This is a public chat log generated from the #semsol IRC channel.
08:44:13
edsu, we brainstormed a tiny bit about sparql endpoint discovery on the sparqlpress wiki page: http://wiki.foaf-project.org/SparqlPress#SPARQL_Endpoint_Discovery
08:47:14
another idea someone mentioned was <link rel="SPARQL" type="application/sparql-results+xml" href="..." />
08:47:59
following the FOAF/RSS discovery convention
11:37:30
bengee: oh nice!
11:37:58
bengee: also, i hope you didn't read my blog post as being snobbish re: php & python
11:38:22
i have some php scars, but they are from php that looked nothing like your ARC code
11:39:01
heh, np
11:39:15
I wouldn't consider php as being elegant myself
11:40:09
i dunno, your ARC code is as elegant as I've ever seen php :)
11:40:22
oh, thanks :)
11:42:33
frankly i'll probably switch lcsh.info over to using ARC, since the bits the major conversion bits that are written in python could easily talk to the ARC triple store via sparql+
11:42:42
s/the bits//
11:44:10
I see
11:45:06
how did you come to write arc, btw
11:46:22
"php can't process rdf in a satisfying way"-style comments
11:46:38
I started with an rdf/xml parser
11:47:55
I also needed some stand-alone components for a php cms I was working on, and RAP felt a bit too black-box-ish to me
11:49:08
what were the components?
11:50:18
mostly a parser and a basic triple/resource processor
11:50:40
but once you start, things get out of hand quickly ;)
11:50:55
so I added a store, and then sparql etc
11:52:29
:-)
11:52:55
what was the triple/resource processor bit?
11:53:01
sorry about all the questions ...
11:53:32
that was just an array structure for easier resource access
11:54:07
kind of like an in memory triple store?
11:54:08
the idea is still part of ARC2's core
11:54:23
innarestin'
11:55:07
not really a store, just a way to get at a resource's object values in a more php-y way
11:56:04
I wasn't a fan of new Triple(new Resource(...), new Resource(...), new Literal(...))
11:56:26
which many RDF APIs use
11:58:19
mortenf created an rdf/xml parser that replaced RAP's Statement(...) object structure with php-native arrays. That switch alone reduced parsing time by 50%
11:58:57
nice
11:59:21
so I went for that approach in ARC
11:59:32
do you know jon phipps btw, working on the national science digital library metadata registry here in the US?
11:59:54
no, don't know him
12:00:00
http://trac.metadataregistry.org/
12:00:23
he's in the semantic web deployment gruop at w3c, doing stuff w/ skos for digital libraries in the US
12:00:32
anyhow, i only mention it because his registry uses ARC
12:00:49
oh, cool
12:00:55
http://trac.metadataregistry.org/browser/registry/trunk/lib
12:01:19
if you are collecting ARC success stories :-)
12:01:53
ah, that's ARC 1
12:02:04
bengee checks his mail archive
12:02:34
yeah, i doubt he's upgraded
12:02:46
he's been going on that codebase for a while now i think
12:08:23
don't think we've been in contact. interesting poject nonetheless
17:05:09
bengee, you should blog that story :)
17:06:11
heh, yeah, maybe ;)