Piers Young made available a first version of his Knowledge Service Methodology deliverable (KSM) and asked for comments and criticism.
Let us first apply KSM to itself. The first paragraph states that knowledge workers spend a considerable amount of time searching for documents. What it does not mention is that once a (relevant) document is found, the next question is: do I read it? If so, I might waste yet more time! For me this applied to KSM. The content is relevant to the things I'm interested in, but it contains too many details in the form of quotations, historical overviews and the like. The summaries on the above link to Piers' weblog can be used to check whether you should read all of it :-).
The direction KSM proposes is "semantic metadata" combined with "context awareness". This appears sensible to me. However, there are two important things to consider. (1) Metadata, as we know it, is very domain oriented. For knowledge workers, other dimensions are also relevant. Examples are the level of detail and the perspective. It may be doable to metadata a document and label "chicken pox" as a virus (domain point of view), but how does one metadata chicken pox documents on dimensions such as: definition, cures, history and so forth? (2) I'm slightly suspicious of the term "context" in relation to computers. If the average search engine included two fields, one labelled "Query" and one labelled "Why?" the results might improve dramatically. Knowledge workers search with a purpose and this purpose might be out of context.
I liked KSM.
Comments