Semantic Web Coordination Group Teleconference

<ivan> last meeting's minutes

PROPOSED to accept last meeting minutes

<mhausenblas> +1

<ivan> +1

<pgroth> +1

RESOLVED to accept http://www.w3.org/2012/02/08-swcg-minutes.html as accurate last meeting's minutes

<pgroth> does it come ever?

<pgroth> summer that is

<mhausenblas> Michael: Just learned that of today 'Axel Polleres no longer represents DERI' - re SPARQL WG

Next meeting will be one hour earlier for Europeans due to US DST.

Axel Polleres is expected to be re-appointed by Siemens.

… and will need to be formally reappointed as SPARQL WG chair.

<ivan> Agenda call

RDFa documents expected to go to CR on Tuesday

RDFa publications: RDFa 1.1 Core, RDFa+XHTML, RDFa Lite (FPWD), RDFa Prime. See http://www.w3.org/blog/SW/2011/12/09/rdfa-lite-1-1-draft-published-rdfa-1-1-primer-updated/

and http://www.w3.org/blog/SW/2011/12/16/new-versions-of-rdfa-core-1-1-and-the-xhtmlrdfa-1-1-drafts/

Linked Data Patterns Workshop, work on a Linked Data Working Group charter

Work on charter continues. Sandro trying to rein in to just addressing RDF

<sandro> proposed: Mission: to standardize a simple and practical application integration technology based on REST and RDF.

Discussion around the draft charter has expanded (probably inappropriately) to involve general patterns, techniques.

Linked Data Platform Working Group DRAFT Charter: http://www.w3.org/2012/01/ldwg-charter.html

<mhausenblas> danbri around?

Ivan: "application integration" is too general a term.

Sandro listed some tasks needed by IBM for their own products.

<sandro> ivan: this is the equivalent of atompub, and what apps do with that is not in the scope of this wg

Ivan: Product-specific concerns are not in the scope of the WG.

<sandro> Ashok_Malhotra: There were people wanting to add things like access control -- what are you thinking about that?

Ashok: Some people want to add access controls and security items to the charter.

<sandro> ivan; The group will define and document REQUIREMENTS for access control (but not standardize the technology

Ivan: This was raised on the mailing list. The group will define and document requirements for access control, but not recommend solutions.

… That is expected to stay in the charter.

Ashok: If we start with data and make it into Linked Data, then we build apps that use that Linked Data, we then have difficulties with parts of RDF. One of the biggest ones is how to handle collections. Another is how to handle very large data sets.

<sandro> Ashok_Malhotra: About Ivan's concern about App Int. We start with lots of data, represented as Linked Data, then we start to write apps working with it. When we do that, we have had difficulty with certain parts of existing RDF. One of the biggest ones is how do you handle collections. The other is how you handle very large datasets. Etc. So, the idea is to think about these and standardize patterns to be able to handle these situations.

<pgroth> +q to ask about why standardization and not best practice

… So we need to think about those problems and possibly standardize patterns that can be used to address these problems.

<sandro> ivan: My problem is: at some point in the time, the goals and mission have to be shown and made understood to the general public. And for me, App Int goes beyond what a WG can do.

Ivan: The mission of the WG needs to be understood and made clear to the general public. If there are additional patterns, that's fine, but "application integration" is still too general a term.

<mhausenblas> Michael: I'm looking at http://www.w3.org/2012/01/ldwg-charter.html - right?

<sandro> Mission: to standardize a simple and practical application integration technology based on REST and RDF.

<Zakim> mhausenblas, you wanted to ask about goals

Sandro: suggests a revised mission statement ^

Ashok: Can we use the term "application integration patterns"?

…instead of "technology"

<mhausenblas> hey! I did not speak!

Sandro: Sure

<sandro> Mission: to standardize a simple and practical application integration patterns based on REST and RDF.

<mhausenblas> [[The mission of the Linked Data Platform (LDP) Working Group, part of the Semantic Web Activity, is to provide standards to develop and deploy interoperable applications based on Linked Data. The scope of the Working Group is to provide profiles of existing Semantic Web standards, as well as to develop further W3C Recommendations in case the current standards do not cover needs arising in the area.]]

Michael: The mission and scope in the current draft differs.

Sandro is trying to change that. Michael wants a concrete proposal.

Ivan: My question will be whether a change of emphasis will be acceptable. The draft charter is too general. Perhaps we need another, separate WG to define Linked Data profiles.

Discussions regarding the W3 process occurred.

Michael: What is the current proposal? Is it Sandro's?

