XDI TC Unofficial Telecon Agenda: Monday 10:00 - 11:30AM PT 2016-06-20

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

XDI TC Notes

Following are the notes of the unofficial telecon of the XDI TC held on:

Date: Monday, 20 June 2016 USA
Time: 10:00AM - 11:30AM Pacific Time (17:00-18:30 UTC)

The TC operates under a standing rule approved 17 July 2008 under which the TC does not hold regular official meetings and conducts all business by electronic ballot only. Unofficial weekly meetings are held to enable discussion among members but no business is conducted nor actions taken.

ATTENDING

Markus Sabadello
Les Chasen
Drummond Reed
Phil Windley
Joseph Boyle

How XDI Builds on RDF

The Introduction section of the XDI Core 1.0 spec includes references to a planned document from the XDI TC entitled “How XDI Builds on RDF”. We have not yet authored this document, but the need for it is growing. Developers need to know how XDI as a semantic messaging and data sharing protocol differs from RDF and Linked Data Platform (LDP) and why they may want to use XDI.

We will review the following table prepared by Drummond and Markus to be included in the “How XDI Builds on RDF” document:

#	XDI Feature	Comparison to RDF and LDP
1	Semantic tree graph model provides 100% addressability of all graph nodes	RDF graphs nodes are not uniquely addressable and RDF blank nodes are not addressable outside an RDF graph
2	Uniform model for context (nesting of graph nodes) across 100% of XDI graph nodes	RDF 1.1 provides only one level of graph nesting (with quads and named graphs) and is not addressable
3	Simple, highly efficient reification (making statements about statements) using inner graphs	RDF reification is notoriously difficult and complex; the XDI solution requires the XDI semantic tree graph model
4	Full decentralized protocol for semantic messaging and data interchange	RDF is not a protocol; Linked Data Platform is less capable
5	XDI link contracts and policy expressions (portable semantic authorization)	These features require the XDI semantic tree graph model; they would be very complex to describe in RDF
6	XDI dictionaries provide highly reusable semantic definitions (the equivalent of natural language words)	RDF uses RDF Schema and the OWL ontology language, which is a different approach than XDI semantic trees
7	Clear legal chains of authority for data and policies	This feature requires the XDI semantic tree graph model; it would be very complex to describe in RDF
8	Simple, standard, uniform versioning model for 100% of XDI graphs	This feature requires the XDI semantic tree graph model; it would be very complex to describe in RDF
9	Immutable addresses with standardized discovery (the XDI Discovery protocol)	RDF resources can be identified with URNs to indicate persistence, but there is no standard URN discovery protocol
10	Absolute and relative addresses usable at any level of a semantic tree	URIs for RDF resources are all absolute; RDF blank node addresses are relative only to the containing RDF graph

We spent some time discussing each one of the above features. Joseph reported on some impressions from conversations he has had with RDF developers.

Joseph asked how important the 100% addressability was. Also, Joseph noted that he finds it hard sometimes to explain the concept of "context" outside the XDI community.

Markus said that “context” can be modeled as a named graph in RDF. One potential source of confusion is that “context” is also an important feature in the JSON-LD serialization format, which is completely unrelated to the XDI “context”.

Joseph said that in his opinion, inner roots are the most complicated feature of XDI. Drummond replied that it is still simpler than RDF reification.

We had a discussion about the terms “p2p”, “decentralized”, “distributed” and “client/server”. Markus mentioned that in his experience there is often confusion and ambiguity about such terms, and that he liked Phil’s post on this topic:

http://www.windley.com/archives/2015/01/re-imagining_decentralized_and_distributed.shtml

Markus said that he liked the phrase “semantic messaging and data interchange”, because it emphasizes that these are closely related.

We briefly discussed versioning, and had the impression that this wasn’t a priority in the RDF community.

Drummond concluded that probably RDF 1.1 can support all the features in the table, but that many of them are easier to do with XDI.

JSON-XD and XDI Dictionaries

We did not discuss this topic but agreed that it will be a priority going forward.

NEXT REGULAR CALL

The next call will be the following week at the usual time (Monday 10AM PT). The link to where agenda items can be posted for the next meeting is: https://docs.google.com/document/d/19oDl0lbb56Grehx2a5flZnhrgnua5l8cVvC_dJ8fTXk/edit?usp=sharing

Drummond won't be able to join the next call because of travel.