Engineering .vs. Science Pt 2
The Market Ticker - Commentary on The Capital Markets
Login or register to improve your experience
Main Navigation
Sarah's Resources You Should See
Full-Text Search & Archives
Leverage, the book
Legal Disclaimer

The content on this site is provided without any warranty, express or implied. All opinions expressed on this site are those of the author and may contain errors or omissions. For investment, legal or other professional advice specific to your situation contact a licensed professional in your jurisdiction.

NO MATERIAL HERE CONSTITUTES "INVESTMENT ADVICE" NOR IS IT A RECOMMENDATION TO BUY OR SELL ANY FINANCIAL INSTRUMENT, INCLUDING BUT NOT LIMITED TO STOCKS, OPTIONS, BONDS OR FUTURES.

Actions you undertake as a consequence of any analysis, opinion or advertisement on this site are your sole responsibility; author(s) may have positions in securities or firms mentioned and have no duty to disclose same.

Market charts, when present, used with permission of TD Ameritrade/ThinkOrSwim Inc. Neither TD Ameritrade or ThinkOrSwim have reviewed, approved or disapproved any content herein.

The Market Ticker content may be sent unmodified to lawmakers via print or electronic means or excerpted online for non-commercial purposes provided full attribution is given and the original article source is linked to. Please contact Karl Denninger for reprint permission in other media, to republish full articles, or for any commercial use (which includes any site where advertising is displayed.)

Submissions or tips on matters of economic or political interest may be sent "over the transom" to The Editor at any time. To be considered for publication your submission must be complete (NOT a "pitch"; those get you blocked as a spammer), include full and correct contact information and be related to an economic or political matter of the day. All submissions become the property of The Market Ticker.

Considering sending spam? Read this first.

2023-01-29 07:00 by Karl Denninger
in Musings , 750 references Ignore this thread
Engineering .vs. Science Pt 2 *
[Comments enabled]
Category thumbnail

My previous piece on this general topic was on the difference between people who take a science view of things and an engineering view of them.  I explained how an engineering view, when faced with an unexpected failure of some sort, immediately stops whatever they're doing, reverts it if something was recently changed.

A "science" view holds that statistical power is enough.

The problem is that each human is not a statistical problem; you are an engineering example.  That is, there is only one of you just like for any individual road over water there is one bridge.  One bridge failure is not tolerable because there may be cars on the bridge when it fails and the people in the cars will die.

The key to an engineering view is that engineering looks at the world as deterministic and thus statistical failures are not permitted.  That is, it looks at the world like physics does even given our imperfect understanding, and thus seeks to place outcomes well beyond confidence intervals.

Physics says that kinetic energy is always 1/2 mV^2.  Never anything else.  2H2 + O2 -> 2H2O + energy, always.  You never get gold, you always get water.  Gravity (on earth) accelerates all objects at 32 ft/sec^2, always.  It does not matter if the object masses one gram or one ton; that is the acceleration.

Here's the functional difference, and what you must understand: Everything in the universe is in fact deterministic.

But wait, you say -- medicine never seems to be.  Nor does weather.  Nor, for that matter, is so-called "climate science"; remember we were told by scientists the polar icecaps would be gone 10 years ago?  Well, they're not.  The weatherman can't even tell you with certainty whether it will rain tomorrow at 2:00 PM and he's wrong more-often than right.  The NOAA forecasters can't give you where the hurricane will hit in five days and how strong it will be either, or if a tornado will form.

Every one of these outcomes is in fact deterministic -- that is, it has exactly one outcome, just like hydrogen and oxygen always produce water and energy when combined at or above the activation level of heat.

Note that it was, for quite some time, believed that a rock and a feather were acted upon differently by gravity and this, it was believed, was part of why birds could fly.  It wasn't until we figured out how to construct a vacuum pump and remove the air from a chamber we proved otherwise.  That which looked not to be deterministic (gravity) in fact is; it didn't look that way because we did not understand enough of the full system in which it applied.

So why is it that when you are told you have cancer the doctor cannot tell you whether or not he can cure it?  At best he can give you some set of odds.  Ever notice that a medical study that claims a "p < 0.001", that is, almost-certainly the result is allegedly real, still does not claim that everyone is cured or helped?  It doesn't have to in order to get that "p" value -- it just has to have an outcome that is wildly improbable to be due to chance.

It is active fraud to use such to claim you should or must do a given thing.  To suggest a course of action is fine, but to cajole, coerce or mandate it is not.

Why?  Because you're always a trial of one.

