Hi Ivan,
Thanks for the response!
On Tuesday 02 September 2008 14:23:05 you wrote:
> One can not
> expect that text from user's input can be placed into query verbatim.
Yeah, we have a pre-processing level that takes care of that.
> Re.
> DESCRIBE ?resource WHERE {
> ?resource dct:title ?free ;
> dct:subject ?var .
> ?var skos:prefLabel ?free .
> ?free bif:contains "Da*" .
> },
>
> there are two errors.
Ah, yes, it was just an example. In practise, we have some OPTIONALs and a
FILTER bound( thing that takes care of all these things.
> The extension like
>
> DESCRIBE ?resource WHERE {
> ?resource bi:any-contains "Foo* and Da*" .
> }
>
> is not very convenient for mixed storage,
Yes, I know, this is a very search-engine-y thing, not at all a database-y
thing, but nonetheless a requirement...
> so I'm not sure it will ever
> appear in the schedule. It may be more practical to materialize some
> view and create either separate graph for the materialization or a
> separate predicate like sub:literals. Depending on needs of an
> application, the view may store concatenations of all literals of a
> subject or a list of short items or a list of concatenated literals that
> are grouped by some criteria.
Right! It requires more work, but OK.
I suspect that part of the performance problem we now have, which may not be
solved by simply using Virtuoso, is that the cost of serializing and
transferring the result set is too high. In that case, I was wondering if I
could somehow exclude the sub:literals predicate from the DESCRIBE result?
Kind regards
Kjetil Kjernsmo
--
Senior Knowledge Engineer
Direct: +47 6783 1136 | Mobile: +47 986 48 234
Email: kjetil.kjern...@computas.com
Web: http://www.computas.com/
| SHARE YOUR KNOWLEDGE |
Computas AS Vollsveien 9, PO Box 482, N-1327 Lysaker | Phone:+47 6783 1000 |
Fax:+47 6783 1001
