Re: [Koha] Do you like the way Koha is sorting facets and facets results ?
Looking at the Perl script that generates Koha facets, in my opinion they are not good and... cannot be better :-) I found a very good solution in the Solr demo of BibLibre. Try http://catalogue.solr-demo.biblibre.com/cgi-bin/koha/opac-search.pl?q=*%3A*&filters= that retrieves all records in the catalog. I think that facets cannot be generated but the indexer itself. Don't know if Zebra can generate them. Hope yopu find a solution. Stefano
Hello Could some librairians and vendors tell me if they like the way Koha is sorting facets and facets results, on a UX point of view ?
Currently, - Results for each facets are sorted alphabeticly. - I can't explain the way factets (Authors, Title, Topics...) are sorted, maybe by popularity, but it is not always the same, depending of the request !
If we look at other catalogues, or non library websites (Amazon...), we don't see the same behavior. - Facets sorting does not change in the same catalogue - Facets results are sorted according to popularity (decrescent occurences), or sometime in a predefined order, sometime both, but never alphabeticly. (in amazon.com , Author and Book series are sorted according to popularity order, but Departement and Format facets in a predefined order. cf. http://www.amazon.com/s/ref=nb_sb_noss_2?url=search-alias%3Dstripbooks&field-keywords=game in worldcat, Content and Public in a predefined order, and the others by popularity : http://www.worldcat.org/search?qt=worldcat_org_all&q=proust )
If you agree, I can make a patch for sorting facets results by popularity (easy to do) and sorting facets according to a defined order (more complicated for me but I will try...).
Regards, M. Saby Rennes 2 University
Stefano and Mathieu, Looking at the Perl script that generates Koha facets, in my opinion they
are not good and... cannot be better :-) I found a very good solution in the Solr demo of BibLibre. Try
http://catalogue.solr-demo.biblibre.com/cgi-bin/koha/opac-search.pl?q=*%3A*&filters= that retrieves all records in the catalog. I think that facets cannot be generated but the indexer itself. Don't know if Zebra can generate them. Hope yopu find a solution.
The in-Koha faceting is, indeed, dreadful. Zebra can generate facets for CHR indexes, which leaves us with a problem for ICU. However, I think there may be a workaround: using key (i.e. untranslated) indexes for facet building. I started to fiddle with this last week but did not get very far yet. Better faceting is one of my goals for the Search rewrite ( http://wiki.koha-community.org/wiki/C_&_P_Search_Rewrite_RFC). QueryParser has a special facet syntax which I look forward to using (eventually) to allow us to add and remove facets from searches. Regards, Jared -- Jared Camins-Esakov Bibliographer, C & P Bibliography Services, LLC (phone) +1 (917) 727-3445 (e-mail) jcamins@cpbibliography.com (web) http://www.cpbibliography.com/
participants (2)
-
Jared Camins-Esakov -
Stefano Bargioni