On Monday 19 June 2006 20:29, Hugh Leslie wrote: > I like the idea of a web service - this fits very nicely with the idea of a > service oriented architecture where you can pick and choose what services > that you want to use. A smart service like this would cache on the local > machine or server so that only updates/or new interactions would pause the > system (although with enough grunt at the other end you might not notice > the pause - done a google search lately!?).
In 2004, when the proof-of-concept drugref interaction server was running (using the interaction data from the Holbrook publication), it turned out that performance was not an issue at all. The OSCAR people in Canada used it, and it seemed to fit their needs. I can start that drug interaction web service again if people want to test it to see how much of a performance problem it might or might not be. http://drugref.org/xmlrpc-interactions.html (ask me by private mail if you want me to fire up the service for testing - the same service also lets you search MIMS interactions and WHO interactions, but the latter two I cannot use on a public server because we never managed to sort out copyright issues, but the work has been done) Horst _______________________________________________ Gpcg_talk mailing list [email protected] http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk
