Response: Spotlight on BIDS

Clive responds to Jan's response.

> Clive, in his response to Isobel's Spotlight on BIDS,
> suggests that '...the fact that html forms make it much
> easier to specify searches which apply to several different
> field simultaneously ... may mean that the facility to
> recycle sets is not so important.'

I said MAY, not WILL. I'm not sure about this. At the moment I'm waiting for some metering to be put into the current service to try to find out a bit more about how it's used, eg how much use is made of "use previous sets". My research suggested that people didn't find it easy to understand what use previous sets did & therefore they tended to retype the search statement in a different way. The ones who did try to use it got it wrong.

> In the olden days when the likes of me were taught to
> construct search strategies,

TAUGHT? get real. how many people who use bids get taught to use it? as far as i can tell no more than 10% of users go on courses and this is probably an optimistic estimate.

> One of the main principles we learnt
> was to break down the subject of the search
> into its constituent parts, enter each part as a separate
> statement then use the building block approach to combine
> the sets to achieve the completed search.
>
> Despite today's more sophisticated and/or user friendly
> interfaces the principle still holds. Clive's example
> "find me refs which have fruit flies in the title AND are
> written by jones OR by bloggs" is fine as far as it goes.
> But what happens if the searcher mis-keys 'flies' as 'files'?
> The search produces no results and the user assumes nothing
> has been written. Split the search into separate steps,
> receive a null response to the 'fruit flies/files' section
> and even the most uncritical user may suspect a problem.

My research does not support this hypothesis. in my experience, even when people were told that a paper by a certain person was available, failing to find it didn't necessarily lead them to question whether their search was correctly formulated.

> When viewing the refs, the user notices the term 'drosophila'
> and wants to include it. Given the choice of searching on
> that term alone and combining the result with that of the
> first compilation or keying in (with correct syntax!) "find
> me... fruit flies OR drosophila in the title AND written by
> jones OR written by bloggs", I know which I'd go for.

Yes, of course, but unfortunately this wouldn't produce the right answer, since what you'd get is a search for:

(drosophila in title) OR (fruit flies in title AND written by jones OR written by bloggs)

Also, you wouldn't have to key it all in. you're making the mistake of thinking in telnet. you'd back up your browser to get the search form back and then edit the text on the form and resubmit it.

> So sorry, Clive, but for once I have to disagree. It had
> to happen one day :-)

I don't think this is going to be a real problem anyway. i'm no longer proposing to drop this facility unless implementation problems become really insuperable, and i no longer think that this is the case. our beautiful friendship can continue unabated.......

best wishes
clive

Contents Page - Electronic Libraries Programme and Project Information
Search Ariadne - Editorial for this issue - News desk - Reviews - Front Cover