Re: Control of concurrent users

[Don Wassink <lbdonw@ust.hk>]

It's true that such statistics are not reliable for journals--but they
do look reasonable for databases.  Frequently a database will require a
login as well as doing IP address checking, and patrons have to go
through the library web page to connect to those databases anyway.  The
logon is scripted, so it is transparent to the patron.  But if they
bookmark the database, it wouldn't work the next time.  Hopefully, our
patrons have learned to bookmark the library database list for all their
databases.  It seems to me, that in general, someone is more likely to
bookmark a favorite journal, than they are a database.

Don Wassink
Hong Kong University of Science & Technology

________
David Goodman wrote:
> 
> Don, don't many of your users access the journal without going through
> your web page-- by using links from other journals or from databases like
> Pubmed and Inspec, or by going directly to the publisher's site?
> 
> David Goodman
> Biology Librarian, and
> Co-Chair, Electronic Journals Task Force
> Princeton University Library
> dgoodman@princeton.edu         http://www.princeton.edu/~biolib/
> phone: 609-258-3235            fax: 609-258-2627
> 
>  --
> 
> Don Wassink wrote:
> >
> > I would not interpret 'all resonable endeavours' to include having the
> > library control the number of concurrent users.  That is unreasonable, and
> > we've never met a vendor with such a requirement.  If your vendor truly
> > requires that, then I would start looking for an acceptable alternative.
> >
> > We routinely and automatically keep statistics on the 'hits' on our
> > web-server, including individual pages, and how many times the script is
> > used to access any particular database.  (I guess there are various
> > software programs that will do this.)  This is useful from a collection
> > development point-of-view, but it can also be useful for negotiating a
> > concurrent user license with a vendor.  If, for example, you can show that
> > a database is accessed on average only 60 times per month, then you can
> > easily argue for a single user license.  Any reputable vendor should
> > recognize the merit of such an argument, or again, time to look somewhere
> > else.
> >
> > Don Wassink