From: Seaborne, A. <and...@hp...> - 2005-04-30 15:33:48
|
Chris Bizer wrote: > Hi Andy and all, Hi Chris, > > first, welcome as new reader on this list. > > By the way, I'm currently planning how we will get Named Graphs and SPARQL > into the next RAP release and maybe you would comment on how our plans fit > with the DAWG and the Jena timeline. > > 1. Daniel Westphal will port NG4J to PHP and starting now and hopes to be > done mid May (including the API, TriX and a DB backend). > Q:Does Jena already have concrete plans on how to support Named Graphs? > Maybe we could implement the same API. Jena will be providing an API because it's needed for SPARQL. ARQ has a updated version of MultiGraph/MultiModels in DataSet/DataSource but the graph and model aspects need separating. It was put firstly to support experimentation of GRAPH in SPARQL and now that is settling down, it's time to work on the interface and see if it migrates to the main Jena package. > > 2. Another Student (Ralf Jacob) will implement SPARQL protocol into our > Joseki clone and might also implement a client library (remoteModel) and > some access control mechanisms (graph, resource and predicate level). He > will start 1st of June and might be done at the end of June. > Q: Will the SPARQL protocol be stable enough by then for being implemented? [I'm not the editor for the protocol so these are my guesses] Good question! Don't know but I would expect that the basic HTTP binding would be clear by then - parameter names may change but the style and design should be stable. Maybe the same for the SOAP one - it needs more machinary to be defined. > Q: Do you know of any other work on access control for RDF servers? I don't know of any work on access control that is specific to the semantic web - the usual security features of your web application container still apply. > > 3. Alberto wrote a SPARQL query language parser for PHP. > http://gargonza.asemantics.org/downloads/sparql.php > http://demo2.asemantics.com/rdf4food/sparql.html > I'm still looking for somebody to update this parser for the new SPARQL > syntax and write the actual query engine. I currently have nobody for this. > So it would be really cool if somebody from this list would volunteer. > Q: Any guess on the timeline when the spec will be stable? I do not expect the syntax to change much. There is one planned revision which should happen soon and it is all fine-tuning now, I hope. In particular, the tokens are stable and that is quite a bit of the work. What technology does the parser use? Hand written or machine generated? ARQ has a number of syntax test cases and these have been checked into DAWG CVS. There are 60-odd test cases. You cxan also use the rest of the ARQ test suite - much of it is scripted using the DAWG manifest vocabularies. > > 4. There are also two people (Hannes Gassert and Tae IL Erdmann) working on > an RDF browser implemented using RAP and Fresnel > (http://simile.mit.edu/repository/fresnel/trunk/docs/manual/FresnelVocabular > y.htm) which might also become part of the next RAP release. > Q to Hannes Gassert: Right or do you want to distribute Arago as a separate > download? > > Chris > And is there an online demo? Andy |