Re: Minutes: XDI TC Telecon Friday 2014-12-05

[Markus Sabadello <markus.sabadello@xdi.org>]

Hello, Peter and Hubert, just wondering if the "RDFify XDI" presentation you gave during this call is in Kavi or elsewhere for people to view?

Markus

On Tue, Dec 9, 2014 at 9:07 AM, Markus Sabadello <markus.sabadello@xdi.org> wrote:

XDI TC Minutes

Following are the minutes of the unofficial telecon of the XDI TC at:

Date:  Friday, 5 December 2014 USA
Time:  09:00AM - 10:30AM Pacific Time (16:00-17:30 UTC)

ATTENDING

Les Chasen
Peter Davis
Hubert Le Van Gong
Drummond Reed
William Dyson
Markus Sabadello
Joseph Boyle

GUESTS

REGRETS

NEWS & UPDATES

The TC wished Drummond a happy birthday!

Report from EWTI

Markus was at a conference in Vienna (European Workshop on Trust and Identity):

http://identityworkshop.eu/

Markus reported that the conference was inspired by Internet Identity Workshop. It was more heavily focused on OpenID Connect and SAML than IIW. Markus held an introduction to XDI session.

Part of the session discussed the differences between the RDF and XDI models and showing how the same data (a person’s email and list of friends) would be modeled in both.

Markus  said he also met a potential implementer.

PRESENTATIONS/DISCUSSIONS

Report from XDI editors subcommittee

https://wiki.oasis-open.org/xdi/XdiOneSpecs
https://github.com/OASIS-XDI-Technical-Committee/xdi-spec-docbook

Peter reported that the Editors TC started to review the updated XDI Dictionary wiki page that Drummond was able to substantially complete over the U.S. Thanksgiving break. The discussion focused mainly on the role of schemas and ontologies in XML and RDF, and how an XDI dictionary was similar yet different.

XDI-RDF

Hubert, Peter and Les have worked on “RDFify XDI”, i.e. using widely known technologies such as RDF, OWL, SPARQL, Linked Data, TriG to realize the XDI goals and principles.

Summary of XDI-RDF by Hubert:

With the recent release of RDF 1.1, the W3C community has made considerable progress towards supporting some of the foundational requirements of XDI. Moreover, related work like the Data Linked Platform specification or the SPARQL protocol over HTTP specification provide key components (graph manipulation, messaging) to the overall XDI software stack. All these specifications are very recent and did not exist when the XDI TC embarked on its vision. With the recent publication of the above standards, and the resurgence of interest in RDF-based semantic graphs, we believe that the TC should align its work with RDF. The main benefits would be to make the TC's work more friendly to the semantic community (especially at the syntax level) but also to be able to leverage the large number of libraries and implementations of RDF 1.1 (thus making our work more appealing to the developper community).

The proposal to the XDI TC is two-fold:

- To create (in parallel to ongoing work) a profile of existing W3C specifications (see list below) along with well established internet standards (e.g. DNS) to fulfill the XDI principles with an internet protocols based stack.

- In addition to this profiling activity, the TC will define missing components where necessary (e.g. policy). This activity will leverage ongoing TC work as much as possible.

List of potential specifications:

RDF 1.1 Concepts and Abstract Syntax (http://www.w3.org/TR/2014/REC-rdf11-concepts-20140225/)

SPARQL 1.1 Graph Store HTTP Protocol (http://www.w3.org/TR/sparql11-http-rdf-update/)

SPARQL 1.1 Protocol (http://www.w3.org/TR/sparql11-protocol/)

Linked Data Platform 1.0 (http://www.w3.org/TR/2014/PR-ldp-20141106/)

Linked Data Platform Paging 1.0 (http://www.w3.org/TR/ldp-paging/)

Linked Data Platform Best Practices and Guidelines

RDF 1.1 TriG (http://www.w3.org/TR/trig/ )

Linked Open Vocabularies - LOV (http://lov.okfn.org/dataset/lov/)

Good Relations Ontology (http://purl.org/goodrelations/v1)

RFC3987 Internationalized Resource Identifiers (IRI) (http://www.ietf.org/rfc/rfc3987.txt)

We discussed whether this proposal would result in additional specs in the list of XDI specs. Peter’s and Hubert’s initial thinking is that it would only require one additional spec, because that spec would largely profile existing RDF 1.1 and LinkedData specs.

ACTION ITEM: Peter and Hubert to suggest a name for the XDI-RDF spec.

Markus’ interpretation was that the proposal was to use RDF and related technologies mostly “as is”, with a few additions, such as specifying an ontology for modeling policies, developing several new discovery mechanisms, and profiling SPARQL queries for certain uses.

Drummond was concerned that this may end out being a bigger job than anticipated because, although the new RDF datasets (named graphs) in RDF 1.1 help, translating all the features of the XDI graph model into RDF 1.1 is more complex than it looks.

Drummond also said that it was critical that this proposal meet three conditions:

1. It would not require changes to “native XDI”.

2. Adding this specification to the TC’s workload must not slow down our work to complete the native XDI 1.0 specification suite.

3. The resulting RDF 1.1 graphs that are compliant with this new specification must be 100% compatible with native XDI 1.0 graphs, i.e., you must be able to introconvert losslessly between a graph serialized in the RDF 1.1 format and one serialized in the XDI JSON 1.0 format.

Joseph asked if Hubert and Peter preferred any of the RDF serialization formats over the others. Peter said he liked TriG and JSON-LD. Joseph also asked if Hubert and Peter are planning to use existing RDF datasets/ontologies. Hubert confirmed this was the case.

XDI Messaging and XDI Bindings contributions

Markus said he would merge his pull request with new contributions, since he hadn’t received any feedback.

XDI Policy and Connections

We did not further discuss this topic.

Transactional integrity

We did not further discuss this topic.

NEXT CALL

The next call is next week at the regular time.