[RDF Interfaces] names for things

Ivan Herman wrote:
> On May 3, 2011, at 18:30 , Nathan wrote:
>> A couple of issues to touch on:

Next up.. having some collisions on reserved words in languages:

- import, have talked it through with a few people and currently 
thinking addAll is about the best alternative (overloaded on graph to 
accept a sequence/array of triples too, which is nice), another 
alternative 'extend' : +/-?

- property, Triple.property, turns out it's heavily used in some 
languages too, Ivan you'll recognize from python (?) proposal is either 
'predicate' -or- i quite like the idea of having Triple.s,p,o instead, 
do we need to be verbose?

- RDFNode.value - still unsure..


and a different issue:

can graph.match return a sequence of triples rather than a Graph? 
basically so it's lighter, especially w/ indexed graph implementations. 
My personal gut instinct is to keep the return type of Graph to prevent 
unexpected functionality, and since pretty much half the time you want a 
quick sequence to output or suchlike, and the other half of the time you 
want a Graph to keep on using for some purpose.

trying to nail as many as these minors before FPWD as possible!

Best,

Nathan

Received on Tuesday, 3 May 2011 18:27:09 UTC