3.6 A Taxonomy of Work
The three initial empirical studies indicate that the perceived difficulties
of enterprise modeling are influenced by the kind of work that one tries
to model, i.e., the nature of the UoD. The taxonomy provided in this section
concerns this issue.
3.6.1 Four categories of work
One way to categorize work is according to the following two dimensions:
The resulting four categories are illustrated in figure 3.4.
Figure 3.4: Four categories of work
Uncertainty of production (replication risk versus design risk)
Form of artifact that is produced (matter versus information)
The rationales for introducing the two dimensions are the following:
The uncertainty of production dimension is introduced based on a
claim in both VPT and PA30 that creative processes were difficult to model.
The form of artifact dimension is introduced based on the claims
in VPT and PA30 that physical processes are more easily agreed upon and
modeled than information-based processes.
Dimensions in the taxonomy
The first dimension in the taxonomy concerns uncertainty of production.
Bollinger and McGowan (1991:36) categorize work according to whether it
is dominated by replication risk or design risk. Replication
risk means that the main challenge is to produce a replica of an already
known product. Hence, the product of the production process is known
in advance, but the production process itself cannot fully be controlled
(it is uncertain at the required level of abstraction).
Design risk, on the other hand, concerns the ability to produce an artifact
at all from a given set of requirements. One cannot in advance say how
the final product of the production process will turn out. Hence, neither
the production process itself nor the outcome of the production process
Along the other dimension, work can be discriminated according to the main
form of the artifact that is produced: Matter or information.
Artifacts dominated by their material aspect are functional due to their
properties adhering to natural laws, while artifacts dominated by their
information aspect derive their functionality from the symbolic meaning
|Manufacturing of shirts is an example of work associated with
replication risk, provided that one has designed a prototype shirt in advance.
Software production is an example of work dominated by design risk,
according to Bollinger and McGowan (ibid.:35).
|A shirt is first an foremost a material artifact, as the intended
purpose of a shirt is to keep the wearer warm or cover his body. These
purposes are met according to physical laws. The valuable aspect of the
shirt is perceived to be its physical qualities (unless the symbolic value
of a specific brand of shirts dominates).
The letter answering an application for a bank loan is dominantly
information, as the important aspect is the meaning of the answer, not
the manifestation. The paper and ink is not valuable to the applicant,
while the information in the letter is.
Categories of work
Combining the two orthogonal dimensions provides the four categories of
work as illustrated in figure 3.4:
The categories may be discussed in terms of perspective making and taking:
Replication risk, matter (RM)
Work has form of mass manufacturing of identical physical products,
e.g., manufacturing of cars and processing of gas.
Design risk, matter (DM)
Work is one-of-a-kind production, e.g., creation of a prototype.
Replication risk, information (RI)
Work is standardized information processing, where at least the form
of the product is known in advance. The main challenge is to perform it
efficiently. An example is claims handling in an insurance company.
Design risk, information (DI)
Work has the form of sense-making and problem solving, trying to construct
a model of some product that will satisfy the given constraints. An example
is at least some activities in the design of software.
The view that categorization of work depends on community of knowing is
supported by Dougherty (1992:189) stating that actors within different
departments do not appreciate the complexity of other actors' work. Suchman
(1995:59) agrees when stating that
Work associated with replication risk occurs within a community of knowing
that has developed a strong perspective, in the sense that their understanding
of the production process and the product is well developed. Work associated
with design risk may be seen as lacking a strong perspective, as neither
the process nor the product are familiar.
Production of material artifacts is a matter of skill, i.e., being able
to perform the physical actions that are required to transform matter into
the required product. Production of information-based artifacts is a matter
of understanding -- of perspective making and perspective taking.
Work may be found to belong within the confines of several categories simultaneously.
However, one of the aspects may be viewed as dominant. Also, the category
is not inherent to the activity, but relative to community of knowing and
their perspective. E.g., as a strong perspective is constructed, work becomes
more dominated by replication risk than by design risk.
"...work has a tendency to disappear at a distance, such that
the further we are removed from the work of others, the more simplified,
often stereotyped, our view of their work becomes."
Hence, an actor with no hands-on experience from working within a domain
may describe the work as less complicated than a seasoned practitioner
within the same domain (and assess work as dominated by replication risk
rather than design risk). The DI type of work is equivalent to what
Boland and Tenkasi (1995) refer to as knowledge-intensive. Their stance
is that effective perspective making and perspective taking is required
in organizations doing knowledge-intensive work. They do not say that
the activities are not valuable in organizations doing other kinds
3.6.2 Categories of work in the empirical studies
Of the categories of work, three out of four were observed in the initial
Work belonging to the DM category was not observed in any of the
The RM category was exemplified by physical production, transportation
and processing of oil and gas, and was modeled in all three projects VPT,
PA30 and Gazz.
The RI category of work was exemplified by the formal aspects of
pre-production activities as modeled in the VPT project (e.g., developing
applications for allotment of promising fields).
Work in the DI category was observed in the Universe of Modeling:
Enterprise modeling is an example of DI type of work. It was also
observed in the VPT project when trying to model research.