Re: A simple test of XPath filter performance

I've added a row in the xpath-filter2 interop matrix and indicate that 
Merlin satisfies the requirement [1]. Christian and Aleksey, can you send 
me a guestimate too? While I completely agree with John on the desirability 
of this result and I think we should push this as far as we can, I don't 
think we can make it a MUST nor am I confident about specifying 
implementation specific type optimizations. I hope that we will *soon* (if 
we don't already) have a spec that is capable of satisfying our 
requirements and it would be better to make it available to folks sooner 
than later given even slow implementations of this are vastly superior to 
the existing specification algorithm.

[1] http://www.w3.org/Signature/2002/02/01-exc-c14n-interop.html
   performance: signature generation and verification over
     sign-xfdl.xml: in 500ms on ~700 bogomips machine. SHOULD