 Freedom of Information Conference 2000
Vicky Reich
HighWire Press
What do librarians want?
My assignment
for you folks today is to answer the question, "What do librarians
want?" In a nutshell what librarians want is freedom of information,
which completely side steps the issue of who pays for it. So don't argue
with me about that! And I'm going to talk to you from my experiences at
working for HighWire Press, which is a department of Stanford University
libraries. It is just reference and technical services. We work with non-profit
scientific societies.
The major
points I'm going to make have to do with market place competition and
choice, content, services that enhance research teaching and learning,
and then perpetual access. I'm not going to speculate much about what's
going to happen in the future.
Increased
competition means choice
From a librarian's
point of view there's decreased diversity and that's been happening through
recent mergers and acquisitions. This is a bad thing. There's been increased
diversity in the market place with initiatives such as BioMed Central,
BioOne, and SPARC,
for example. These are good things. At HighWire our role in the market
place has primarily been to help stabilise the diversity, to help non-profit
societies do well in the on-line environment. So my experience is working
with the highly cited and, therefore, frequently read primarily biomedical
but broadly science journal literature. HighWire works with about 65 publishers.
We have 206 titles on-line and add a few more titles every week.
On-line
is easier to access than paper
We have 275,000
full-text articles and if you add up the total number of abstracts and
full-text it comes to about 716,000. So how many individuals are using
this information? At least 3 million a month and the important point here
is that when I compare data from the year 1999, that number had grown
73% and the data we deliver, measured in terra-bites per month, is over
a terra-bite of data delivered to more than 3 million hosts per month.
That number has really jumped in a year. It's increased by more than 126%.
About half
of the journals we have on-line have three to five years of back files,
three eighths of the journals go back two years or less, and one eighth
have five or more years. This content is used and it is used a lot. If
you assume that when an article is published use is at 100% what you see
- and this is pretty constant across all titles - is that after three
months of publication, use is down to 13% of the original, and after 6
months it falls to just 7% of the original. As far as we can tell it stays
around 7% for a long time. This means that if you assume an average article
has 500 users per article per month, after it is half a year old there
are still 35 people a month wanting to read it. My point here is that
old on-line content is more useful than old paper content because it's
easier to access. And, while I don't have comparative data about how libraries
are used I have been a reference librarian for twenty-five years and,
believe me, on-line is used more.
As a librarian
I believe we need many systems and approaches to perpetual access because,
to be frank, we're going to be dead. Who the hell knows what's going to
work and I just don't want to leave it to chance.
What
we don't want - delays
This is a
talk about what librarians want and also what they don't want. What they
do not want is for content to be delayed for people getting access to
information through institutional subscriptions. This is not a widespread
practise but it is not uncommon either. Publishers tend to differentiate
between individual and institutional subscriptions but it is becoming
even more critical not to disenfranchise institutional readers. The time
between publications is shortening quickly. Shortening times between office
admission and publication, and manuscript acceptance and publication are
becoming simultaneous events and, to be blunt, the issue has become a
relic.
Services
enhancing research
We have two
services at that enhance research teaching and learning. The first service
is linking. We have approximately 2 million links from bibliographic references
to full text articles and references and it's absolutely true that people
link to literature, they don't necessarily search literature. About half
a million of the links are to free full text articles and when articles
are published on HighWire they are indexed in PubMed within 24 hours.
That is a phenomenal service to the community.
The second
service I want to talk to you about are alerts. We have three kinds of
learning services on HighWire, tables of contents, subject and author,
and new titles. We have about a million alerts that go to about a third
of a million readers and this database of alerts is growing at about 100
000 alerts a month.
Ensuring
perpetual access
The very
last point that I want to make, which is important to librarians, has
to do with perpetual access. I'm purposefully not calling this archiving
and I also want to make clear that this is not a HighWire project. This
is a project that I am involved in as a Stanford University Librarian.
The project
I'm involved with is called LOCKSS - Lots of Copies Keeps Stuff Safe.
This is a project funded by the NSF and the Stanford Libraries. LOCKSS
is a software protocol that will allow libraries, or anyone, to locally
store and manage web content. If it works it will be decentralised, distributed,
highly replicated, easy to use, and inexpensive to operate. The protocol
allows these locally held catches to compare the integrity of the content
and to self repair if any damage or loss occurs. So again, as a librarian,
what's important to me is that the content retains integrity, functionality,
and accessibility.
| 
