On Jan 25, 2008, at 3:47 PM, Kyle Banerjee wrote:
> To satisfy an information need, you need to know a little about the
> context of the need. If you know a little about who is asking a
> question and why, you're much more likely to give a useful answer. Our
> obsession with making sure we have as little context as possible to
> work with is why everyone wonders why our services aren't keeping pace
> with the growing number of alternatives our users have.
Above echoes my perspective. In our increasingly networked
environment it is less easy to understand the context of a library
patron. They aren't standing right in front of you so you can't
estimate their age, status, nor expertise. We rarely record or
remember anything about them so you can't build up a history of
questions/usage. We get little feedback whether or not particular
solutions were useful or helpful.
It is for these reasons, plus rising user expectations, that I
advocate the "my" approach to library Web services. [1] More
specifically, in an academic library setting, I advocate building a
relationship with the college/university's registrar and human
resources departments. Next, get regular dumps from their rosters
including names, email address, departments, status, majors, classes
taught, classes taken, etc. (You are probably already getting much of
this data already to feed to your ILS circulation module.)
Once you get this information use the patron their characteristics as
the basis for classify your content. Departments, majors, classes
taught, and classes taken can all map to library subjects. Subjects
are associated with library resources (indexes, databases, etc.).
When a person opts to sign into your website you can look up their
name and display recommended resources based on who they are and
their particular characteristics. A website could then allow people
to add and subtract from these recommendations. Now that a person has
a set of recommendations, you can find others (freshman or people in
English 101) and answer the question, people with some of your
characteristics also have this. Once you have recommendations the
system can count the number of times particular things are used. Once
these tallies are calculated you can say what is popular (or not).
Along the way allow people to review, link, blog, tag, etc.
All of this relies on knowing more about the patron. Who are they?
What characteristics do they have? Library systems need to be about
more than just stuff in libraries. They need to be about creating
relationships between stuff and users.
[1] BTW, in the most recent (January 2008) issue of College &
Research Libraries there is an article called Engaging Users: The
Future Of Academic Library Web Sites by Shu Liu. It mentions the "my
library" concept a number of times and says library Web sites should:
1) switch the focus to... targeted and customizable tools, 2) design
a seamless, one-stop information environment..., and 3)... give users
opportunities to express, share, and learn.
--
Eric Lease Morgan
University Libraries of Notre Dame
Received on Fri Jan 25 2008 - 17:24:14 EST