- From: Christopher Gutteridge <totl@soton.ac.uk>
- Date: Mon, 19 Nov 2018 11:58:33 +0000
- To: "<public-lod@w3.org>" <public-lod@w3.org>
- Message-ID: <4c37ef84-c427-e374-86f8-7d43f3b99c8e@soton.ac.uk>
Hi, I built this demo a few years back but it might be worth sharing to give people ideas. It was part of a system called "Alicorn" I prototyped to make it easy to template a website on top of SPARQL. (It was inspired by someone telling me it was very expensive to do) Each URL on the site maps to a URI in a namespace. To serve a page it looks up the rdf:type of the URI and uses that to pick a template & "path expression". The template was just PHP written using my jquery inspired PHP library, <http://graphite.ecs.soton.ac.uk/><http://graphite.ecs.soton.ac.uk/> which has it's faults, but is good for getting stuff done. The path expression was a mix of SPARQL1.1's paths and ideas from regular expressions. It describes all the relationships you want from the URI and is expanded into a great dirty SPARQL query: <http://graphite.ecs.soton.ac.uk/pathdemo><http://graphite.ecs.soton.ac.uk/pathdemo> -- the default example now returns far too much cruft as dbpedia has addd http://dbpedia.org/ontology/wikiPageWikiLink These are ideas I've stopped playing with, but they might inspire future work. This was intended as a system to make life easy on the human. It's certainly not the most efficient! -- Christopher Gutteridge <totl@soton.ac.uk><mailto:totl@soton.ac.uk> You should read our team blog at http://blog.soton.ac.uk/webteam/
Received on Monday, 19 November 2018 11:59:00 UTC