I am thinking to switch from ARC2 to either Jena or Sesame for SPARQL 1.1 support and the need for inference/reasoning tools. However, I would like to keep working mainly from PHP with MySQL. I've seen that Sesame can be installed as a server and has a native store that can run on MySQL, and have found a PHP wrapper for accessing the Sesame server through HTTP requests.
I have not yet been able to find any information if any of the existing inference/reasoning tools can also be used through an API (that I can then use from PHP) Does anyone know if this is possible?
Also, more general: I am just not so good yet with the command line and server things, nor do I have experience with Java. It looks like just installing the Sesame server could already be quite a task for me since I have no idea yet what a "Java Servlet Container" is or what TOMCAT is. That said, in your experience / advice: Will I get away with setting a RDF server + Mysql triple store + reasoner up and then accessing them from PHP? Or will I have to enter the Java world at some point?
Any advice is greatly appreciated.
You can use a Sesame server directly from PHP, using a generic PHP SPARQL endpoint library such as RAP or even just rolling your own processing using JSON, see these pointers by Kingsley Idehen (they assume you use Virtuoso, but it should work for any SPARQL endpoint implementation).
Sesame Server has a RESTful API and functions as a SPARQL endpoint. For some pointers on getting started with Sesame, there is an excellent blog by Jeni Tennison which takes you through it step by step.
(ObDisclosure: I am on the Sesame development team)