IRC log for #koha, 2006-09-11

All times shown according to UTC.

Time S Nick Message
10:40 fr0zen i'm having some problems with z3950 daemon.  i run it and it immediately ends.  the log says "Bareword "Net::Z3950::RecordSyntax::USMARC" not allowed while "strict subs" in use at /usr/local/koha/intranet/script​s/z3950daemon/processz3950queue line 260.
10:40 fr0zen Bareword "Net::Z3950::RecordSyntax::UNIMARC" not allowed while "strict subs" in use at /usr/local/koha/intranet/script​s/z3950daemon/processz3950queue line 261.
10:40 fr0zen Execution of /usr/local/koha/intranet/script​s/z3950daemon/processz3950queue aborted due to compilation errors."
11:35 thd kados: are you there?
11:43 kados hi
11:46 thd kados: so it is easy to see how the relations work for Zebra 2 Alvin XML indexing where you are at most dealing with a trans-adaptation work concept and related biblios and authorities in the meta-bibliographic record.
11:47 thd kados: where it becomes just a little trickier is for indexing authority hierarchies themselves.
11:48 kados why do we need to index authority hierarchies?
11:49 thd kados: the problem there is that depth is arbitrary and I have not yet formulated a good design for arbitrary depth outside of Zope which uses OOPS inheritance to manage that.
11:50 thd kados: so that we can include or exclude large parts of a hierarchical subject thesaurus or classification system at one time
11:50 thd kados: for efficient searching
11:51 thd kados: or browsing hierarchical trees
11:51 kados perhaps an example?
11:53 thd kados: suppose we have hierarchical place names in our thesaurus and we want to include all North American place names in our search but Exclude all European place names.
11:55 thd kados: So we want to avoid records from Athens, Greece and Berlin, Germany while including Athens, Ohio and Berlin, Pennsylvania if there is such a Berlin.
11:56 thd kados: the point here is not about possible similar names, that is just an added Benefit
11:58 thd kados: the point is with two specifications one can include and exclude a very large number of terms where one has an understood hierarchical relationship with a small number of values without enumerating every child element such as every place name in North America or every place name in Europe
11:59 thd kados: does that make sense?

| Channels | #koha index | Today | | Search | Google Search | Plain-Text | plain, newest first | summary