> >More seriously, we are straying onto metasearch requirements 
> territory
> >here. I suspect that one of their requirements and a reason I suspect
> >that might give for not considering SRU/SRW is that they do expect a
> >single string to work on an arbitrary list of servers!
> 
> Surely you mean that they *don't* expect this?

No - I meant "do" since as you go on to say

> if the metasearch
> engine builders seem unimpressed by efforts like SRW, I'm 
> guessing it's
> because they have long-since resigned to the fact that they 
> not only have
> to generate different queries across servers, in many cases 
> they will have
> to submit those queries through completely different 
> protocols

I can quite imagine (especially from those new to the field) a group of
metasearch developed deciding that SRU/SRW is unsuitable because it
doesn't address this and going off and try to develop something that
does. I can remember in the UK fighting a sometimes uphill battle trying
to explain that Z39.50 wasn't fundamentally broken and useless because
it didn't address this!

Matthew