[WEB4LIB] Re: NOTAmazon OR Google as a "corporate" role model: iTunes

[Steve Morris]

> Agreed.  Browsing by genre, subject, format, intended audience, and any
> other facet available (or a combination of these) is very useful.
> Browsing as a result of a search is even better.

That last item is the key point.  The user wants that single little search
box where they can type things in, to assert control over the experience. 
Libraries (and librarians) are good at organizing things, setting up
browse situations, etc.  We have to meet the user half-way--let them do
that quick search they so much want to do, and then try to fish them out
of their failed searches into our structure.  Try to respond with results,
but also do ‘search characterization’ and present resources, gateways, and
specialized starting points contextually.  Stats often show ‘Subject
guides’ are a non-starter with users, but in a failed search a contextual
pointer to ‘entomology resources’ might get some interest—just need try to
make those browse/portal tools seem tasty and attainable and available in
context.

We need to start interacting with the user in the search space they way we
do at the reference desk--learn from failure.  For example, if 8 users
come to the desk in one day asking for the OED dictionary set, we show
where the set is, but we also put the data about user failure to use by
improving signage, putting the resource more directly in the path of the
user, etc.  We need to get into the logs and start responding to user
failure in the search space by building ancillary indexes that allow us to
respond with something.

One issue: to the extent that we provide the user with a _single_ search
box, we should expect that they are searching not just for content, but
also for services, facilities, support, tools, expertise.  Our logs show
us this.  Google vanilla doesn't separate these things out, why should the
user expect we would.  25 years ago or whenever it was, library automation
consisted of taking those cabinets full of cards and putting them in the
computer, and then those serials indexes--but the library was always more
than the books and serials: photocopy machines, study rooms, media viewing
equipment, students tracking down study mates, etc.  Now with the advent
of information commons and the proliferation of specialized services like
digital media labs and GIS labs, facilities and services are even more of
an issue.  And to complicate, for the edu’s the line between library
facilities/services and campus resources is blurring (same for public
libraries vis-à-vis government agencies).  Any single search system
focusing just on content will be necessarily limited in scope.

Steve Morris
Head, Digital Library Initiatives
North Carolina State University Libraries
Steven_Morris at ncsu.edu
ph: (919) 515-1361