The Nature of Knowledge Work

I was writing a generational history of electronic data exchange yesterday, and in the description of the first generation explained why we call batch files ‘batch’ files. They are produced by batch jobs. Batch jobs are one of the three classic methods of production that antedated the computer age: job or ‘one-off’ production, batch production, and flow or ‘mass’ production. And these classic methods are still the primary underlying conceptual model of production we all share.

In the same kind of way, the business process model we all share comes to us by way of Adam Smith, Frederick Winslow Taylor, and the twin pillars of modern process engineering, Peter Drucker and W. Edwards Deming.

Drucker actually coined the term ‘knowledge worker.’ And in the healthcare industry, that is what most of us are. Sure, there are companies manufacturing drugs, durable medical equipment, and so on. But most of us are communicating and making decisions about healthcare – clinical decisions, financial decisions.

In the software business we document business processes. The tools vary, and the level of rigor varies, as methodologies come and go.

But there always seems to be a conceptual gap we have to leap over going from a process model where we have identified a new automation boundary, and the actual design of the software components and the orchestration and choreography needed to realize it.

I have come to think that in part it is because we don’t have a clear understanding of the nature of knowledge work in the age of computing. And that we suffer the shackles of our Drucker-Deming conceptual history. Which makes it hard to express a ‘classic’ business workflow as software .

I have come up with an approach I have refined over the years that I think is directionally right. (That classic process modeling doesn’t work well for knowledge work seems to be widely recognized. I have not done a rigorous search for prior art on new models – which may be in large part me indulging in the conceit of creation – if you know of any, please add links in the comments. )

I have a lot more context for this that I will post about. But for now, here is the core meta-model.

Stay tuned.

One thought on “The Nature of Knowledge Work

Leave a comment