W3C home > Mailing lists > Public > www-zig@w3.org > September 2004

RE: Z39.50 Searching

From: LeVan,Ralph <levan@oclc.org>
Date: Wed, 15 Sep 2004 14:14:49 -0400
Message-ID: <E5431CF93E29F9478878F623E5B9CE9809DF7771@oa3-server.oa.oclc.org>
To: Heiko Jansen <jansen@hbz-nrw.de>, www-zig@w3.org

It's not obvious to me why you get these different results.  The 4=6
(Structure = WordList) does not mean that they must come from the same
field.  There is an implicit operator between the words in the list.
Perhaps it isn't simply AND for that database.

An experiment would be to repeat the search, omitting attribute types 2, 3
and 5, since they were also omitted from the ANDed search.

If you still get different results, I'd contact the database provider.

Hope this helps!

Ralph


> -----Original Message-----
> From: Heiko Jansen [mailto:jansen@hbz-nrw.de]
> Sent: Wednesday, September 15, 2004 12:52 PM
> To: www-zig@w3.org
> Subject: Z39.50 Searching
> 
> 
> Dear all.
> 
> Correctly formulating a Z39.50 query that means what I think it means
> seems to
> be a difficult task ;-)
> 
> I was searching a certain database for documents having the 3 words
> "sicherheit", "internet", and "firewalls" in their "title" and found one
> hit
> with this query:
> @attr 1=4 @attr 4=6 @attr 2=3 @attr 3=3 @attr 5=100 @attr 6=1 "sicherheit
> internet firewalls"
> 
> The title was sth. like "Firewalls und Sicherheit im Internet." Later on I
> found another title "Einrichten von Internet firewalls : Sicherheit im
> Internet gewährleisten" using a different search tool. At first I thought
> that the servers index no. 4 did not contain the remainder of the title,
> but
> I could find the additional title (together with the one found by the
> first
> query) with this query:
> @and @and @attr 1=4 "internet" @attr 1=4 "firewalls" @attr 1=4
> "sicherheit"
> 
> Now when I asked one of the developers of the server software why that was
> the
> case he told me that the "attr 4=6" in the first query made the
> difference,
> indicating that all words from the "word list" must occur in the same
> "field".
> 
> I have some questions on that with the most basic one being this: What do
> the
> expressions "field" and "subfield" mean concerning bib1 queries?
> Since MAB puts title and remainder in different "fields" 331/335 while
> USMARC
> puts them in the same "field" 245 but different subfields $a/$b and a
> server
> probably supports both record syntaxes it would make no sense to correlate
> the two applications of the terms because there´s no way of knowing which
> one
> applies?!
> In my eyes it would make even less sense if the "fields/subfields" from
> the
> standard would apply to any internal field of the database (because the
> user
> usually has no idea of them).
> 
> If "field" means "index" ("use"-Attrib.) here, shouldn´t then the first
> query
> yield all hits from both queries mentioned above? There is attr 3=3
> (Position
> = any pos. in field) which IMHO should apply to all words of the word list
> separately.
> If "field" means only a part of the index: what would then be the correct
> result?
> 
> Puzzled greetings
> Heiko
> --
> ________________________________________________
> Dipl.-Bibl. Heiko Jansen
> Hochschulbibliothekszentrum NRW (HBZ)
> Postfach 27 04 51, 50510 Koeln, Deutschland
> Telefon +49-221-40075-150, Fax +49-221-40075-190
> Juelicher Str. 6, 50674 Koeln, Deutschland
> jansen@hbz-nrw.de
> GPG Fingerprint:
> F65F BA22 2833 C196 440B  B766 2BD1 41D5 13D4 6609
Received on Wednesday, 15 September 2004 18:14:51 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 29 October 2009 06:12:23 GMT