An earlier post suggested that Minsky's model might help motivate some discussion related to 'oops, loops, and oops. The WSJ couched the discussion in terms of lies told by economists. We could probably talk about untruths (not to imply lies, rather those things that are meant to overcome uncertainty and are bolstered by optimism) told by suppliers.
But, let's use a couple of economic notions from Minsky and see if there might be a program / project parallel.
For one, in terms of borrowers and a cycle, he described three levels. These are the hedge borrowers (where something of value might be thought of as collateral - existing as a real thing at the moment), speculative borrowers (where leverage starts to raise its head), and Ponzi borrowers (yes, this concept can be applied to the multi-layering found in an extreme outsourcing arrangement).
The movement through these types (and, there are probably many more levels that one could differentiate) goes from the more real to the abstract (I'm being nice here and pointing more to ignorant manipulations rather than explicit malfeasance).
That one needs to hedge has become clear, and this can make financial sense. Can one 'hedge' in engineering? That is a question we must ask and discuss. Some related issues are earned-value analysis, clarity, and the like.
Well, one could definitely talk about 'speculative' use (and, perhaps, even 'ponzi') in terms of some analytics and their assumptions. The main question is how to guard against this; good people, in short.
Now, in terms, of a product, testing is a key convergent phenomenon that has to have higher priority than we saw in one instance.
Too, Minsky talked about some other attributes associated with a cycle, using displacement, boom, euphoria, profit taking, and panic, from early on to late. Without going into detail, it's not hard to map this same type of thing with projects. What denotes more about 'panic' than sending oodles of people out to a project? The mythical man-month applies to more than software.
Well, control is an operative word as a means to dampen instability. And, the old watchwords come to play, such as measurement, vigilance, and, I might add, attitudes that are anti-hubris.
08/01/2013 -- Ben cannot unwind or taper down; he has too many Doves. We'll have to get back to the king thing (yes, the divine rights of the CEO, new royalty, in other words) and dampening of these types by a new outlook (Magna-Carta'ísh).
12/17/2008 -- We'll use made-off in lieu of ponzi, henceforth.