RE: [PEAR-DEV] rdf in pear .. here we go

> From: Alan Knowles [mailto:alan@akbkhome.com] > Sent: Wednesday, November 19, 2003 1:12 AM > Comments, hey what else do you expect from here :) > > Had a quick look around.. - some thoughts from just looking at it.. > > ADODB -> pear db needs changing in the model bit. Well we should not hurry that one imho. They are open to switching to a PEAR backend. I think MDB's schema management would even offer a good incentive. But it shouldn't be much work. > Their litteral/types eg. > > | $model2->add(new Statement($someDoc, new > > Resource("http://www.example.org/myVocabulary/language"), > new > Literal("English"))); > > Utilizing a class to represent every type, including strings, urls etc.. > does look like a huge overkill.. - perhaps it should follow the more > standard logic, of using native types, and allow a single generic > datatype description eg. > ||new > RAP_Type(32,||"http://www.w3.org/TR/xmlschema-2/integer"||||);| $age = new Literal("26"); $age->setDatatype("http://www.w3.org/TR/xmlschema-2/integer"); I think they need to be objects. Of course you could use arrays, but I think this would reduce the usefulness of the class. > xml parser - not sure if it could use the pear one... I can check it out. > extends Object (which only implements toString???) Well I don't know what their plans are here. > Looks like quite a bit of work.... Indeed it is. They are quite ahead of the game in terms of standard compatibility. They are listed on the W3C page etc. They said they can handle about 100k statements. The Java alternative can handle in the millions. But I think its quite an impressive feat as it is. Regards, Lukas