On Mon, 6 Sep 2004, Theo van Veen wrote:
> Given this discussion I think it helps that whatever we decide on this
> subject, it might be useful to specify a few default index sets and try
> to avoid identical names in the different index sets (for as far as we
> have control). I can imagine that we use dc, rec and srw as default. We
> may add dcterms for qualified dublin core names and cld for names from
> collection level descriptions. Only when a name is used from a different
> context set the prefix needs to be specified.
(a) We already have identical index names in dc and rec, eg language and
identifier.
(b) It would be a nightmare for parsers to -have- to know about the
context sets.
(c) CQL can be used outside of SRW, so we shouldn't specify a rule that
references anything other than absolutely essential stuff (eg cql.)
and to counter the above
(d) It's actually what we have already, as if you leave off the prefix
and don't specify a default in the query, then the server gets to
decide what to do with the index name. So if you want to do what
you've suggested, then go right ahead.
"An index name always includes a base name and may also include a prefix,
the name of the context set of which the index is a part. If not supplied,
the default context is determined by the server."
Rob
,'/:. Dr Robert Sanderson ([log in to unmask])
,'-/::::. http://www.o-r-g.org/~azaroth/
,'--/::(@)::. Special Collections and Archives, extension 3142
,'---/::::::::::. University of Liverpool
____/:::::::::::::. L5R Shop: http://www.cardsnotwords.com/
I L L U M I N A T I
|