RE: Lack of Deming Philosophy in Education, bounding of systems

["Vic Forte" <vic@vicf.com>]

> I believe the above comment may highlight where my confusion may arise.  I
> do not see the underlying assumption that a system can be black boxed and
> have external inputs and outputs.  I do not see how the inputs and outputs
> of a system can be displayed graphically, or how the definition of the
> system can allow for external inputs or outputs.  I believe the black box,
> process, or function model cannot replace the system model and as
> alternate
> models, the system model is the more accurate.

A "system" is an operational definition. It is not something that exists
"out there". We bring this operation definition to reality in order to limit
the number of aspects of reality that we have to consider. It is simply a
tool - a kind of filter.

> none of the arrows shown is unterminated.  There are no arrows going out
> into the unknown, nor arrows coming in from outside.  There is simply
> nothing within the system to generate nor handle the hanging
> arrows from the
> black box model.

This is a very good point.

As I understood it, the concept of the black box is part of General Systems
Theory.

In General Systems theory there are many orders of system. The simple black
box that I think you are referring to is the first order of system. One
input, one output. Example - a piping bag. A second order system has
feedback i.e. an information arrow flowing from the output to the input
signifying negative feedback. (Example - a thermostat). A third order system
is predictive. The higher orders are capable of learning - viz. animals, and
creative and capable of questioning the purpose itself through values viz.
humans. The Deming model is an example of the highest order model which is
capable of redesigning itself towards a purpose. There comes a point when
the idea of a black box simply breaks down at the higher orders, and ceases
to be useful. But within systems thinking there are loads of different tools
to help us conceptualise the world that we live in. But all models serve the
purpose in hand only. Sorry by the way if I am teaching granny to suck eggs.

I do accept that Dr Deming avoided putting boxes in his diagram, and he
explains why.

> For an aim to be meaningful, it must define what is done, who it is done
> for, and who provides the resources to accomplish it.  The result of an
> operation can only be evaluated in the context of the user of the output.
> The feasibility of an operation can only be determined based on
> the supplier
> of the input.  Hence, the aim must define the suppliers and users for any
> operation.

A point which we would all do well to note. I will certainly give this some
reflection in my daily work. Thank you.

If we bear in mind that black boxing is no more than a tool for restricting
the options that we have to think about, then it is pretty clear that each
black box can be exploded to examine the black boxes inside of it, and so
on, but without overloading us with information and excessive detail.
Similarly each black box is really only understood as a component of the
larger black box in which it is embedded. No serious proponent of systems
thinking would suggest otherwise. Stafford Beer recommends that we focus
simultaneously on at least three recursions of system. With the "system in
focus" in the middle. But in his amazing little book "Diagnosing the system
for organisations." you see no black boxes.