So why is it that we can't actually tell you where the hurricane is going to hit a week out, whether the tornado will come (and where if it does), whether the medicine will cure you or whether you need your umbrella in two hours?

Simple: We do not actually understand the thing we are analyzing; at best we have only a partial understanding.  When it comes to biochemistry that understanding beyond the basics (e.g. metabolism of foods into ATP and thus your survival) is in fact quite poor in both depth and breadth.  This is also true for essentially all large-area physical systems on Earth and indeed through the universe.

An example from the planet we live on: Despite the claims that CO2 emissions are driving the warming of the earth nobody, to my knowledge, has been able to accurately provide the carbonate buffering reaction between the atmosphere and the oceans.  This is in fact critical to being able to accurately model anything related to same because there is far more carbon dioxide in the ocean in the form of carbonate than that which is in the atmosphere as a gas.  Without being able to describe this buffering reaction accurately (and several scientific groups have realized, after much experimentation that they have no accurate idea at all how to do so) you can't possibly determine how this will all balance out -- or, even more-importantly, which is the driver and which is the driven element -- or whether both in fact are in different places and times!

We lack the understanding required to be able to accurately describe it and, despite decades of trying, we are nowhere near being able to do so.

Yet in fact the outcome -- that is, the buffering reaction -- is deterministic and, if we understood what we were analyzing we could tell you 100% of the time what was going to happen.  The reason we can't is that we're ignorant of significant parts of what's being treated or predicted.

That's all it is folks.

There are myriad buffering reactions in every living thing.  We do not understand how most of them work accurately enough to describe them with a formula, function or model yet all are utterly essential to survival.  I can list a dozen off the top of my head and every single one of them is essential to the continuation of that organism's life.

The claims of anyone in these fields are not facts -- they're guesses.  Perhaps educated guesses but they are guesses and you must never confuse them with facts.

Policy, especially binding policy, is never legitimately made from a guess.  That is no different than religion; I may believe in God, but that belief is a guess.  I cannot prove it and you cannot disprove it; we both lack the ability to reach a deterministic outcome.  There is one but neither of us know what it is and by the time we know its too late to change our minds.

A climate "scientist" might tell you that we must reduce CO2 because he claims that humans emitting CO2 will cause the earth to get warmer and the sea level to rise, both of which are claimed to be bad and must be avoided.  I'll leave the "warmer" thing out for a minute and focus on the sea level aspect.  Said "scientist" in fact has made two claims, not one: First, that the sea level will rise (the bad outcome) and second that CO2 emissions by man are the cause -- that is, if we stop doing that the sea level will not rise.

This claim has multiple problems not the least of which is that he can't design an experiment to validate his hypothesis because said scientist doesn't have a spare earth laying around that is identical to this one upon which to run his experiment nor does he have the hundred years or more to prove he's right.  Therefore all he's got is back-fit mathematics which do not meet the scientific method that, I remind you, requires a hypothesis, formulation of an experiment in which only the desired variables are changed, recording and analysis of results and then publication of all of it so the results can be replicated by any interested party.

A back-fit model is never scientific; it is not, by definition, a test of a hypothesis.

Remember, the alleged "scientist" has not only claimed an outcome will occur he claims he knows why.  He has no evidence for either of these claims other than a historical back test which, anyone who's worked around any system that has a lot of unknowns (e.g. the stock market, for example) will tell you almost never validates on a forward basis.  These sorts of models don't even have the record of a coin toss; they nearly always fail to be predictive.  There are plenty of people who have blown up their trading accounts believing they have found the exception to this rule and nobody that I've ever heard of who has even a decent record of being right with what they've discovered in that regard in any system that exhibits evidence of non-deterministic, as we see it, behavior.  If such could be done the person who did it would wind up with all the money, obviously.

Let's ask what happens if he's wrong about the reason the sea level will rise?  Let's first presume that he's right in part and the sea level does rise.  He can't prove that his claim of cause is correct, however, as he cannot show determinism; he doesn't have enough facts to produce a deterministic result.  If he turns out to be wrong we take all the costs, societal, economic and otherwise, to reduce CO2 emissions and yet the people will still get screwed because the sea level rises anyway!

That is the alleged "solution" makes it worse than if you did nothing; the people still get hosed by the sea level rise but you first confiscated a large amount of their money by raising the cost of power, transportation, heating, cooling and agriculture so your "solution" screws them twice!

The engineering view of this problem, assuming the fear is that sea level will rise and destroy property and people is deemed both reasonable and worth defending against, is to build walls and otherwise insulate people and property from the sea level rise, or move the people and things out of the way so they don't get flooded out.  That, provided you do so beyond a reasonable confidence interval of said rise and are far enough on the safe side of it, will always work.  That is the engineering solution; it is deterministic in that provided you build the wall to the correct height with the specified materials of a given strength and/or move the stuff the bad result will not occur.

