> We *know* the dirt-simple MS API works (sorta); I think we should > get just a > LITTLE more ambitious in the first iteration, then keep getting more and > more complex/conformant/powerful until we come to the point of diminishing > returns. > > So, I'd pose the question: "What's the next step beyond > selectNodes() that > we can agree on as useful? Is there some tweak for namespace > awareness that > we can add ... or should we address the context issue first, the variable > binding issue first, or what? > > Is anyone comfortable with this approach? > Yes, definitely. I think the namespace issue is the most critical out the door. I would vote for addressing that first. -aaronReceived on Friday, 5 May 2000 17:31:49 GMT
This archive was generated by hypermail 2.2.0 + w3c-0.30 : Friday, 25 March 2005 11:19:25 GMT