Ivan: Yes

<Zakim> pgroth, you wanted to ask about why standardization and not best practice

Paul: Are you concerned about the "extra" recommendations? Shouldn't these be best practices? I am concerned that the WG could take a long time.

Sandro: Hopes that narrowing the focus will help.
... AC review will probably narrow the focus anyway.

<Zakim> davidwood, you wanted to respond to Ashok, Ivan.

<mhausenblas> PROPOSAL: The mission of the LDP WG is to standardize a simple and practical RESTful application integration technology that is Linked Data compliant.

<mhausenblas> Michael: Note, we should say 'LD compliant' rather than RDF (there *may* be an alternative serialization in JSON, see for example http://webofdata.wordpress.com/2012/02/05/json-http-data-links/)

<mhausenblas> Michael: Note as well that we would need to state what simple and practical means in a second sentence to follow this first sentence.

<sandro> davidwood: It seems to be that ... large enterprises have noticed problems with using RDF in anger ... and we each have our solutions ... this group has a huge motivation to get it a solution right. We owe it to ourselves to refine these patterns and get them out there.

<pgroth> +1 for guidance

<pgroth> point is *guidance*

<sandro> sandro: David, I'd really like to see your list of technologies you see necessary there, in case we need to enumerate them in the charter.

<mhausenblas> Michael: For example, simple == TCO, ROI, implementation complexity and practical could be defined via the use case (80/20 solution)

<sandro> davidwood: Still working on it, cf callimachus. Main thing is to expose existing data instead of migrating data to RDF. That's the kind of pattern that's working for us. It's a little early to suggest that be a W3C recommendation.

<Zakim> mhausenblas, you wanted to talk about a concrete wording of the mission

… but it could evolve into a best practice Note

Michael: Has a proposal. Do you want it now?

Sandro: The CG doesn't decide, but a strawpoll would be helpful.

<mhausenblas> PROPOSAL: The mission of the LDP WG is to standardize a simple and practical RESTful application integration technology that is Linked Data compliant.

(Scribe off) I think I know what "Linked Data compliant" means now, but not what it will mean next year.

<pgroth> I've got a counter proposal :-)

<sandro> mhausenblas, "RESTful" not "REST": "Linked Data" not "RDF". We might want to use JSON.

<sandro> mhausenblas: If we confine ourselves to purely RDF serializations (RDF/XML, Turtle, RDFa), then [ something about JSON ]

<sandro> sandro: So you don't want to rule out JSON?

<sandro> mhausenblas: Right.

<pgroth> PROPOSAL: The mission of the LDP WG is to recommend a set of simple and practical RESTful application integration patterns that are Linked Data compliant.

<bhyland> we're getting warmer, +1 to this one.

+1 to Paul's proposal

<sandro> pgroth: "patterns" instead of "technologies"

<mhausenblas> +1 to Paul

<Ashok_Malhotra> +1 to Paul

<sandro> mhausenblas: Any objections, negative thoughts on this?

Tom: Is the draft charter public?

Ivan: Not yet

<sandro> sandro: It's public access, but not really ready for wide-spread review

The document is public access, but not ready for public review

Bernadette: That is a good refinement. Section 2.1 (LD Profiles) is still unclear.

<sandro> ivan: I expect the mention of OWL will disappear in the next version

Ivan: The eventual charter will probably not mention OWL.
... Does it make sense to propose another group for LD profiles?

<mhausenblas> Michael: Sandro can you mark http://www.w3.org/2012/01/ldwg-charter.html as DEPRECATED in big bold red letters and provide a link from there to the new version, please?

<Zakim> davidwood, you wanted to ask about a separate group formation

It seems to me that the LOD community needs will remain separate from the needs of the enterprise community.

<bhyland> davidwood: The needs of the enterprise (using LD) behind the firewall vs the needs of LOD publishers (on public Web) will remain different. The line is clear thru the firewall.

<Zakim> mhausenblas, you wanted to talk about audience

<bhyland> davidwood: Does this activity have the appetite for another WG to focus on these differences?

Ivan: Maybe

Michael: Lots of people care about the public Web, but there is also an enterprise need. The LDP WG draft charter should focus on the enterprise.
... Should be clear about focus of the group in an "out of scope" section.

<bhyland> I appreciate the opportunity to iterate together, thank you.

Ontology for Media Resources

A Recommendation for Ontology for Media Resources was recently approved.

The chairs could not join us today.

…so we won't discuss.

AOB

Michael: schema.org extensions should be added to the agenda in two weeks.

ADJOURNED.