Why the sea rises from the engineering perspective is irrelevant.

When the pandemic hit in the first couple of months it was clear that young, healthy people were at statistically no risk (materially less than the flu) yet older, more-morbid people were at serious risk I put forward an engineering-style solution, albeit an imperfect one as we had wildly insufficient knowledge to get a decent confidence interval, which was ignored.  That is, the sanitarium model which was used for tuberculosis.  That absolutely would have worked far better than what we did because it did not rely on anything that we did not know was correct; there was no element of guesswork in the solution.  The only people allowed in and out of facilities housing said high-risk people would be those who had seroconverted and thus were known unable to acquire or transmit the infection.  We had no choice but to accept the "who's living there and has tested negative now is ok" but that risk only had to be accepted once for a given facility at the very outset when few people were infected at-large.  This meant putting up housing (e.g. rented RVs) at said facilities on site for workers who hadn't seroconverted and paying them whatever was required to work and stay there with food and other essentials brought in and sanitized.  If they rotated out for any reason they could not come back in until and unless they had seroconverted.  As the young, healthy people in the general population got the virus, shook it off and did seroconvert they could be hired to work safely and not have to stay on-site.  Within a couple months with no attempt to contain spread among the low-risk side of the population there would have been tens of millions of available seroconverted workers and those who found the premise of on-site lodging onerous could have been replaced.  There were no unknowns that could result in ineffectiveness; while we might have had some failures here and there due to human mistake (people are not perfect) beyond that it would have with near-certainty prevented the infection from getting into those facilities and very few high-risk people would have died.  Instead we did the exact opposite in several states and shoved infected people into those buildings, attempting to rely on masks and testing to prevent transmission. The masks and testing regimes repeatedly failed as they were based on statistical reductions which we had every reason to believe was irrelevant (once you cross the threshold of enough virus to become infected how much more gets into you doesn't materially matter with a virus since viruses replicate exponentially) and many died because the fools implemented a scientific method focused on probability and statistical reduction in emitted particles rather than an engineering approach that relied on deterministic process designed to be well beyond reasonable confidence intervals.

Engineering is always deterministic because it has to be as a discipline; if its not people die and the engineers who did not employ deterministic methods are held responsible for the failure.  Scientific methods only are deterministic and thus interchangeable on a functional basis when all the variables are known and correct.  The practitioners of scientific methods are almost-never held responsible when they're wrong; when was the last time a hurricane forecaster was charged with manslaughter when he incorrectly predicted where the storm was going or failed to predict the intensification of the Cat 2 storm to a Cat 3 or 4?  The engineering answer is always superior since it does not rely on that which is not known to be correct -- or even known at allIf I do not know, for example, what the physical load a bridge pier that is driven to 80' can take without displacement under a specific set of conditions then I have to measure that before I can accept 80' as a suitable depth.

The scientific answer often kills people when all the inputs and variables are either not known or incorrect because it is a guess and guesses are frequently wrong.  The more unknown variables the worse the guess will be.

Facts are absolutes.  Physics is a set of facts.  Chemistry is a subset of physics, when you get down to it; it describes the physical interaction of atoms and molecules, which are comprised of protons, neutrons and electrons (and then subatomic particles beyond that.)

If and when we ever manage to understand biological systems sufficiently we will reach the point of determinism in medicine.  We will not say "you have a chance of beating this condition"; we will know what the outcome will be and whether the condition can be resolved or not -- and if so what you must consume or do to resolve it.

The same is true for "climate"; what is currently proclaimed may be a scientific process but it is not engineering and must never be used to drive policy because we simply do not understand what we're studying well enough to make accurate predictions nor establish causation.  The predictions that have been made have almost-all been proved wrong and as such they don't even qualify as educated guesses.  To make policy decisions on that basis is to make the wrong decision in virtually every instance, that is to screw people in some form or fashion who then have the bad thing happen anyway.

Some day we will reach an engineering level of understanding when it comes to medicine, climate, and many other things -- just as we have with chemistry and, at least at the atomic level, physics.

That day is not today, whether we are talking about climate, weather or all manner of biological things around us - including medicine.

Go to responses (registration required to post)
 



 
No Comments Yet.....
Login Register Top Blog Top Blog Topics FAQ
Page 1 of 68  First123456789Last
Login Register Top Blog Top Blog Topics FAQ
Page 1 of 68  First123456789Last