On Tue, 19 Nov 2002 08:31:55 Jim Elwell wrote: ... >Sorry I was not more clear, Marcus.
That's alright, Jim. Your usual point-to-point approach below does work a lot better, thank you! :-) > Here is what I am trying to say: > >(1) You claim that the management tool of "optimization" must take into >account all peripheral constraints and parameters. > The taking of 'all peripheral constraints and parameters' is a concept that would have the best chance of yielding the best *model* to study a phenomenon. The more one does it the better one would expect the emerging model to be. Now, sometimes one may not *necessarily* need to be THAT comprehensive to come up with a decent *working* model (i.e. a model that works well!). However, it's generally true that *especially* when one studies human-related issues the above is indeed desirable if one is to find such "decent" model. >(2) I claim, with evidence from Frederick Hayek and W. Edwards Deming (and >you apparently agree), that much of the necessary information in a complex >system such as the US economy is both unknown and unknowable. > That may indeed be true, but when faced with 'unknowns' or 'unknowables' one must *try* to *simulate* parameters/definitions that may serve as good **explanatory** tools for phenomena under study. This particular aspect is something that scientists do a lot when they try to 'explain' certain things. There is a term in scientific jargon that one uses for this but it's unfortunately escaping me right now (sorry... I've been trying hard to remember what it is, but the word just doesn't come up, darn...). I think it's heuristic, but I'm not 100% sure. The other possible term that just popped in my head now is called empirical! Anyways, the likes of BMI factors for example fall under that category. I.e. people "come up" "out of the blue" with some "concocted formula" that ends up explaining reasonably well a phenomenon under study. So, in the absence of some "hard data" this is an actractive avenue to explore. >(3) Therefore, the logical conclusion is that the management tool of >"optimization" is not appropriate to this particular problem, i.e., >metricating the USA. > Again, I don't think I was talking about using optimization as a tool for 'metricating the US', Jim, but rather discussing the issue of 'maximizing freedom' you volunteered to discuss. But, ok since you're at it, please allow me to give you my humble personal opinion on this use of optimization tools. Optimization techniques are extremely powerful tools in management science. It's extremely valuable especially in light of very competitive environments in which we live in in the business world today. However, such tool is only as good as one is able to create and simulate a model that mirrors the phenomenon in question!!! Therefore, when faced with difficulties like the ones you so aptly volunteered here one would have to resort to clever alternatives that MAY yield satisfactory results. Evidently BEFORE a proposed model is used one would have to TEST it to make sure it's performing well for well controlled experiments and data. Once that stage is done and we're happy and confident about it, THEN we try to use it for REAL LIFE situations. Using this technique to find out the best approach to metricating the US would certainly be a very challenging one, a one I must concede I'm not certain would be successful. So, I'd probably prefer to rely on experiences outside the US to make a recommendation on a course of action for it, hence my insistance on proposing the "regulated approach", since they're known to work and/or to have worked in the past! >Marcus, I am not sure we are ever going to agree on #3. So I'll just ask >you this: > >If you were made USA Measurement Czar for a day, and you did your academic >magic and applied all your optimization tools and took into account all of >the parameters that were available, etc., and the end result of your >analysis is to the effect of: > >"The most optimal solution is to metricate the USA over the next 20 years." > >Would you be willing to accept this? >... With the most enthusiastic YES, my dear friend. *PROVIDED* we would, at the end of those 20 years, arrive at a **FULL** metricated state of affairs in the US!!! I'd have no problem with that. :-) (but our bet was for 10 years...) On the other hand I continue to doubt that this would ever happen WITHOUT strategic plans of action *to get there*! Do you know what I mean? Planning, organizing and implementing IMHO are *everything*! I can't see something as intricate and complex to do as metrication requiring anything less to work, or to successfully happen in a totally uncontrolled chaotic environment. Regards to you, too, Jim. Marcus ____________________________________________________________ Get 25MB of email storage with Lycos Mail Plus! Sign up today -- http://www.mail.lycos.com/brandPage.shtml?pageId=plus
