> Date: Tue, 9 Dec 2003 13:27:40 +0000
> From: Robert Sanderson <[log in to unmask]>
> > dc.title <>/stem monies
> > when using the Bib-1 attribute set would be an error (???).
> Unless you can turn the <> relation into a not boolean.
I am not convinced that a well-behaved implemetation has licence do
au=kernighan and ti <> unix
may be a subtly different query from
au=kernighan not ti=unix
(Example: if the back-end works by querying an SQL database, the
latter query will find records with author=kernighan and title
undefined, whereas the former will not.)
> > Finally, exact, any, and all are interesting cases. There is no
> > 'any' in Bib-1 (unless I have an old printout). 'all' I guess is
> > 'word list'.
I don't think it was ever the intention to use "word list" for this.
I seem to remember there was a proposal to add explicit "any of these
words" and "all these words" attributes to the BIB-1 attribute set,
but I can't find any mention of them in
Am I imagining things? Or what happened to that proposal?
When using the attribute architecture, of course, there are
format/structure attributes for these concepts: @attr 9=3 and @attr
9=2 respectively. Although that, too, was up in the air for a long
time, wasn't it? Does what's current on the site at
represent the outcome of that discussion, or did it run out of steam
so that what's there now is what was always there?
> For any, I turn the clause into a tree of OR clauses in bib1, AND
> clauses for all. Yes, this isn't quite correct, as the client will
> likely have different 'word' extraction to the server.
Yup. I think we all agree that this is The Wrong Thing, but it seems
to be the right thing to do anyway :-)
/o ) \/ Mike Taylor <[log in to unmask]> http://www.miketaylor.org.uk
)_v__/\ "... about as similar as two completely dissimilar things
in a pod" -- Black Adder.
Listen to my wife's new CD of kids' music, _Child's Play_, at