Casey,
I would like to suggest a rethinking to make your wonderful catalog even better. It is important for everyone, including librarians and I think especially catalogers, to think outside the box, or in this case, outside the catalog. The very concept of the "library collection" is changing and includes things that could never have been included before, plus there is the ability to include more traditional library materials into a catalog search that were ignored before.
So, when someone gets a zero search, where could they search where they would not get a zero? If you redirect their search to, e.g. Google Books, or the Internet Archive, or Intute, or Google Scholar, or whatever, you are helping users more easily to search for materials that are readily available. This is how I have tried to work it in my catalog, where I have a tiny collection, but I leverage what is on the web.
For example, my catalog has nothing on "Nepal history"
http://www.galileo.aur.it/cgi-bin/koha/opac-search.pl?op=do_search&marclist=bibliosubject.subject&operator=contains&value=nepal history
but when the users get zero, they are presented with a choice to extend their search. When you click this, you enter a new area, where you can search materials arranged in different groups that I have selected. These could be reorganized in all kinds of ways. I also try to add plenty of help. I think it's pretty clear what I am trying to do (however unsuccessfully), and I am sure that others can see how the whole thing can be improved. (I have already had some suggestions that I am trying to include) In any case, my users quickly and easily discover that they have all kinds of things relating to the history of Nepal.
What I am suggesting is that you think boldly. When somebody gets "zero" today it's rarely true in reality, and in fact it should be a great opportunity to show what is *really available* so long as we reconsider what our "collection" is (ie. library-selected areas of the WWW) and then we do our best to give access to that library-selected area in as coherent and reliable way as possible. The users have already redefined "the collection" in their own minds, we need to do so as well.
I hope this makes sense and I hope you consider adding the subjects to your catalog.
Jim Weinheimer
> Jim,
>
> The controlled vocabulary of library data is part of what makes the
> search autocomplete work, in my opinion.
>
> My example just uses the data from the cataloged records. I very much
> like your suggestion that it follow the subject authorities when
> offering suggestions. Michael Klein of BPL demonstrated a very good
> authority-based subject autocomplete at a New England code4lib a few
> months ago, and I've been hoping to implement something like that ever
> since.
>
> The difficulty so far is that few collections are as expansive as the
> thesauri they draw their metadata from, and coordinating the authority
> data with the actual contents of that collection is necessary to avoid
> offering dead-end suggestions. That, and there aren't many sources of
> electronic authority data.
>
> --Casey
>
>
> On Feb 19, 2009, at 9:31 AM, Weinheimer Jim wrote:
>
> > I like this a lot. It seems even to work with authority files, so
> > for example, if I enter chaikovskii it will give me: tchaikovsky,
> > peter ilich, 1840-1893.
> >
> > When I enter "labor and laboring classes" I get two hits, when I
> > should get "working class" and "labor movement" and
> maybe some items
> > from the time before the heading changed. But that would be more
> > sophisticated.
> >
> > Very nice!
> >
> > Jim Weinheimer
> >
> >> Casey Bisson wrote:
> >> > The autocomplete that I implemented in Scriblio
> searches across
> >> title,
> >> > author, subject (and others), and suggests possible
> matches in
> >> any of
> >> A0> them. You can see it here:
> >> >
> >> > http://library.plymouth.edu/
> >> >
> >> > I've been skeptical of the value of autocomplete
> (there are a
> >> number
> >> > signs that suggest that it slows down the user
> experience or leads
> >> > them to prematurely narrow their searches), but I've
> been
> >> pleased at
> >> > how it works against library data. I'm especially
> happy at how it
> >> > allows users to begin a search without first having to
> select
> >> which
> >> > field to search.
> >> >
> >> > --Casey
> >> >
> >> > On Feb 18, 2009, at 12:31 PM, Mike Cunningham wrote:
> >> >
> >> >> I thought someone on this list might be interested
> in this. I now
> >> >> have a proof of concept running on our staging
> port of an
> >> >> autocomplete/live search feature for the opac. You
> may have
> >> seen this
> >> >> type of feature on sites like Amazon or Zip.ca. It
> comes back
> >> with
> >> >> different results depending on which search field
> the user
> >> selects
> >> >> (title, keyword, author, subject, call number).
> Since there is no
> >> >> keyword index per se, keyword also uses the title
> index. The
> >> feature
> >> >> kicks in after the user has typed at least 3
> characters.
> >> >>
> >> >> I'm curious if anyone knows of other library
> catalogues that do
> >> >> something like this. Someone pointed out to me that
> >> BiblioCommons has
> >> >> a feature like this. Does anyone know of others?
> >> >>
> >> >> You can try it out here:
> >> >>
> >> >> http://search.cambridgelibraries.ca:2082/search
> >> >>
> >> >> There is some background info on the feature here:
> >> >>
> >> >> http://ex-libris.ca/?p=694
> >> >>
> >> >> Mike
> >> >>
> >> >> --
> >> >> Mike Cunningham
> >> >> Web Services Librarian
> >> >> Cambridge Libraries
> >> >
> >> >
> >>
> >>
> >> --
> >> -----------------------------------
> >> Karen Coyle / Digital Library Consultant
> >> kcoyle@kcoyle.net http://www.kcoyle.net
> >> ph.: 510-540-7596 skype: kcoylenet
> >> fx.: 510-848-3913
> >> mo.: 510-435-8234
> >> ------------------------------------
Received on Fri Feb 20 2009 - 03:08:16 EST