Minutes of Feb 4 Property Graphs CG Telcon

Thanks to Kelvin for scribing

Property Graphs CG Telcon, Feb 4, 2014

<Ashok> present: Michael, Ashok, Kelvin, Andy, Gregg, Ted, Zhe

[12:03] <Kelvin> Kelvin elected(!!) as scribe
[12:03] <Kelvin> Vote to approve minutes of last meeting
[12:03] <Ashok>http://lists.w3.org/Archives/Public/public-propertygraphs/2014Jan/0040.html
[12:04] <Kelvin> No objection to approval
[12:04] <Kelvin> Ashok: Phil unable to join us
[12:05] <Kelvin> Ashok: At inflection point. Have done some use case work and discussed outline of a charter. Now need to decide if there is sufficient support to start this work and will people implement
[12:05] <Kelvin> Ashok: Getting "questioning looks" from W3C about this
[12:06] <Kelvin> Ashok to Ted "If we do this will OpenLink implement" Ted replied "Yes"
[12:06] <Kelvin> Kelvin asks "Implement what?"
[12:07] <Kelvin> Ashok: Asking that W3C create a WG for a standard Property Graph data model
[12:07] <Kelvin> Ashok: Open Link and Oracle would be 2 impls but real question is, will this be widely accepted
[12:07] <Kelvin> Ashok: Not sure how to answer that question
[12:08] <AndyS> Q to Oracle : is this implement PG in addition to Oracle spatial or to replace? Or is Oracle spatial considered PG already?
[12:08] <Ashok> Kelvin:Not sure we have critical mass so industry would get behind this
[12:09] <zwu2> in addition
[12:09] <Kelvin> Kelvin echo'd concern that we don't have enough of the key players in this CG like Tinkerpop/Titan, Neo to judge if there is buy-in here
[12:10] <zwu2> we may be able to make things compatible :)
[12:11] <Kelvin> Kelvin tried to make it clear he is very nervous we do not have the buy-in. Ashok offered to start asking people in e-mail if they would implement a standard
[12:12] <Kelvin> Kelvin said he would like to hear from more of the key players both open source and commercial
[12:12] <Kelvin> Answer to Oracle question was "new feature in addition to RDF capabilitie

[12:13] <Kelvin> Analytical functions like page ranking
[12:15] <Kelvin> Ashok will e-mail key players and ask if they would implement and maybe even become part of writing a Data Model spec
[12:15] <Kelvin> Ashok perhaps we could start writing a document to describe what the standard might look like and that might help us get a WG started
[12:17] <Kelvin> Andy: Only way getting it done in a year is realistic is if a lot of work is done up front to get things going. Spec therefore needs to be done in conjunction with implementations
[12:17] <Kelvin> Andy: Otherwise how do you know if you are solving the right problem?
[12:18] <Kelvin> Andy: Need a lot more solid basis going in or else very little reason for people to turn up and actively participate
[12:19] <Kelvin> Kelvin: I agree parallel implementation, ideally in open source would help grow confidence that such a spec has merit
[12:24] <Ashok> Ashok:Should we start outlining a spec?
[12:24] <Ashok> Kelvin:Yes, that would help
[12:24] <AndyS> See https://github.com/tinkerpop/blueprints/graphs/contributors
[12:25] <Kelvin> Kelvin: I think a writeup of "what we have in mind" would help with the discussion with the graph players
[12:26] <Kelvin> Kelvin: and if a WG is formed a blank slate is never a good idea so a seed spec definitely helps
[12:26] <Kelvin> Andy pointed out that the URL (above) shows a list of implementors
[12:26] <zwu2> yes
[12:33] <Kelvin> Several people commented that the various graph players like Neo and Titan have declined to join the W3C effort. Ashok mentioned that Marko R from Aurelius said we should "join them"
[12:37] <Kelvin> Andy asked where all the people who expressed interest in this work at the workshop are?   Ashok replied he has tried hard to get them to join
[12:40] <Kelvin> Kelvin suggested we build a table of possible implementors and document their interest level
[12:41] <Ashok> Ashok:Yes, that's a good idea!
[12:42] <Kelvin> Kelvin asserted that it is more important to get the "graph technology providers" behind this initially than the big graph users (like Facebook)
[12:42] <Kelvin> There was no disagreement with that assertion
[12:46] <Kelvin> Kelvin started a discussion that maybe we should have a short writeup of what our spec might address
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
[12:49] <Kelvin> Suggestion was made we could take the current Blueprints "spec" and document things that might be added to make that more powerful
[12:52] <Kelvin> Meeting adjourned at 12:52pm EST
-- 
All the best, Ashok