See also: IRC log, previous 2006-05-15
Steven: I've given 2 talks in the past two
months including RDFa
... one to a Dutch HCI conference about future interfaces
... I talked about how RDFa can improve the user experience by permitting the
browser to do more
... then gave a modified version of the same talk to a Web 2.0 conference
... was less technical; I talked about how RDFa could mean less work for
people producing Web sites
... also, I was interviewed on radio and for print magazines; should be
published in the next month
... a business-oriented Dutch IT newsletter
... I have a link to the radio interview on my home page
... business people have been very receptive to the whole story
Elias: I will be focussing on RDFa for IBM
... My name is pronounced like "eh-LEE-us"
... Wing Yung (IBM) just added RDFa support on top of jena; may have a demo
later
Ben: as we're in limbo between the two WGs let's focus on the implementations for now
Ralph: Guus and Tom Baker are out during July/August, so SWD starts in September
Mark: the tutorial document has some things that do not quite match the spec
Ralph: -> SemWeb Deployment WG charter
Ben: there is an aggressive schedule in the SWD
charter
... there are some hard issues raised on the mailing list
... but we should not try to resolve them until [SWD formally starts in]
September
... however, it's ok to fix things in editor's drafts
Mark: it's just that people trying to implement
RDFa have noted conflicting details between the two documents
... things like the use of the role attribute
... the primer uses role in a certain way but the spec doesn't talk about
role
<benadida> Task Force mailing list
Ben: the most important thing we can do for the next 6 weeks is update our issues list
<EliasT> I sent a few bugs in the spec that don't require WG approval but need patching.
<EliasT> Here are my comments
Steven: there are some unanswered comments
... we should make sure we've addressed comments
Ben: Lee Feigenbaum also sent some spec comments
<EliasT> Lee's comments
ACTION: Ben update the issues list [recorded in http://www.w3.org/2006/07/18-HTMLTF-minutes.html#action01]
ACTION: Ben make sure RDFA bookmarklet runs locally [recorded in http://www.w3.org/2006/05/15-htmltf-minutes.html#action08] [CONTINUES]
Ben: I propose to withdraw this as it was mostly for the Edinburg demos
Ralph: is it not interesting outside of a demo context?
Ben: but it's only useful to run the bookmarklet locally if you're also running a local Web server, due to javascript security sandboxing
Mark: if a user chooses to configure it to run locally, would it be useful?
Ben: but that configuration would open a security hole for other javascript
Mark: I'm not sure there's no way to configure things to only support certain javascript
Elias: do what tabulator does?
<EliasT> http://esw.w3.org/topic/SparqlCalendarDemoUsage
Elias: signed applet code base principal support
Ben: bookmarklet does not the same one, but I'm
glad to see the changes
... if you want to give the bookmarklet any rights then the page you're
viewing when you activate the bookmarklet would inherit the same rights
... the attack paths are complicated but we should not recommend that people
open these vulnerabilities
Elias: though the name of the option 'signed.applets.codebase_principal_support' does imply signed applets, tabulator is not an applet and is using this option
Ben: it does appear that the tabulator approach
is much better than the previous hack
... our goal should be to help people try out RDFa
-- action continues
Mark: you'd also considered using greasemonkey
Ben: yep
ACTION: Ben start separate mail threads on remaining discussion topics [recorded in http://www.w3.org/2005/12/06-swbp-minutes#action04] [CONTINUES]
ACTION: Ben to draft full response to Bjoern's 2004 email [recorded in http://www.w3.org/2006/01/24-swbp-minutes.html#action03] [CONTINUES]
Ben: I got some great feedback from Steven on draft response
ACTION: Ben write a prototype hGRDDL profile for XHTML 1 [recorded in http://www.w3.org/2006/05/08-htmltf-minutes.html#action06] [CONTINUES]
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] [CONTINUES]
Steven: XHTML2 editors' draft is just about
ready.
... remaining issue is the new wording to describe role
... hope to have the draft ready before the next telecon
Ben: will this correct the spec mistakes?
