Recent posts have had an
economic flavor as this topic is important to engineering using a larger definition of economics that gets beyond money and scarce resources.
In a sense, any decision process (and, folks, computational processes are nothing but decisions) has aspects related to
economics.
Minsky's model really deals with how far we can carry relationship before having to check-in with reality. For instance, feedback loops work (such as, automated flight) since they tie into sensors.
What are the data that can drive a project or an economy? Well, that is the subject of the
7'oop7 blog, in part. It is of interest here, to boot.
So, getting back to Minsky, he proposed what might be termed a hierarchy of values and their measurables. That is, we have things that are more concrete (like a plane that can actually fly given fuel, pilot, and other things) and more abstract (like some plan - even if computionally supported- or design - until proven). Then, we have (essentially) extrapolations from that.
Our abstractions can get far removed from reality (whatever that is, we'll get into). In fact, some mathematicians (those with the pure hearts) take great glee in being thusly removed. Whereas, those who have to actually build things (like airplanes) cannot remove themselves too far from what is real.
So, dealing with the real, we have to like hedging, though that concept has been usurped by particular types of funded thinking. It's the first-order
abstraction that will work and be in control, given that certain rules are met.
Next up is speculation. Think of this as the realm where the
hype versus hypothesis dynamics play out. Again, this can be controlled, though risk is a definite element here.
Too, it is here where management starts to wreck havoc (the old pun that in a manager's meeting, water runs uphill is apropos) due to their having insufficient data or having improper grasp of what goes behind the data (an argument that data is function-based and not merely related to capture/retrieval operations).
One might say that we need the hedge realm in order for the systems engineer to work. Then, we need the speculative for the business and economic goals to play out. Note, though, that the speculative is only as good as its underlying hedging ontology.
Finally, we have the Ponzi realm, according to Minsky, for which all bets are off. It's the realm of those with either too much time on their hands or no clue about the impacts that their decisions have on others (like second-hand smoke, we ought not allow secondary and tertiary and more economics effect to be played out in the hands of those without ethics) or the like.
From a project viewpoint, the house-of-cards upon which rests some computational ontologies are not unlike the Ponzi. We'll get back to that in the context of how the
quasi-empirical necessitates certain types of project activity, such as
testing early and often.
Yes, Minsky had some powerful insight here that engineers need to think about.
Remarks:
11/04/2010 -- Big Ben is still putting us
at risk and
trashing the savers.
09/12/2009 -- Sandbox was used without definition. Let's discuss
that concept.
12/17/2008 -- We'll use
made-off in lieu of ponzi, henceforth.
09/14/2008 --
As seen in the
7'oops7 grouping related to Minsky's model, examples abound of finance going awry, very much due to
moral hazards being allowed to continue in the name of
capitalism, leading to the need for bailouts as we've seen much of, of late. Several arguments get in the way. For instance, these are complicated issues; yet, some complications are self-created in order to continue the play ground (
gaming). And, we don't know a whole lot; why? Well, we're dealing with the dismal science. Yet, even the premiere sciences have (or will, at some point, have) to deal with the issues of
quasi-empiricism and what they might mean. So, let's get the sandbox defined, and in place, for those who need the thrill. The more mature minds ought to be considering how to establish a more solid basis for the future (the metaphor: not only is the current financial scheme on sand, it's continually washed by waves; we need to build the land of continents that can stand the test of time. ).
08/03/2008 --
This post showed how
Minsky's model for the inevitable outcome of
gaming in finance could be an analog for engineering management issues. To take the analysis further,
money has to be looked at in depth in an attempt to show why we've lined up on the side of the
gamers and to attempt to look at alternative methods that are not so
painful to the hapless.
Modified: 08/24/2011
