On 20/01/2014 06:41, "Andy Seaborne" <[email protected]> wrote:
>The Google Summer of Code cycle is about to begin again: > >http://www.google-melange.com/gsoc/events/google/gsoc2014 > >March 10 - student application period starts >March 21 - student application period ends >April 21 - acceptances > >If there are any ideas, then we should get them into JITA with tag GSoC. > Also I wondered about things that would as part fo the project need >investigation, needing a significant proportion of the time on deciding >what to do as well as some amount of doing. > >A few ideas that occurred to me, in no particular order, which I'll >think about more to make sure that they are realistic in the time >available: > >+ Prototype an in-memory parallel dataset implementation. > >+ New API(s) -- given Java8, what would a new RDF API look like? > >+ Enhance graph capabilities (e.g. shortest path and other graph >algorithms). > >+ Binary protocol for SPARQL. Maybe one day YarcData will let me open source the stuff I've done in this area, taking my prototype and expanding it would make a nice GSoC project. Rob > >+ Temporary tables in SPARQL - both in one query and also for combining >query and update in one request. > >+ Query planning where there are property functions like text:query >involved (see recent users@) > >+ A data validation, transformation (and correction!) pipeline. > > Andy
