Re: VSS collaboration (Data TF) meeting today?

Hi Benjamin,

As it happens I am attending Open Geospatial Consortium (OGC) meetings
this week and was in a session that included trying to unravel the
confusion of the different uses of SHEX, SHACL and some other tools. I
think Graph is more common and approachable but we can run both in
parallel to contrast.

Graph alternates are about and Neo4j's is indeed proprietary but they
are committing to GQL which was part of the discussion Dave Raggett and
I had with them.

https://www.gqlstandards.org/

https://neo4j.com/press-releases/query-language-graph-databases-international-standard/

More from OGC and W3C Spatial Data on the Web Interest Group (where we
coordinate with them extensively), we may have some ontologists that
want to play with us and do cross SDO on routing, trip and observation
(involving VSSo) ontologies. This is stemming from the transportation
workshop as are several other things. They will be making public their
engineering report on a new routing web api and accompanying ontology
ideas. The API uses JSON at present, in particular GeoJSON and they're
torn between it and JSON-LD. 

On Mon, 2019-11-18 at 15:34 +0100, klotz@eurecom.fr wrote:
> Hi Ted and all,
> 
> > Benjamin gave his insight on the various graph database engines,
> > that
> > basically the various open source ones require considerable more
> > custom
> > programming to provide services on top of them. His and general
> > opinion
> > seem to favor Neo4j.
> 
> I had some thoughts about this and I see basically 2 directions for
> us:
> - RDF/RDFS/OWL and W3C standards: that would require some effort in  
> the design and maintenance of the DB and query interface.
> - Existing DBMS platforms (most likely Neo4j), which come with
> their  
> prorietarw query languages (cypher for instance) but would be
> deployed  
> more quickly.
> 
> I saw an interest in demonstrating the benefit of the VSS model  
> through GraphQL, and that is of course possible with both
> approaches:  
> with HypergraphQL [1] (or equivalent) on linked data or a GRAND
> stack  
> [2] including neo4j.
> 
> It mostly depends on the community using the interface. I have a  
> personnal preference for the open standard way (RDF/OWL...) so I
> can  
> use directly ontologies (like VSSo) without a need for adaptation  
> (e.g. fro URIs to local IDs) and export JSON-LD data, but I
> understand  
> that this is not mandatory for graphQL interface demonstrations.
> 
> [1] https://www.hypergraphql.org/
> [2] https://grandstack.io/
> 
> Best regards,
> Benjamin
> 
> > I met with several people from Neo4j by phone about getting an
> > enterprise license donated, as it has additional features over the
> > community edition. They were favorable and will likely meet one in
> > person at Open Geospatial Consortium this week. They may also
> > extend to
> > us usage of their cloud service. I'll setup a community edition on
> > the
> > server we have at MIT for now.
> > 
> > Since Patrick is sending regrets for Tuesday, perhaps we can take
> > some
> > of the WG call tomorrow to discuss next steps on VSS Graph project.
> > 
> > https://www.w3.org/auto/wg/wiki/Vss_data
> > 
> > On Thu, 2019-11-14 at 09:52 +0100, Ulf Bjorkengren wrote:
> > > Will there be a meeting today at 16.00 CET to discuss the VSS
> > > collaboration project?
> > > 
> > > BR
> > > Ulf
> > > --
> > > Ulf Bjorkengren
> > > Geotab
> > > Senior Connectivity Strategist?|?Ph. D.
> > > Mobile +45 53562142
> > > Visit www.geotab.com
> > > 
> > >  Join us at Connect 2020San DiegoJanuary 13 - 16, 2020
> > > 
> > > Register Now!
> > > 
> > > 
> > 
> > --
> > Ted Guild <ted@w3.org>
> > W3C Automotive Lead
> > https://www.w3.org/auto
> > 
> > 
> 
> 
> 
> -------------------------------------------------------------------
> ------------
> This message was sent using EURECOM Webmail: 
> http://webmail.eurecom.fr
> 
> 
-- 
Ted Guild <ted@w3.org>
W3C Automotive Lead
https://www.w3.org/auto