Mark: I'll have to check -- the issue we've been discussing is the wording for 'role'
ACTION: Ben fix content negotiation on his current personal Web hosting site to make http://ben.adida.net/card (w/o ".html") work again [recorded in http://www.w3.org/2006/07/18-HTMLTF-minutes.html#action07]
move to next agendum
Ben: there were 3 talks about RDFa
... a W3C AC Meeting lightning talk
... I've posted my slides
... there was a lively discussion afterward
... there were two talks at WWW2006 conference; one from developer's point of
view, other just from SemWeb point of view
<EliasT> Ben's Presentations for WWW2006
<EliasT> rdfa presentations
Ben: there were microformats talks before each
one so I pointed out the additional benefits of RDFa, including scaling
... a developer from AOL building widgets looking for general ways to
describe data was interested in the features of RDFa over microformats
... some questions about XHTML modules; we don't have complete answers to
these yet
<EliasT> Kevin Lawver of AOL
Ben: sweetwiki is using RDFa; they gave me a bit of material to include in a slide
Mark: it's a good time to come in with a full solution as people are starting to look at microformats more closely
-> Paper>>SweetWiki-Semantic WEb Enabled Technologies in Wiki
Steven: the tone of the microformats
presentations were "we create microformats and you use them"
... in retrospect, about the CURIE debate, people really did not understand
what the issues were
... people really needed a tutorial about CURIEs before the debate
... and the actual debate unraveled over things that were not specific to
CURIEs
Ben: if we drop CURIEs we lose a lot of our bnode support, so it's not just the IPTC issue
Mark: I'm only proposing that in the XHTML2
spec we refer only to QName
... we can update later assuming CURIE is a superset of QName
Elias: it was good to focus on the differences
added by RDFa
... the philosophy of RDFa is what makes it important for us versus
microformats; scaling, etc.
... the problem is in the timliness of the presentations; people are saying
"but microformats are already here"
... this just shows the urgency of our work
Ben: my javascript implementation has been
moved to www.w3.org
... I will send a summary about this to the WG
... nothing much has changed since the end of May
... there are now dated versions
... I would really like to make this work in greasemonkey
<EliasT> RDFa service.
<EliasT> Now part of RDFLib
Elias: I wrote a sax-based parser to extract RDFa; the code was not complex
<EliasT> http://ben.adida.net/card.xhtml
Elias: then I deployed a service to do this extraction, dumping the output as triples
<EliasT> http://torrez.us/srevices/rdfa/[url]
<EliasT> SELECT * FROM http://torrez.us/srevices/rdfa/[url] WHERE { ... }
Elias: so now we can put RDFa anywhere in the Web and query it with SPARQL
<EliasT> http://abdera.watson.ibm.com:8080/browser/
<EliasT> http://wingerz.com/blog/?p=35
Elias: we just deployed a browser yesterday
<EliasT> http://torrez.us/archives/2006/07/17/471/
Elias: see Wing's post
... we implemented the ATOM protocol to post entries to our endpoint
... we store these posts as RDF
... if the content-type is XHTML we run the java parser and add more triples
to the graph
... we only store the RDF version
<EliasT> a SPARQL endpoint
<EliasT> urn:lsid:abdera.watson.ibm.com:entries:397859049
Elias: click on graph, add the URI to ATOM entry, uncheck introspection
Elias: what this shows is a data store service
that you update using ATOM
... and you extend this using RDFa
<Steven> Doesn't seem to work in Opera
<Ralph> I don't seem to be having much success with Mozilla either
Elias: Tim Bray may demo this at OSCON. I've fixed all the bugs that TimB found
Mark: I don't like that role is becoming rdf:type
Ben: do we agree that rel does not use '[]'?
Mark: yes, we only mention '[]' in the CURIE
document where we say the brackets are only needed if there is a possibility
of confusion between a URI and a CURIE
... so Wing's blog example is
wrong
Ben: the GRDDL transform for RDFa/XHTML2 will recognized the values of rel that are defined in the HTML spec and expand them to xhtml:<value>; other values for rel will be handled differently
Elias: I am constantly faced with how to make embedding of RDF practical
[adjourned]
<benadida> can we meet next week same time?
<Ralph> same time next week OK with me
<benadida> Elias, thanks for a fantastic demo
<Ralph> +1
<benadida> send me an email list of people to credit?
<benadida> sorry, I mean, send me, by email, a list of people to credit
<EliasT> k. I willl.