Creating an Internal Content Management
By Greg Sennema
Here at Calvin College, we created Hobbes, a hybrid of CMS and intranet that
organizes Web site content and a variety of internal tools to help our librarians
complete their daily tasks.
When I became digital resources and reference/instruction librarian in 1999
at the Hekman Library at Calvin College and Calvin Theological Seminary, one
of my first tasks was to redesign the library's Web site. Like many of you
who maintain Web links to library resources, I quickly learned that offering
access to a large number of resources from an even larger number of static
Web pages can never really be described as efficient or professionally fulfilling.
In recent years, many libraries have created dynamic Web sites that are database-driven,
that is, maintained and populated with content stored within what can be loosely
termed a content management system (CMS). In 2001, Jed Koops, the Hekman Library
systems programmer, and I created a CMS called Hobbes (library staff liked
the play on the words "Calvin and Hobbes"). Since then, Hobbes has grown into
a hybrid of CMS and intranet to include not only Web site content, but also
a variety of internal tools used by librarians to help them complete some of
their daily tasks.
Hekman is a small academic library serving more than 4,200 students, 370
faculty, and many community patrons. The majority of the library's 1.7 million
items (books, e-books, journals, e-journals, microfiche, government documents,
etc.) are spread across 100,000 square feet over four floors. Hekman employs
the full-time equivalent of 17 staff, including eight professional librarians
who also serve as liaisons to all of the college's 25 academic departments.
Access to all library resources is provided through the online catalog (Sirsi),
a digital media archive (Sirsi), and our Web site, the Hekman Digital Library
(HDL). Since its creation in 1997, the HDL has grown from a mundane list of
links and library information to the primary point of access to all of the
library's resources. In 2001, I created the current version of the HDL.
Calvin and Hobbes: Managing Our Content Online
Hobbes is a Web-based tool that uses Common Gateway Interface (CGI) scripts
written in Perl to store, query, and return results from data stored in a series
of related SQL tables. Because it is Web-based, librarians can easily manage
data without knowing Perl or SQL and can be authenticated into the system using
the college's existing online directory. By authenticating through a continually
updated college database of login names and passwords, we did not need to create
our own. Users added to the Hobbes environment have immediate access without
needing to remember yet another ID and password.
As a CMS, I initially envisioned Hobbes as a tool to help me to efficiently
manage the library's Web pages, specifically the growing number of subscription-based
resources. In the months after Jed and I implemented Hobbes, we realized that
it could be used to host tools that served the library's internal communication
needs as well. Over the course of 2 years, we added tools as they were thought
of and created, and placed them according to their function under tabs that
are strung along the top of the Hobbes environment (see Figure
1). Each librarian
is given access to the tabs and tools she needs based on her profile within
Hobbes. Here I will describe each tab and some of the major tools (or tables)
placed within it. With each description, I have included our label in parentheses
to help you compare what you are reading to the screen shots. To see how some
of Hobbes' resources are displayed to the public, please visit the HDL at http://www.calvin.edu/library.
Manager Tab (Manager)
The Manager tab is used primarily by Jed and me to administer the contents
within Hobbes. Since there are quite a few tools under the Manager tab (see
Figure 1), I'll briefly explain a number of these before moving on to the functions
of the other tabs.
Databases (TofDbs): Each research database record stored in a table of databases
(TofDbs) contains the following fields that allow the resources to be displayed
on the HDL in a variety of helpful ways:
AlphabeticalBased on the first letter; can also include
See Also references
TypeCitation index, abstracts, full text
AuthenticationWhether the resource is accessed through
IP authentication, for free on the Web, on a CD-ROM, or through
a print index
Full-text typeDescribes a database's full-text content,
such as journal or newspaper articles, e-books, or even primary source
VendorMost useful for our librarians
SubjectIn our case, based on the college's academic disciplines
Peer-reviewed/refereedHas a field where instructions
can be included on how to limit a database search to peer-reviewed/
Citation examplesIncludes a field that shows examples
of citation styles for material found in the database. We created a
Web page on how to cite electronic resources, and this page displays a
list of full-text databases. When clicked, it brings the user to an information
page that offers examples of how to cite the full-text content.
DescriptorsTerms that describe the database and its contents,
making it possible for patrons to search for appropriate databases. However,
we turned off the search feature on the HDL, because many patrons thought
that by searching for a database, they were actually searching the
contents of the database itself.
Microforms (TofMic): Web pages that describe our microform sets are stored
as records in a table of microforms. In addition to a brief description of
the microform set, each TofMic record includes detailed contents information.
For some sets, this means a reel-by-reel description or a detailed bibliography.
For sets that have individual MARC records in our online catalog, we display
a WebCat search box that limits searches to the appropriate microform set.
Reference (TofRef): The college's academic departments each have an assigned
library liaison. One role of the liaison is to maintain a Web page that highlights
library resources that are important to the department, such as databases,
microform sets, and reference sets. Each liaison can add Web resources that
she considers important for her departments.
Staff (Staff): In addition to displaying staff information on the HDL, creating
a table of staff members allows each of the librarians to be authenticated
into Hobbes. Like TofDbs records, the staff record contains many fields that
allow them to be used in a variety of ways:
Login name (based on the college's LDAP server)
StatusLibrarian, support staff, student assistant
Office location, office hours, and phone number
ResponsibilityThese show up in the HDL's A-Z Index, so
that users can look for, say, the cataloging librarian
Birthday (so we all know when to expect treats!)
Access levelThis determines the Hobbes tabs (and tables
within the tabs) that each staff member has access to. For example, the
circulation manager only has access to the Lib Hours, Minutes, and Lib Resources
A-Z Index of this Web site (A-Z Index): Entries that appear on the HDL's "A-Z
Index of this web site" bring together records from several tables into one
place, including the A-Z Index table, the TofDbs, TofMic, TofRef, and Staff
table (see Figure 2).
Icons (Icons): As Figure 2 shows, we made the A-Z Index more visually appealing
after implementing a database of icons that is itself linked to from other
E-journals (Ejournal): Based on the experience gained from implementing Hobbes,
Jed and I decided to try our hand at managing our growing collection of e-journals.
The Ejournal tool in Hobbes is designed to accept a collection in the form
of batch loads of comma-delimited files (created from lists from aggregated
databases and our own e-journal lists). Using a complicated script, the tool
creates e-journal records for each title based on ISSN; any subsequent e-journals
that are loaded as part of another collection are not added as unique records,
but rather as a holding to the main e-journal record. Initially displayed to
the public within a separate database called the E-journal Locator, we eventually
figured out how to import these e-journal records into our online catalog,
thereby offering our patrons one place to find all our journals, print or online.
Passwords (Passwords): This table contains password records for resources
that require login information, such as e-journals, database administrative
modules, or database usage statistics. Creating one place for these allows
the librarians to easily share this important information.
Quotations (Quotes): The HDL home page displays a unique quote each time
it is loaded, and each of these quotations is stored in the Quotes database.
Additional tables control the "message of the day" (WebCat motd) that appears
in the banner of our online catalog.
Reference Desk Tab (RefDesk)
The default Hobbes screen for the reference librarians is the RefDesk tab,
which displays the tools they use while working at the reference desk. The
top half of the screen displays any questions that have been submitted by a
patron using the Ask a Librarian Web form. The librarian can choose to immediately
answer a question or assign it to another librarian. (See Figure
for all these questions are internally tabulated and can be accessed in Hobbes'
The bottom half of the RefDesk screen displays a RefBlog (a Weblog) where
librarians post reference-related entries. RefBlog entries can also be tagged
as sample reference questions to supply anecdotal evidence in statistical reports.
All of the information within the RefDesk tab is searchable.
Reference Desk Schedule Tab (RefDesk Sched)
With constant shift changes being scribbled on the print copy of the reference
desk schedule, or with shifts being missed because they could not consult the
schedule from home, our librarians welcomed the addition of an online reference
desk schedule in Hobbes (see Figure 4). Four weeks' worth of shifts are displayed
at one time, and changes can be made simply by clicking on a librarian's name
in a particular time slot.1
Library Statistics Tab (Stats)
Before Hobbes, librarians kept track of their own statistics in individual
Microsoft Excel files, a program that many knew well enough to enter data into,
but not well enough to creatively display or manipulate that data. Various
statistical tables are included in the Stats tab, as is a utility to create
reports based on date and other criteria.
Library Meeting Minutes (Minutes)
The Minutes tab displays folders that contain minutes from the various groups
of library staff that hold regular meetings. After being completed in Microsoft
Word, the secretary logs into Hobbes and uploads the minutes into the appropriate
folder for future reference. Also, an e-mail with the attached minutes is distributed
to staff members who have this preference selected in their staff table record.
Library Resources Tab
While the librarians are able to share documents on a common drive on the
campus network, it became apparent that some information could be better shared
by being displayed within Hobbes. The Lib Resources tab contains a wide variety
of links to resources such as a fund status report (used by liaisons to track
departmental book budgets); employee policies; evaluation documents; and library
labels, signs, and graphics.
Library Hours Tab (Lib Hours)
Hekman Library's hours are not easy to describe with text, so they are displayed
on the HDL in a 2-month calendar format. After being incorporated into Hobbes,
the tedious process of updating this Web page was replaced by a table that
utilizes an existing calendar.
Library News Tab (LibNews)
After attending a session on RSS2 at the 2003 Computers in Libraries
conference, it didn't take me long to wonder how we could incorporate this
into Hobbes, and it took even less time for Jed to actually implement this
feature. The LibNews tab displays information from RSS feeds, with the name
of the feed on the left and the contents of the feed displayed within a box
on the right.
This tab contains links to the administrative modules of five locally created
databases.3 As with e-journals, records can be added, edited, and
deleted, and bulk amounts of data can be imported using delimited files.
on bar beneath the tabs)
Since Hobbes displays information based on a librarian's login, we were able
to add some tools that provided unique data for a given librarian. For example,
each person is able to upload his or her browser bookmarks from the office
computer to use at the reference desk or from home. Librarians are also able
to upload liaison newsletters into Hobbes, which in turn are displayed on their
HDL staff page.
The Future of Hobbes
Jed and I created Hobbes on the fly, with no blueprint for what the "product" would
eventually become. Consequently, we gave many of the resources names and locations
that made sense to us, but would potentially be quite confusing for the librarians
who use the system, or for our successors were we to leave Calvin. ("Uh, what
exactly is a 'Tofdbs_cit_label'?") With this in mind, most Hobbes tools have
a text box for instructions that can uniquely identify the resource, and say
why the tool was created and what HDL pages contain its dynamic data. (See
Figure 1.) In the near future, I plan to flesh out these instructions as we
consider plans to make Hobbes control the remaining HDL pages. From time to
time, we find other ways of displaying the data, and as other tools are added,
we rename, re-create, and reorganize some of the tabs.
Even though Hobbes is yet another resource vying for attention in already-full
schedules, it was not difficult to convince the librarians that Hobbes was
something they should be using. They like that library information has been
moved from files on a shared network directory to an environment where it can
be stored, displayed, and manipulated for the benefit of both the library staff
and the patrons we serve.
The biggest challenges that we face are the maintenance required to keep
Hobbes' information up-to-date and keeping the complex scripts working properly.
Creating a CMS like Hobbes required extensive programming skills. Despite Hekman
Library's relatively small size, I was fortunate that it also employed a systems
programmer. Other libraries considering an undertaking like Hobbes would not
only have to evaluate the skills of available employees, but also consider
the time needed to create and maintain such a system.
1. For additional information, see Sennema, Greg.
" Managing the reference desk online." Reference
Services Review 31 no. 3 (2003): 257263.
2. For additional information on RSS, see Tennant, Roy.
" Feed your head: keeping up by using RSS."
Journal 128 no. 9 (2003): 30.
3. You can find information on these five databases at http://www.calvin.edu/library/databases/index.stm.
also Sennema, Greg.
" The Christian Reformed Church
Periodical Index: a local solution to indexing periodicals."
Journal of Religious & Theological
Information 5 no. 3/4 (2002): 119134.
Adalian, Paul T. and Judy Swanson.
" Locally developed
web-enabled databases: new roles and opportunities
Reference Services Review 29 no. 3
Brown, Katrina L. and Anne M. Candreva.
database-driven web content."
netconnect (Fall 2002): 24.
Browning, Paul and Mike Lowndes.
" Content management systems: who needs
Ariadne 30 December
Fagan, Jody Condit.
" Transition to a unified system: using Perl to
drive library databases and enhance web site functionality."
Library 19 no. 3 (2001): 168179.
Klein, Leo Robert.
" Your web sitemixing up web site management."
Journal netconnect (Spring
" Designing a database-driven web site, or, the evolution
of the Infolguana."
Computers in Libraries 20
no. 9 (2000): 26.
" Database-backed library web pages."
The Electronic Library 19 no. 6 (2001): 424432.
Greg Sennema is digital resources and reference/instruction librarian at Calvin
College and Calvin Theological Seminary in Grand Rapids, Mich. He holds a master's
of information studies from the University of Toronto. His e-mail address is