> My SRU client interface WILL receive queries such as the following --
> people WILL enter queries such as these:
>   * foo bar
>   * repetitive task
>   * virtual libraries
>   * International Technology Education Association

Just my 2 cents here: I think it is a misconception to build a SRU interface
that a user will be directly interacting with. It is my understanding that
SRU is a machine API, for programs to use, not for people to directly
interact with...unless they are just goofing around with a GET string in
their browser.

In the above situation I would put a layer in between the the user form
which submits the data, and the operation performing the SRU/CQL search.
This layer would receive the user input, massage accordingly, talk to the
SRU service, and return the results.

//Ed