RE: Why no namespace node KindTest?

> > One of the things I learnt from James was that when someone argues 
> > against a specification on the grounds that it is impossible to 
> > implement efficiently, they are usually wrong.
> 
> Could you elaborate? E.g, why is implementation efficiency, 
> usually, not a valid reason?

Because if the spec is clean and well-defined, then you can usually find a
way of implementing it efficiently. You shouldn't do the design until you
have got a specification, and if your design gets in the way of a clean
specification then it's probably not a good design anyway.

That's not to say it's always wrong to consider implementation factors, for
example the restriction in XSLT that you can't create attributes for an
element after creating its children is probably justified. I only said
"usually"!

Michael Kay