SPARQL is to linked data as SQL is to relational databases. It is a query language for semi-structured, open-ended data. SPARQL is also an increasingly popular way of exposing large datasets, such as those provided by the UK government.
Currently, the only substantial bit of jQuery used is its abstraction for JSON-P. This may change as we look to concrete application use-cases. Beyond jQuery, the library uses Yahoo!'s YQL service, specifically Dave Beckett's Triplr table, to wrap around arbitrary SPARQL endpoints (which often don't support either CORS or JSON-P). It is hoped that this part of the library can be abstracted out soon.
There are many things left to do with this library, including but not limited to:
- Abstracting out the execution system to support different endpoints and (maybe) any local SPARQL implementations.
- Support for more sorts of query than just SELECT (including update operators).
- Integration with follow-on processing, such as DOM population, data analysis, summarisation and visualisation.
- Production of detailed documentation.
- Lots and lots of tests.
If you have any ideas about these things, or need to solve them for your work, please get in touch and contribute to the project!