- From: Danny Ayers <danny.ayers@gmail.com>
- Date: Thu, 18 Aug 2011 01:03:25 +0200
- To: Giovanni Tummarello <giovanni.tummarello@deri.org>
- Cc: Semantic Web at W3C <semantic-web@w3.org>, Linking Open Data <public-lod@w3.org>, Andy Seaborne <andy.seaborne@gmail.com>
Nice work! > Due to requirements of query functionalities and aggregates, Sparallax > currently only works on Virtuoso SPARQL endpoints. (do other > triplestores have aggregates? if so please let us know and we'll try > to support other syntaxes as well) Which aggregate functions are needed? ARQ has some support (very possibly more than) listed here: http://jena.sourceforge.net/ARQ/group-by.html [Andy, are there any more query examples around? I can't seem to get count(*) working here] I see there are Graph and Dataset fields on the form - how are they used? What happens with an endpoint that only has a default, unnamed graph? > Try sparallax on our test datasets or put just the URL of your SPARQL > endpoint (but make sure you have materialized your RDFs triples, it's > needed) materialized? (I assume that's not the httpRange-14 extension for dogs :) What else can be done to the data to make it more Sparallax friendly? I noticed clues in the form boxes for abstractprop and imageprop, also the config file - http://code.google.com/p/freebase-parallax/source/browse/app-parallax/sparallax/scripts/config.xml Do I take it from the rdfs:Class entry there that it uses RDFS to decide on collections? Sorry for the raft of questions, but I recently put LD browser on my todo list, completely forgetting about Parallax and all the boxes that already ticks, and now Sparallax... so I want to get an idea of how it works to avoid reinventing the wheel (in a square shape). Cheers, Danny. -- http://dannyayers.com
Received on Wednesday, 17 August 2011 23:04:02 UTC