3.3 Organisational memory systems continued
3.3.1 Metaphors for organisational memory systems
Section 2 argued for a model of knowledge deriving from the situated interpretation of abstract representations. There is an active process by which different interpretations may result from a given information source. This is in contrast to the popular notion that knowledge can be unproblematically encoded and digitally stored and accessed.
Bannon and Kuutti (1996) argue that the term ‘organizational memory’ is widely used to mean a repository based on an implicit ‘memory as bin’ metaphor, whereby material is unproblematically added and extracted. When we look at how human memory actually works, the cognitive sciences show that ‘memory as reconstruction’ is a much better model. Memories are not simply retrieved according to a database model, but are reconstructed in the context of our own understanding of the world, who is asking, and for what purpose. The task for organisational memory design is better conceived as the provision of resources for reconstructing and negotiating meaning (there are often different recollections of ‘what happened’ and different perceptions of ‘what this now means’). This is, therefore, one important difference between a concern with knowledge technologies for human interpretation and action, and databases serving information for machines.
Bannon and Kuutti emphasise the important role that ‘talk’ and ‘narrative’ seem to play, according to a variety of studies into how knowledge is shared in organisational contexts. Although they do not develop this theme's implications for group memory technologies in any detail, the implication is that systems that do not recognise this natural process may not be successful. (We return to this theme in ‘Stories for sharing tacit/informal knowledge’ in Section 4.2.)
A useful ‘design space’ for organisational memory has been proposed by van Heijst et al. (1998) (Figure 6). A design space articulates two or more dimensions that highlight important differences between designs. The authors also suggest example technologies.
Note that both collection and distribution can be active or passive. The metaphorical label in each cell implies the use of very different kinds of technologies. It can be seen that an organisation might make use of all four kinds of system on many scales, from small teams to enterprise-wide collection and distribution.
Consider the concept of organisational memory and the four metaphors in Figure 4: knowledge attic, sponge, publisher and pump.
Which metaphors best characterise the way your organisation's information collection and distribution systems (computerised or otherwise) work (or fail to work)?
Reflect on the use of active/passive criteria in relation to both collection and distribution. Do we assume that the lower-right quadrant is the ideal type, or are there merits in the other quadrants?
Most organisations may have elements of all four types, even if these are highly dependent on individuals who act as sponges’, ‘hoarders’ ‘pushers’ and networkers. Certainly, most will have ‘attics’ and many will have ‘publishers’ (for example, the circulation of documents or memos to selected colleagues). We might have some doubts about targeting all our efforts on the ‘knowledge pump’ since its apparent mode of operation is highly dependent on timing. Perhaps we also require a knowledge attic’ where we can store knowledge we are not able to exploit at present; though, ideally, in a form open to search and retrieval by software agents who can ‘pump’ relevant fragments to the right people at the right time (the challenge, of course, is how such agents keep track of the users’ contexts, so that what they send is relevant).
The ‘holy grail’ is a technology which promotes the integration of information and experience in order to build new layers of meaning and higher levels of understanding. Doing this completely automatically is a long-term research challenge, and probably only possible in very tightly restricted fields. A shorter-term target is to produce collaboration tools which mediate and enrich human reflection and discussion, but without overwhelming participants (who only have limited time to engage in such forums) and, ideally, adding value to a conventional online forum by drawing attention to relevant past discussions and contents in the ‘attic’. At present, research laboratories are prototyping such systems: they are not products.