Page 143 -

126 Chapter 4

The construction of taxonomy involves identifying, defi ning, comparing, and
grouping elements ( Lambe 2007 ). Organizational knowledge taxonomies, however,
are not driven by basic fi rst principles or “ real ” attributes, but by consensus. All the
organizational stakeholders need to agree on the classifi cation scheme to be used to
derive the taxonomy — it cannot be theoretical but must be empirical — this is how we
code this type of knowledge in our work. The reason for this is that unlike traditional
taxonomies, such as the fi rst comprehensive biological species taxonomy developed
by Linnaeus (1767), the purpose of an organizational taxonomy is not to come up
with a universally accepted way of describing reality. Rather, an organizational tax-
onomy is a mixture of a depiction of concrete components and abstract concepts that
together make up the context of that particular company. Consensus is vital because
the taxonomy serves to help achieve the goals of the organization and it does
this by helping knowledge workers communicate better, code knowledge better, and
organize this coded knowledge in such a way that it can be used by everyone today
and by workers of the future when they need to retrieve and make use of this
knowledge.
A taxonomy is a classifi cation scheme that groups related items together, often
names the types of relationships concepts have to one another, and provides some
notion of more general categories versus examples or specifi c instances of a category.
Classifi cation schemes can be very personalized, such as the names we give our per-
sonal e-mail folders or PC desktop fi les. There is no problem as there is typically only
one user — you (and hopefully you can remember how you named your folders!). But
what happens if we are working with someone else? We usually refl ect a bit more
before typing in the e-mail subject heading and before naming a fi le to be sent as an
attachment. Why? The names must make sense to you but also to the recipient. In
the same way, we have no choice but to standardize a bit more and to achieve some
sort of consensus if there are a number of people working with the same content. At
the very basic level, a consensus on naming different versions of a document that has
multiple authors will be needed. The organizational level will require the highest level
of standardization and consensus. Perfect consensus is rarely feasible (and is not very
cost-effective), so we are fortunate to have a way of “ cheating ” : together with the
knowledge dictionary, it is often a good idea to develop an organizational thesaurus.
The thesaurus will contain all the synonyms and cross-references prevalent in the
organization. For example, one group may have decided against using the term knowl-
edge management and prefer knowledge sharing , and yet another division may adopt
knowledge networks . All three would appear in the thesaurus, with KM highlighted as
the formally accepted term for the organization as a whole, while allowing for some