[Bigdata-developers] Higher level reasoning

[Bryan T. <br...@sy...>]

All,

There appears to be quite a bit of interest in supporting higher level reasoning against bigdata.  

I'd like to organize a conference call (Skype) to:

 - Make introductions;
 - Identify core requirements;
 - Identify an approach for which there is some level of consensus *and* on which people would be interested in collaborating.

I see the following as possible goals for such an effort:

- API hooks for other languages (Prolog, Lisp).  This would allow people to provide their own integrations.

- Embedded support for Prolog in SPARQL extension operators. The prolog program would see solutions as they streamed through the query.  The prolog program could issue subqueries against bigdata when filtering or expanding solutions.

- Query rewriting to support rule-oriented model theories, such as the OWL2 RL profile or application defined rules.

- Support for incremental materialization of entailments and view maintenance in a manner which is aware of bigdata's concurrency control architecture.

Please rely either on list or directly with your level of interest and availability for a 1 hour call within the next 1-2 weeks.

Thanks,
Bryan