Michael Kay wrote:
>>How do you know it's 98% conformant?
>
> Of course any number is arm-waving. If one had a set of conformance tests,
> one could say how many of the tests are passed; but even then that doesn't
> exactly correlate to how much of the specification has been implemented.
So I take it one doesn't have a set of conformance tests? The 98% number
is entirely subjective, or rather, expert opinion?
If so, short of soliciting your expert opinion on the several other
schema processors out there, it's not obvious how to reconcile Henry
Thompson's assertion that the conformance problem is largely solved with
the contradictory evidence given by Rick Jelliffe and others.
Bob Foster
