On 4/12/06, K.G. Schneider <[log in to unmask]> wrote:
> Do users actually "determine relevance" or do they have faith in Google to
> provide the best results on the first results page?
I'd say people use a "click and try n times, before refine search
until relevance is fulfilled" technique. But again, this is *totally*
dependant on what they're searching for; known or unknown ;
- books by Frank Herbert (specific enough to get some results)
- Jungs philosophy in fiction (general enough to cause bleeds)
- good SciFi (general enough to cause bleeding)
- oil crisis metaphors (specific and general at the same time)
All of the above can lead to "Dune" by Frank Herbert. What is it's
relevance to the above searches? It's a book by Herbert, it certainly
contains Jungs philosophy, it's a good SciFi book, and has indeed the
metaphors as part of its concept. And to top it all, it's still a
popular book. So I could say "The Dosadi Experiment" and all the same
is true, except the popularity. Who is to say that former is preferred
over the latter? Google will give us the former, never the latter.
For libraries, this is an interesting problem to solve, because
popularity, at least in my view, is mostly a misnomer in searching for
information. Popularity in Google is measured by people actually
putting in the links, which means they point to something *because*
there is something interesting that way. In the library catalogs there
is no such thing.
We've got an experiment running here which uses tags to do this last
bit for us; people and librarians alike can tag books which will boost
their ratings. An anonymous tag denotes popularity (unless stated
otherwise), while a reference librarian boosts importance. Another
fields I'm digging into is using search term logs to do some of this
as well, generating "heat" for items ... close to popular, but can be
very time-based (unlike links which stays around) if you don't feed
the flame, it eventually will die out (or in this case, repurposed).
Anyways, just a few thoughts and ideas.
Alex
--
"Ultimately, all things are known because you want to believe you know."
- Frank Herbert
__ http://shelter.nu/ __________________________________________________
