Thursday, May 31, 2007

Scale and Scope-Creep

Our mental model of the nature of the world dominates our thinking on what kind of problems we should be starting with.

The classic view, emphasized repeatedly by academics to new students, is to keep the focus narrow, to work on the smallest problem possible and do it well. Large problems are "bad", and very large problems are "world hunger" or impossible:



But, if we look at problems like antenna radation with near and far fields that are both easy, and just a middle field that is difficult, we have to ask if the actual curve doesn't look more like this one, with "easy" parts at each end and the "hard" part in the middle. At the low end, one thing dominates and other terms can be ignored. At the high end, a different thing dominates, and everything else can be ignored. It's only in the middle that nothing can be ignored and the problem becomes too hard to do.

An example was the water in the faucet. At a molecular level, we can model the motion of several, possibly 100 molecules. More gets harder and harder. But if we keep going up to the level of ten to the tenth molecules, we get solvable problems, just with different terms. We now have terms like "water pressure" and "volume" and "flow rate" that mean nothing at the molecular level. At the water level, looking back to the molecular level, it now looks hard, what with all the probability distribution functions and quantum mechanical effects and waves instead of particles, etc.





But, as Marsden Bloise pointed out in an article I read long ago that simply transfixed me, the reality of LIFE is that it has a "curiously laminated" quality, with levels that make sense to us (cells, organs, systems, people, teams, companies, planets" separated by stuff between the easy levels, like filling in an OREO brand cookie, that is squishy and hard to analyze.

That model, then, is more like this picture:



The conclusion is that LIFE has levels at which there are meaningful concepts, separated by spaces in which we can't find meaningful concepts. Each of the levels we know of has academics studying it, and they are treated as if they were entirely different universes, not different parts of the very same single LIFE object.

So we have Politics and Sociology and Psychology an Biology and Cell Biology, which hardly ever talk to each other, but, in reality, are just different aspects of the same LIFE entity that humans are in the middle of. As we begin to understand that these levels interact, perhaps even causally as seen from above (but not from below), we begin to figure out that there is really only one large complex thing here, not many small distinct separate things. We have to unfragment what we know about LIFE and reassemble the pieces.

ANYWAY, my point is that sometimes LARGER is actually EASIER, as the study of "water" is much easier than the study of "quantum mechanics of a dense population of H20 molecules."

So, the same thing is true, in my mind, about organizational functioning. There is no point in saying we are going to "solve" the individual and small-team problems first, and then, if there is time left over, move on to the much harder department and company scale problems we face.

First, we'd be missing all the easy cherry-picking solutions at the higher levels.
Second, since everything is connected to everything else, like some huge "mobile" hanging structure, it's actually not possible to "solve" any one level without at least partly solving the level above it. We've found that out in Public Health and Psychology -- there's little point in trying to change one person or one tribe, because, when we walk away, the surrounding systems push back on them and the person or tribe reverts to their old behavior.

The best solution, then, would actually address every level simultaneously, and ask the question of "What would be a win-win-win solution here?"

In Public Health, that would mean asking what would simultaneously address personal, corporate business, city, state, and population level needs, and national identity needs? Instead, too much effort is put into trying to sub-optimize the problem, and solve "environmental health" at the expense of jobs, that then backfires because the unemployed workers then have worse health than before.

Larger is smaller. Bigger is Easier. It is neither ill-advised nor a waste of time nor unimaginably complex to address large issues ahead of small ones. Like water versus molecules, sometimes the larger problems are much easier to solve.

And, like my post on the fragmentation of social groups, if we don't address the larger problem, the gaps and holes that neglect produces manages to defeat, neutralize, and devastate all the work we did "solving" local problems. All our work is wasted if we train people to perfection and no one thinks the problem that arises is theirs to fix.

Similarly, the Toyota Production Model and "lean" thinking can help us greatly in addressing local efficiencies, and even set up some global concepts in terms of "pull" and the impact of stopping a whole line because anyone, at any level, is having a hard time doing their own job because of difficulties someone else exported to them. Those are great concepts.

But those concepts still start at the bottom and work upwards. They leave the separate and mostly independent base uncovered of starting at the top and working downwards, and finally meeting in the middle.

It seems that both ends should be worked on simultaneously, for best results.




Ref: Near field, far field.
MC Escher - Ascending and Descending.

No comments: