Philip E. Agre

Department of Information Studies

University of California, Los Angeles

Los Angeles, California 90095-1520

USA

pagre@ucla.edu

http://polaris.gseis.ucla.edu/pagre/

On the Horizon 7(5), 1999, pages 8-11.

Copyright 1999 by Jossey-Bass.

Please do not quote from this version, which may differ slightly from the
version that appears in print.

1800 words.

The higher education community is planning for a world in which information
technology (IT) will be so pervasive that the very institution of higher
education will change. Of course, IT probably can be used to improve higher
education. But IT is exceedingly flexible, and we will face numerous choices
about how best to apply it. Some of those choices are straightforward matters
of efficiency, best left to technical experts. Other choices will require
us to reflect carefully on the values that a university ought to express.
If educators have learned anything from attempts to improve life using IT,
it is that significant improvements are possible only when institutions are
rethought. But in order to rethink institutions in a responsible way, we
first need language to describe them.

Let us consider an example. In a 1998 letter to University of California
faculty members, Sandra Weiss (1998) -- then chair of the University system's
Academic Council -- discussed course articulation, which she defines as
"the degree to which students can build an additive degree program by taking
courses either at different institutions or at the different campuses of one
institution" (all quotes page 5). (This same idea is called modularity
in Britain, where it was central to the higher education policies of the
Margaret Thatcher and John Major governments.) Weiss contends that flexible
course articulation is important because "we have moved into an era where
individual campuses are becoming part of a larger academic community -- a
'global academic village,' so to speak". IT helps drive this trend, and
Weiss further explains that "[f]or technology-mediated coursework, we need
to identify comparable content across courses that would be acceptable for
transfer and also grapple with our expectations regarding traditional 'face
to face contact' between professor and student and among students themselves."

This sort of commentary weighs against the stereotype of professors (or
"academic elites," as the new jargon would have it) as Luddites engaged in
bull-headed resistance to technologically driven institutional change. In
her letter, Weiss, a professor in the School of Nursing at UC San Francisco,
supports my own impression that faculty hierarchies are in fact contemplating
fundamental changes. Nonetheless, I believe that faculty and administrators
alike should step back and evaluate more fully what choices they are making.
What are the trade-offs associated with articulation? And how are these
related to IT in particular?

Ontological Standardization: Making Disparate Systems Uniform

The main tradition of computer system design involves a phenomenon that
might be called ontological standardization. In writing a computer
program, the first challenge is to define the ontology that the program's
data structures will reflect -- that is, to define what sorts of things the
world is made of and, therefore, what sorts of data objects will be created
and stored through the program's operation. In technical terms, this is
called a data model (Simsion 1994). In the case of higher education,
an ontology might include people, job titles, departments, courses, majors,
and grades. The resulting program will work correctly only if everything that
the program represents can be contained within one of those six categories.

In the old, un-networked world, different universities developed their
ontologies somewhat independently of one another. Some trends, including
the frequent movement of administrators from one university to another,
did enable what Walter Powell and Paul DiMaggio (1991) call institutional
isomorphism: a pattern of analogies between the workings of different
organizations. But in the world of networked computing, the forces of
institutional isomorphism are amplified greatly. If a student chooses to
study at only one out of hundreds of different four-year schools, it matters
little whether the internal workings of those schools can be mapped onto
one another. But if, at the opposite extreme, a student chooses from among
hundreds of schools for each course or even each class meeting, then the
schools need to ensure that their definitions of course or class
meeting, among other important terms, are the same.

Ontological standardization, then, is what happens when separate organizations
in a given institutional field are required to make uniform the most
fundamental categories of their internal workings. Until recently, the issue
has arisen primarily in the context of mergers between corporations; if the
computer systems of two merging companies cannot talk to one another -- say,
because their definitions of employee and sales do not match --
then genuine havoc can result. Now, however, the same issue can arise in a
wide variety of institutional contexts, even when separate organizations are
not formally merging.

The Implications for Higher Education

Institutions of higher education now compete on the basis of their distinctive
programs: one economics department, for example, might be ranked above another
in a magazine survey. Departments can design the curricula for their majors
according to their own distinctive approach, and students can choose the
program that fits best with their values and goals. Universities can design
their core curricula according to an overall educational philosophy. Because
decisions about program philosophy and course content are made by the faculty,
the contents of and boundaries between courses are flexible; they can be
changed to suit evolving circumstances, not least the interests of the best
students.

A radical increase in articulation would threaten this flexibility. Allowing
students to earn academic credit at multiple campuses is often a good thing.
It can create financial tension between the campuses, but it also permits
students to save money, discover new interests, or overcome imperfect high
school transcripts. The University of California recognizes this fact;
for decades, it has encouraged students to transfer from community colleges
into the UC system, even though the transfer students sometimes find this
transition difficult. But if it becomes radically easier to transfer course
credit between schools, thus effectively enabling students to assemble their
college education a la carte from among the offerings of a large number
of potentially quite different programs, then, I will argue, intellectual
diversity may suffer.

Observe, first of all, that articulation requires ontological standardization.
At present, with a low degree of articulation, much bureaucracy is required to
determine which courses at school A count for which requirements at school B.
As the degree of articulation grows, this ad hoc scheme becomes impractical
(Hawkins 1999: 15). Thus, to the extent that degree programs are assembled
from bits and pieces of coursework provided by different campuses, the very
concept of a course (and a grade, and so on) must be the same on
each campus.

It does not follow that ontological standardization implies articulation.
Institutions of higher education could systematically standardize their
ontologies in order to employ compatible software, or to achieve economies
of scale in the training of their staff, without trying to standardize the
contents of particular courses. Problems nontheless arise. Even though
it does not directly affect course content, ontological standardization
would make it difficult (i.e., even more difficult than it already is) for
a single institution to choose a radically different path -- eliminating the
boundaries between courses altogether, say, or replacing grades with narrative
evaluations. All of the standardized schools would reap the benefits of
interoperability, and the outlier would suffer competitively as a result.
If a qualitatively different and better way of organizing higher education is
out there waiting to be discovered, the lock-in of a single set of standards
(Shapiro and Varian 1998) might render it impracticable. Most importantly,
ontological standardization would remove a major barrier to articulation,
and would thereby increase the incentives to standardize course contents.
Those incentives, of course, are not entirely absent in a print-based world:
textbooks, for example, respond to the economies of scale that can be achieved
by standardizing the contents of introductory courses. But with technology,
the opportunities for content standardization are much greater.

The institutional consequences of this trend would be even worse. A radical
increase in articulation would require the internal modularity of degree
programs to be coordinated centrally, or at least negotiated among numerous
independent universities. The result would be less flexibility and greater
uniformity. The power to adjust fine details of a curriculum inevitably
would shift from faculty members to accrediting organizations, university
administrators, and other professional coordinators. Faculty may effectively
lose the right to design their own syllabi. This is a matter for serious
concern. The diversity of thinking and teaching at universities has long been
important to the health of a free society; that is why professors are awarded
tenure once they have passed several competitive hurdles and proven their
abilities. Faculty are in day-to-day contact with the particular students
who have chosen them and are thus in a much better position than a centralized
organization to assess their evolving needs. Increased articulation through
the use of IT would thus endanger the institutional conditions that guarantee
a diversity of intellectual approaches.

Important Choices

As the higher education community decides how to use IT, then, it faces
important choices. Before advanced communications technologies became
widespread, educational decentralization and diversity were promoted by
the limitations of the physical world. Universities were distant from each
other geographically, and it was relatively difficult to transfer people
and practices between them; consequently, different universities evolved
along somewhat independent paths. Now, however, that independence -- that
separate evolution and diversity of educational approach -- exists only if
educators actively choose to foster it. And because the local benefits of
standardization are easier to quantify than the global benefits of diversity,
broad awareness of the issues is crucial. A new generation of students,
never having encountered higher education before, may not even recognize
the dangers of a centrally planned educational economy or an intellectually
homogeneous society.

Although many cyberspace visionaries have asserted that IT inevitably brings
decentralization and choice to the world, this analysis of institutional
isomorphism suggests that the opposite might be closer to the truth. The
practice of ontological standardization first arose in military and industrial
settings in which centralized coordination did not threaten important societal
values. Higher education, however, is a different story. In order to use
IT constructively, educators may have to reinvent it, reconceiving what it
means to define an ontology, to standardize practices, and to fit modular
components into a whole. At a minimum, we should weigh as fully as possible
the potential consequences of each individual standard. Particular standards,
such as for the university's financial processes, may have no significant
impact on education. Others can perhaps be designed carefully to gain
the benefits of interoperability without contributing to an intellectual
monoculture (cf. Hanseth, Monteiro, & Hatling 1996). But some standards
may grease an already slippery slope that endangers the university's social
purpose, and those standards should be rethought. So let us use technology
when it helps us do our good work better. But let us not permit the
traditional practices of technology to dictate important, value-laden changes
in our institutions. Technology is supposed to serve human purposes, but the
burden of technology is that we must choose carefully how to apply it so that
we do not sacrifice individuality.

Acknowledgements

I appreciate the thoughtful comments of Barbara Horgan and the anonymous
referees.

References

Hanseth, O., Monteiro, E., & Hatling, M. (1996). Developing information
infrastructure: The tension between standardization and flexibility.
Science, Technology, and Human Values 21(4): 407-426.

Hawkins, B. L. (1999) Distributed learning and institutional restructuring.
Educom Review 34(4): 12-15, 42-44.

Powell, W. W., & DiMaggio, P. J. (Eds.). (1991). The New Institutionalism
in Organizational Analysis. Chicago: University of Chicago Press.

Shapiro, S. & Varian H. (1998). Information Rules: A Strategic Guide to the
Network Economy. Boston: Harvard Business School Press.

Simsion, G. C. (1994). Data Modeling Essentials: Analysis, Design, and
Innovation. New York: Van Nostrand Reinhold.

Weiss S. (1998, May). Notes from the chair. Notice. (David Krogh,
Editor; University of California Academic Senate; University of California
Office of the President; 1111 Franklin Street, 12th floor; Oakland, CA 94607;
USA. On the Web at <http://www.ucop.edu/senate/notice/my8notc.pdf>http://www.ucop.edu/senate/notice/my8notc.pdf>.)