Slaughtering Sacred Cows: Simplicity

Simplicity is good, but it isn't everything

Just because something is a sacred cow doesn't mean it is a bad thing; rather, it could merely be over-rated. Simplicity is one such idea, and I will explain its paradoxical nature from all four sides of the STEM fields.

Now then, before I start, if anyone reading this has ever read The Simplicity Cycle by Dan Ward, please forget everything in that book. Goodreads might call Ward an "award-winning engineer," but I'm going to let you in on a secret: regardless of his education, he has never worked as an engineer, unlike yours truly. If he had, then he would be aware that real engineers don't like to over-complicate things, and a real engineer would have either reduced his 200-page shit-piece down to an article no longer than the one you're about to read, or would have marketed the book as "history" rather than a "field guide." For those of you who are not engineers and don't already know this, the real engineering adage isn't "keep it simple, stupid," rather, it is this:

A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away. - Antoine de Saint-Exupéry

Instead of beginning with science in order to illustrate my point, I'll start with a couple pieces of technology which will hopefully pique your interest. There are two instances I know of when a complicated mechanism was developed in order to perform a certain task, only for a much simpler mechanism to be developed for that exact same task at a later date. The first example is the automatic apple peeler, which was invented in the late 19th century. Here is the video that made me aware of this early version:

Undoubtedly, some of you have seen a much simpler machine that performs the same task, which was invented roughly fifty years later. The strange part is that absolutely nothing prevented the later design from being built with the technology from the earlier period, and that holds true for the other device that I have in mind as an example: the gun.

The wheellock ignition mechanism was invented at the beginning of the 16th century, and used on both pistols and early long guns called arquebuses (muskets didn't appear until the 17th century). They were rare, and wielded only by nobility in a bizarre era of warfare known as the "pike and shot period," which is one of several reasons that I am so fascinated by the Thirty Years' War. The wheellock was such a complicated mechanism that only watchmakers could manufacture them, hence being so rare. Most guns of that time used a matchlock ignition, which was much simpler at the cost of being much less reliable, much slower to fire, and completely useless if the humidity was too high, because the priming powder had to be fully exposed immediately prior to firing. The matchlock is fairly self-explanatory, but if you're not already familiar with the wheellock mechanism, I'd recommend this video:

Roughly two hundred years after the wheellock was invented, the flintlock had fully displaced both it and the matchlock. The flintlock is much simpler than the wheellock, thus facilitating mass production of locks (not all gunsmiths made their own locks, just as some gunsmiths today don't make their own barrels), faster to fire than the matchlock, and since the priming powder remains covered right up until the trigger is pulled, it is much more reliable in high humidity (still not 100% though, I ought to know, I own one). There is also absolutely no reason that the flintlock could not be manufactured using tools and techniques that were available in the 16th or even 15th century, considering some of the rather impressive mechanical devices that smiths of the time could produce, such as Götz von Berlichingen's iron hand, made in 1530. In fact, the earliest mechanisms that we would recognise as "proto-flintlocks" appeared within a century after the earliest wheellocks, but they didn't become popular until much later. Now, I could go on about certain firearm innovations being far older than most people realise, but then you'd be here all day. If you're really curious, just peruse the Forgotten Weapons YouTube channel. BTW no, the Viking Atgeir wasn't a gun, that was an elaborate April Fool's joke. The real Atgeir probably looked more like a glaive (which would have been a very unusual weapon for the time), assuming it existed at all.

I'm sure there are other mechanisms out there that illustrate this basic principle (certainly, steam engines and internal combustion engines alike have changed quite a bit), but now I need to move on to maths and science in order to illustrate the shortcomings of simplicity. There paradox here is best illustrated by the counterpart to Occam's Razor: the more you learn about anything, the more complicated it always turns out to be.

The star of this particular section is Sir Isaac Newton, who invented not only mechanics, but also calculus, which was necessary in order to explain his astronomical observations. 

But Sasha!

No! I am not talking about Leibniz here, particularly since modern calculus is mostly Newtonian in origin. Anyway...

Newton almost single-handedly invented mechanics... almost. He neglected to tackle the subject of energy, but that's not really relevant to what I'm talking about. In Newtonian mechanics, the only types of orbits mathematically possible are conic sections, i.e. circles, ellipses, parabolas, and hyperbolas. In order to map out an accurate prediction of any orbit, one must be able to determine the change in both position and velocity, plug those into the formula for universal gravitation (which Newton derived), and then, calculus allows us to derive a formula for the exact path of any given body's orbit. Granted, this works only when there are two bodies involved; it is insufficient to map out epicycles, and thus cannot be used to predict the orbit of a planet's moon relative to the Sun. Newton's method made for near-perfect predictions of the orbits of every single planet except one: Mercury.

For centuries, it was assumed that the perturbations in Mercury's orbit were the result of there being another planetoid nearby exerting force on it. No such object was ever found, and so a crisis arose. Newtonian mechanics works in 99.9% of instances, so to say that it is worthless is incorrect, but there are other factors at play. These factors have a negligible effect in 99.9% of instances, but a rather noticeable one in the other 0.1%. Albert Einstein put forth an explanation in 1915 called general relativity. The primary difference between general relativity and Newtonian mechanics is that the latter assumes gravitational fields to be flat (three-dimensional space), whereas the former assumes them to be concave (four-dimensional space), and the difference in depth greatly affects gravitation in close proximity to massive objects, such as stars. This also ties in with special relativity (which actually came first, despite its name), but I'm not going to get into that here. Universal gravitation still applies, but distance and mass are not the only things that must be accounted for. In other words, Newtonian mechanics can be explained in terms of general relativity, but not the other way round. On the same token, it is possible to explain algebra in terms of calculus, but not the other way round. The same is true of virtually every scientific paradigm shift. 

I'll concede that I'm not the greatest science teacher, so I'll simply leave you with the source material: part of my physics curriculum while I was being homeschooled was a series of VHS tapes (bloody hell, I'm getting old) created by Dr David L Goodstein of the California Institute of Technology called "The Mechanical Universe and Beyond." The first video below is one about Newtonian mechanics as it relates to the solar system. Feel free to skip it if you already know basic physics. The second video is one about Einstein's refinement of those same ideas. This series never actually gets into general relativity, only special relativity, but for now, just keep in mind that gravity has an effect on the flow of both light and time. Yes, it really is that weird, and even though Einstein's theory isn't perfect, for the most part, it has been confirmed.

Hopefully you've learned something. I'm going to wrap this up here, lest I over-complicate things. Simplicity may be something you want to strive for, but sometimes it simply isn't achievable. The simplest possible solution to a problem might still be unfathomably complicated at first glance. Complicated tasks require complex tools, and don't let anyone tell you otherwise. Na shledanou!