[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Minutes: XDI TC Telecon Friday 2015-02-13
Following are the minutes of the unofficial telecon of the XDI TC at:
Date: Friday, 13 February 2015 USA
Time: 09:00AM - 10:30AM Pacific Time (16:00-17:30 UTC)
Peter Davis
Markus Sabadello
Drummond Reed
Joseph Boyle
Phil Windley
Dan Blum
Les Chasen
Phil summarized the announcements from IBM that’s radically P2P. The video starts out with the phrase, “I come not to praise the smart home, but to bury it.”
JSON Triples/Quad Serialization Format
Further review XDI Triples and XDI Quad (Tree) format:
https://wiki.oasis-open.org/xdi/XdiTriplesSerialization
https://wiki.oasis-open.org/xdi/XdiQuadTreeSerialization
Drummond reviewed the new documentation for the quad tree format. Markus then showed how this has been implemented in the XDI2 Converter tool (for writing, not yet reading).
https://server.xdi2.org/XDIConverter
Dan asked about the rationale for the new format. Drummond explained the origin of the new format is the ongoing discussion about the relationship between XDI and RDF 1.1. Markus also showed RDF mapping functionality of the XDI2 Converter tool.
There was consensus that this format appears to meet all our needs for becoming the single standard JSON serialization format, and also the standard format used for display. Markus suggested to use the MIME type application/xdi+json for the format. The only other format we should need is the single line display format, which Drummond proposes to call “statement format”, since every line represents a complete XDI statement.
ACTION ITEM: DRUMMOND AND JOSEPH to sync up on discussing the use of single forward slash for prefixing relations.
ACTION ITEM: DRUMMOND and MARKUS to sync up on defining additional rules for serialization of implied context statements.
Further review XDI Variables:
https://wiki.oasis-open.org/xdi/XdiVariables
We reviewed the new documentation on the page, including the new matching rules. The default is now that a typed variable can match any number of instances to any depth. To constrain matching, you must use a defined variable, and specify the constraints using XDI statements in the definition of the variable.
The <$n> attribute can be used to constrain the multiplicity of instances the variable may match, i.e., <$n>/&/”1” would limit it to one match.
The <$depth> attribute can be used to constrain the depth of matching, i.e., <$depth>/&/”1-2” would limit it to two context levels deep.
Markus suggested that another attribute (e.g. <$offset>) would be needed to achieve paging, e.g. to match ordered members 20 to 29 in a collection.
Drummond noted that there is no single variable for matching all types of entities, as there is for attributes. He suggested that another character could be assigned for this purpose, but Markus really didn’t like that idea. So it remains an open question.
We realized that this work on variables is the foundation for an XDI query language we had been envisioning.
Drummond proposes to clarify (and simplify) root node syntax in the ABNF so that parentheses only surround single arcs as they do with attributes, collections, variables, etc. For an example see the XDI Discovery page:
https://wiki.oasis-open.org/xdi/XdiDiscovery
It was late in the call, so Markus, Joseph, and Drummond discussed this and did not yet come to a conclusion. Since this is important for both the Core ABNF and for discovery, they agreed to have a dedicated discussion about this next week. Markus wondered how exactly the discovery process would work. Joseph asked if the “single arc” constraint for peer roots would also affect inner roots.
ACTION ITEM: CORE EDITORS to have a dedicated discussion on this proposal.
The next call is next week at the regular time.
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]