former than the latter
... there are long discussions on this question
... it's not clearly a standard that you can name a statement and then start
talking about that statement itself
Mark: I'd like to use the named graph approaches that are being discussed
Elias: SPARQL uses named graphs though they're not yet accepted
Mark: we want the name to apply to whatever the
RDF community decides is meant
... we want RDFa to specify this at a higher level
... e.g. for IPTC we want the creator statement to say who is responsible for
making the statement
Elias: I can't justify making reification a requirement when its specification is very unclear
<EliasT> .. http://lists.w3.org/Archives/Public/semantic-web/2006Mar/0159.html
Elias: there is not a clear definition so
people are reluctant to use reification
... I'm not in favor of making this a requirement
Ben: I've heard similar feedback from the
community as Elias
... that reification is poorly specified
Mark: I think we should leave this issue open
and come back to it
... RDFa is different from RDF/XML in that we _do_ have an element that
represents the triples
... we have an element that is the object that carries the statement
... RDF/XML does not have a node that carries the triples
... since, for example, we do have a LINK element maybe we can make
statements about the LINK element
Elias: yes, we have a better scenario to make
use of this but it's still only a possibility
... I do not use reification because of its cloudy definition
Mark: what if we call it something different?
Elias: we want to talk about who made the
statement; e.g. provenance
... we'd have the capability to add provenance to any statement on an HTML
page?
Ben: right
Mark: I've often thought this could be harmonized; "who added these nodes to this document" and "who added this metadata to this document?"
Elias: couldn't we give a subject to the LINK
and add a predicate directly to that subject?
... if what we're pointing to is the element that contains the statement
...
... but an element can contain more than one statement
Mark: but we still have a problem; what does it
mean to add @title to a LINK? What is the subject of the title property?
... it's not a property of the document or of the target @href, so title must
be a property of something else
<benadida> Mark is looking at: http://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/2006Oct/0030.html
Ben: see <link rel="meta" title="ICRA" href="OurICRAPolicy.rdf" />
Mark: consider a blockquote with a child
<meta property='cite'>
... we're all happy that this is a property of the blockquote
... can also be specified equivalently as <blockquote cite='x'>
... if we say the @title= example works the same way, then the title property
must be a property of the statement
Elias: are we trying to name a conceptual statement made by the link element or the link element itself?
Mark: @rel=bookmark is used by Opera and an
@title= is used as a kind of label; the text to show to the user
... <meta property='title' ...> is then a fantastic technique
Elias: what if the link has @rel=, @rev=, ... -- does the @title apply to all of the statements?
<MarkB_> <> xh:meta <OurCrapPolicy.rdf> .
<MarkB_> _:link0 xh:title "ICRA" .
Mark: the problem is that these two statements aren't connected
Ralph: it's possible that we coud assert that the document says <> xh:link _:link0
Elias: yes, if you want provenance you have to
add all this yourself
... it's part of RDF to be able to express any relationships you want
Mark: we're saying that there are already existing HTML features and we want to specify what they mean [in RDF]
Elias: but we can create specific subjects on
the fly and attach the properties to these without having to use
reification
... I'm wary of pursuing RDF reification
... maybe meta is special
... @rel, @ref, @href are special but properties like @title are properties
of the link
... attributes on the element are _attributes_ of the element
... but we have another axis where both subject and object can be external to
the document
<MarkB_> <> xh:link _:link0
Elias: the element itself is the concrete thing we have to which we can attach the title property
<MarkB_> _:link0 xh:rel xh:meta .
<MarkB_> _:link0 xh:title "ICRA" .
<MarkB_> _:link0 xh : subject <> .
<MarkB_> _:link0 xh : href "OurCrapPolicy.href" .
Mark: the point of my example is that this
really is reification
... so we've just reinvented reification
<EliasT> ... http://www.w3.org/TR/swbp-n-aryRelations/
Elias: it's not bad to reinvent reification
... but the specific RDFS vocabulary [subject, predicate, object] is not
well-defined
... n-aryRelations uses a technique
... RSS wants to say several things about 'content'; the content relation has
several values
... so RSS uses intermediate bnodes
ACTION: Elias summarize the reification discussion [recorded in http://www.w3.org/2006/10/16-htmltf-minutes.html#action11]
<EliasT> .. http://ioctl.org/rdf/usementionmyarse
Mark: the n-aryRelation document has a note toward the end about reification
Ben: I'm willing to work on both primer and syntax documents this week
<Zakim> RalphS, you wanted to note Web Content Labels as another use case
Ralph: Elias is quite right that this is a long
and complicated discussion
... the Content Labels
Incubator Group has the same use case: making statements like "who made
this statement"
... Elias's summary will be very useful
... interested in the POV from implementors and theorists.
... at some point, the practice needs to take precedence
Ralph: I think it's complicated from a theoretical point of view but the application developers know exactly what they want the apps to do and we ought to be able to specify that
Ben: let's plan to talk about RDF containers next week
[adjourned]