RE: conflict resolution issue

>The trouble about this is that there are many cases where it's obvious to
>the stylesheet author that one match pattern is more selective than
another,
>but where it requires a fair bit of analysis to deduce this. For example

>match="para[p and q]"
>  implies
>match="para[p[1] or q[last()]]"
>  but does not imply
>match="para[p[2] or q[last()-1]]"

Perhaps, but the case of 'use the rule with the most predicates' (e.g.
match="a/b[c]" overrides match="a/b" would be fairly easy to implement (I
assume).
Cheers,
Matt