>   http://listserv.loc.gov/cgi-bin/wa?A2=ind0501&L=ZNG&D=0&I=-3&P=4512
> 
> If you have half an hour, read through the rest of the thread 
> as well which is enlightening, IMO.

Wow. I vaguely remember this thread, but I didn't care enough to
participate at the time. I really don't want to re-open a discussion
that has been so thoroughly debated, but I think we need to have some
closure on it. Allow me to summarize:

There was some opinion (primarily by Matthew Dovey) that
cql.serverChoice should be used only as a proxy for some other index
that the client would normally use. The opposing opinion (primarily by
Rob Sanderson and Mike Taylor) was that cql.serverChoice can do whatever
it needs to be useful, regardless of other existing indexes.
Unfortunately, the thread somewhat dies out after Rob makes this
statement (Jan 10, 2005):

> Okay, then my opinion is that the serverChoice definition 
> is incorrect and needs fixing.

If we follow Matthew's interpretation (and the current wording for the
definition of cql.serverChoice/cql.anywhere), a new keyword index would
be very useful. But I agree with Rob and Mike that cql.serverChoice is
perfectly fine IF we modify the definitions to make it clear that
cql.serverChoice can serve this function. 

In sitting down to construct a server for the first time, it was obvious
to me that cql.serverChoice should be used for my keyword search. I just
had this nagging feeling that I was doing something wrong, because what
I was doing didn't quite fit the definition as written. 

---- Ryan Scherle
---- Digital Library Program
---- Indiana University