See also: IRC log
<trackbot> Date: 07 July 2011
easy test -- mute one, speak ... see which is quieted
<scor> manu: not set yet
<manu> scribe: Ted
<manu> scribenick: MacTed
<manu> Introduction: Niklas Lindstrom
lindstream: consultant in Sweden,
working with RDF for 6+ years, semantics and fidelity of
expression are prime interests
... coming back to RDFa from API perspective, but now
considering Microdata/mapping concerns
<manu> http://www.w3.org/2011/06/30-html-wg-minutes.html#item09
manu: HTML WG has discussed what to do about TAG note; basic response is "need information"
Steven_: all the more reason why we should send a formal objection
manu: better that the TAG does
that
... HTML WG has basically asked TAG to make their concerns more
clear and formal, before they will act
Steven_: if TAG does so, that's fine. if not, then we should step up.
manu: that's the path we'll take
<manu> http://lists.w3.org/Archives/Public/public-rdfa-wg/2011Jul/0007.html
manu: ...reviewing official
position email draft, position by position...
... "CONCERN: Multiple specifications for the same task" --
effectively, same concern as TAG has expressed.
... not "which spec should exist?" but "should there be two
specs?"
gkellogg: 2 key questions --
should there be 2 markups, and what is range of data to be
marked up?
... i.e., what is the purpose of metadata markup in HTML?
presumably to generate data compatible with other W3C specs,
e.g., RDF
(discussion)
<manu> MacTed: This is a range issue - what use cases are supported?
lindstream: apart from use cases
(what), microdata raises concern of *how*
... how much do you expose to an expert as opposed to a
newbie?
manu: is the important thing, how much complexity do new people have to deal with straight-away?
lindstream: that's the gist. e.g., if something looks like a date, is that enough, or does it have to be specifically typed, etc?
<scor> bear in mind that the reason why md was created is because RDFa was covering too many use cases and was overkill for the basic web developers... the TF should focus on identifying practical use cases which RDFa covers (and not no md) which make sense for regular web developers (the audience of md)
manu: `CONCERN: Consensus on "No Change"` -- TAG must know that this really isn't an option
lindstream: it's very bad to have two specs for the same thing...
manu: `CONCERN: Key implementers
will choose to not be involved.` XHTML and XForms both had task
forces to figure out best way forward; neither was successful
because key players didn't play
... advice is, make sure the list we've sent gets
involved
... `CONCERN: Agreement and then non-action` -- everybody
agrees on plan, but nobody acts on agreement...
... advice is to actively review commitments and report on
follow-through
... `CONCERN: Slow creation of Task Force` -- if it takes too
long to create TF, other WG (RDFa, HTML, others?) timelines are
jeopardized
... advice is to prioritize this TF effort
lindstream: conflicting specs have happened before. this is opportunity for W3C to show they can handle such conflicts...
Steven_: does TAG think that creating this TF is their job?
manu: nobody wants the job...
everybody's waiting to see who steps forward
... we'll nudge TAG to accept responsibility
... `CONCERN: TAG Note is not actionable`
... advice: formalize objections & concerns, so that
something actually comes of this
<scribe> ACTION: manu to revise email to TAG based on today's discussion, with "last comments" due ASAP mailing list [recorded in http://www.w3.org/2011/07/07-rdfa-minutes.html#action01]
<trackbot> Created ACTION-87 - Revise email to TAG based on today's discussion, with "last comments" due ASAP mailing list [on Manu Sporny - due 2011-07-14].
<manu> http://manu.sporny.org/rdfa/rdfa-core-simplified/diff-20110331.html
lindstream: Microdata seems to be
surface data with no fidelity.
... can quickly be mapped to triples, if you base off URI of
current document
... vocab and profile mechanisms could be used to add
fidelity
... somewhat akin to GRDDL
<manu> This is somewhat akin to the JSON-LD coercion approach: http://json-ld.org/spec/ED/20110615/#type-coercion
manu: points out similarity of JSON-LD coercion approach: http://json-ld.org/spec/ED/20110615/#type-coercion
lindstream: and also to
GLUON...
... this might also answer how to represent RDF connections in
RDFa
gkellogg: putting profile in RDFa Advanced seems to make this a less useful path forward...?
<Zakim> manu, you wanted to note browser manufacturers don't like this.
manu: browser vendors are
expressing preference for non-RDFa APIs, because they think
RDFa API makes caching and other performance issues harder to
address
... trouble with profile is that browser has to stop
processing, retrieve profile doc, and then resume
processing
... browser vendors just don't like profile...
lindstream: put profile in Basic, but allow basic API processing *without* profile application (and thus low fidelity). Advanced brings profile processing (and high fidelity)...
manu: should be taken seriously,
as a key browser implementer...
... maybe we should sit on WHAT WG, mozilla, other channels,
and pester them about API thoughts
... we do want a path that makes it easy for browser vendors to
implement
... "easy to implement in Javascript doesn't mean easy to
implement in a performant way"
lindstream: performance concerns seem addressable by low-fidelity microdata-ish RDFa Basic.
<manu> MacTed: It seems that the browser vendors are mostly concerned about human interaction - most of what we're takling about is machine interaction.
<manu> MacTed: Basic to Advanced progression makes a great deal of sense to me - initial pass - low fidelity. Once there is something there for human to deal w/ then we do more advanced / less performant mechanism.
lindstream: interesting consideration is "end users" vs "re-users" even more than "human" vs "machine"
<scor> manu: we should not be pestering them
<scor> manu1: ^^
manu: anyone willing to sit on WHAT-WG channel and ask questions?
lindstream: will be polling relevant existing contacts...
<scor> +1
manu: right, this is not about
pestering, but about checking for concerns with APIs, and
figuring out how best to address...
... correcting misundersatndings, etc.
trackbot, end call
<trackbot> Sorry, MacTed, I don't understand 'trackbot, end call'. Please refer to http://www.w3.org/2005/06/tracker/irc for help
trackbot, end conference
This is scribe.perl Revision: 1.136 of Date: 2011/05/12 12:01:43 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/quited/quieted/ Found Scribe: Ted Found ScribeNick: MacTed Default Present: scor, manu, MacTed, Steven, gkellogg, lindstream, ShaneM Present: scor manu MacTed Steven gkellogg lindstream ShaneM Agenda: http://lists.w3.org/Archives/Public/public-rdfa-wg/2011Jul/0006.html Found Date: 07 Jul 2011 Guessing minutes URL: http://www.w3.org/2011/07/07-rdfa-minutes.html People with action items: manu[End of scribe.perl diagnostic output]