See also: IRC log
<Arnaud> barely awake, not up for scribing, sorry
<phila> kurzum: (Sebastian Hellmann) introduces himself
<phila> ... talked about his PhD on RDF mapping, internationalization format etc.
<phila> ... other work relates to legalistic Linked Data cloud
<phila> ... there's a new domain called linguistics - new datasets that I'm involved with.
<phila> ... I have 6 PhD students here in Leipzig where Soeren was previously
<phila> Kerstin: Works for Astra Zeneca. Involved in Sem Web mostly for clinical trials data. Also more general stuff. have talked with Eric about Shapes etc
<Kerstin> http://www.phusewiki.org/wiki/index.php?title=Semantic_Technology
<phila> phila: Asks others for quick intros from others
<danbri_> http://www.w3.org/People/Arnaud/
<Kerstin> https://twitter.com/kerfors
<scribe> chair: PhilA
<scribe> scribe: bhyland
<kurzum> http://rdfunit.aksw.org/demo/ provides an implementation for reading DSP profiles and also IBM resource shapes
<kurzum> created by one of my colleagues in leipzg
dwood: Asked when the older W3C working group co-chairs should roll out?
phila: Explained he'd like to see some seasoned members remain involved to the tune of 1 hour/mos to join the coordination call, if possible.
Arnaud: Discussed LDP WG updates ...
Option is to stay in CR in perpetuity vs. go to last call again. Issues with getting time commitment for an implementation ...
<phila> http://www.w3.org/TR/ldp/#ldpic is the feature at risk in the new LCWD
Arnaud: Just published LDP Paging & went to LC
<davidwood> Arnaud, The Callimachus Project should have an implementation test report to you within 3-4 weeks. Apologies for the delay.
Also, LD Patch, a patch format so you can patch for defining expressing a sequence of ops for LD resources. Used for HTTP PATCH method, see https://dvcs.w3.org/hg/ldpwg/raw-file/ldpatch/ldpatch.html
<danbri_> odrl? http://www.w3.org/TR/odrl/
<phila> ODRL
Editor's Draft is available for review. Arnaud said it is too early to tell if there will be sufficient public feedback.
Phila: Access Control is
anticipated to be a 'charterable' topic for a future LDP
Group.
... Is soliciting input from other W3C WG chairs to see if
there is synergy / interest.
... The bar to starting a new W3C working group is high, so we
have to make sure we do sufficient research & make sure
there is sufficient interest.
DavidWood: The topic of Access Control is critically important but inextricably linked to a topic where everyone has a 'favorite dog' in the race.
… What I like about the LDP spec(s) is the simplicity of the recommendations, avoiding the Access Control 'hairball'.
Arnaud: I would prefer to see LDP recommendations be a set of detailed 'modules' that people can select to implement/adopt.
Davidwood: So how do you foster interoperability?
Arnaud: Today, people are adopting other means of doing Access Control, yet they are implementing parts of the LDP.
Davidwood: I don't want to take up the time of the DACG's time today to deal with this, but we can offer some concrete input having done an implementation of the real life issues we've faced dealing with Access Control. It is an important issue.
danbri_: RE: CSV on the Web WG, see http://www.w3.org/2013/csvw/wiki/Main_Page.
There has been recent discussion is around mapping language for CSV
<danbri_> see also this thread, http://lists.w3.org/Archives/Public/public-csv-wg/2014Sep/0006.html
<danbri_> sorry, had to drop off.
<kurzum> there is a mapping wiki with syntax for publicdata.eu
<kurzum> http://wiki.publicdata.eu/wiki/Csv2rdf:0a7c3648-9e7e-4e9c-8647-a0579fe19fed
<kurzum> to annotate publicdata csv files
RE: Clinical interest group calls, per EricP ...
Group discussing clinical observations ...
Would coordinate with the HCLS Group. Good pairing because people expect respective groups to understand semantic technologies & HL7, respectively.
Has 4 co-chairs but none are on the call today.
Phila: It is going ok, lots of discussion, and working on getting words on paper. We're meeting at TPAC in October 2014.
<phila> Data Shapes Charter
<phila> (proposed)
EricP: on history of Data Shapes Working Group
Desire for RDF data validation/interface definition on SemWeb ...
How do we balance the social issues that balance the diverse interests of those who have invested in different approaches to address interpretation of shapes or optimize SPARQL queries & create UIs.
<phila> TPAC registrants
… If we meet F2F in TPAC and discuss approaches, that would be ideal.
Arnaud: I've been asked to chair & I agreed, but I'm looking for a good strong co-chair
<phila> Main TPAC page
… We were at this workshop last year (2013) and there were concerns about whether standards should lead or lag specifications.
… The members of the workshop generally were in agreement back in 2013. However, when it came time to charter the working group, there was more contention on the public list.
… We'll see how this pans out. There are companies who are quite far down the road with implementations but we feel going through the W3C process would be beneficial long term.
<ericP> TPAC reg (pasted re: shapes WG registration)
<kurzum> kurzum -> Sebastian Hellmann
<danbri> how does this relate to http://json-ld.org/spec/latest/rdf-graph-normalization/ ?
<danbri> "This document outlines an algorithm for normalizing RDF graphs such that these operations can be performed on the normalized graphs."
Phila: During the formation of the Data Shapes WG Charter, there was much humming & hawing around Graph Normalization.
ManuS has proposed what he believes to be a credible solution.
… that he proposed during his recent SemTech 2014 keynote.
<davidwood> I have not found a need for graph normalization, but that doesn't mean there isn't one.
Phila: Is graph normalization central to the Shapes Charter?
<phila> No, it's not in that charter, bhyland
<kurzum> My opinion is, that this is a research issue. Is there a method to do a diff? and decide equality
Correction: Discussion was whether graph normalization *should* be added to the charter?
<danbri> Shapes is interesting, because it gives an RDF version of the "is this data valid or not?" mechanisms people are familiar from in XML etc.
<Zakim> ericP, you wanted to say that shapes can give you the ordered traversal of a graph which can produce a normalized graph.
<danbri> RDFS/OWL expressly don't do that.
Kurzum: There are higher priority items to be included in his opinion. Graph normalization may be more suitable as a research topic.
EricP: Described use cases where it would be useful, e.g., ordered traversal of graph.
<Kerstin> I've a very practical need: a good way to represent value sets, e.g. the so called controlled terminiligies drawn fromCDISC / NCIt, see http://kerfors.blogspot.se/2013/10/the-future-of-cdisc-cts.html
Kurzum: SEMANTICS is a 10 year old event, started as academic research conference, now moving towards industry engagement.
… Industry is interested in Linked Data, but we need bridges to make it seem more relevant.
<phila> bhyland: Dave and I attended SemTech(Biz) in San Jose
<phila> ... new owner added Biz to emphasise business application
<phila> ... booths with strong Tech had more visitors than the marketing types. Tightly coupled with NoSQL conference
<phila> ... we (3 Round Stones) were next to Cloudera (2 $bn VC funding) and Mongo. Lots of people don't get what a graph is
<phila> davidwood: I'be neen to about half the SemTech events. Didn't think it was too useful for business but it was fun.
davidwood: I previously thought SemTech was more of a conference to connect with my peers, but not really to do 'business'. In fact, after attending about 5 over the last 10 years, it was a good conference for our company to conduct business.
<phila> ... Can be hard to justify going. This year we decided to sponsor and speak. The crowd has changed. More people interested than used to be the case
… There was more discussion around partnering & commercial work getting done.
Phila: We're out of time. Would like to meet once more before TPAC. Next meeting is 15-Oct. Thanks all!