19 November 2005The Anthropik NetworkBenjamin Shender
A while back I came to realize that people seem to want things to be complicated. Simple solutions are uniformly treated with skepticism, while complicated solutions are hailed as being salvation. And this routinely repeats itself despite comparative success rates. Simple and elegant solutions always work better than complicated and contrived solutions. Why? As you might expect, the answer is very simple.
The universe runs on very simple. Evolution, very simple. Gravity, very simple. Light, very simple. Every time a scientist comes up with a complicated way to explain something, a better scientist comes along later and explains it better and simpler. The universe seems to run on the axiom: keep it simple, stupid, if only because complicated things tend to run afoul of entropy, another simple thing. I think civilized humans just like making things complicated because it fits into the world view better than simple things. Perhaps the words of Niels Bohr will help us here: "Stop telling God what to do!" The universe is what it is; making it something it is not is quite the silly waste of time. There isn't enough magic, beauty, and wondrous awe in the universe as is you have to make it excessively complicated to appease your idea of what the universe "should be"?
Think of various problems you have solved. Think of which solutions you are most proud of. I bet that they're the simplest ones, the ones that work with the problem to solve it. I know mine are. Alternatively, civilization seems to run on contrived solutions. Instead of systemically solving problems we tack on a "fix" to the system that is already there. Which causes a substantial increase in complexity.
Take recycling for example. We found that our landfills were filling up and that we were running low on resources. So we tack recycling on to the end of the manufacturing process (gather resources -> create object -> use object). We do this instead of manufacturing items with multiple prolonged uses. Baby food jars are an excellent unintentional example of this. Once the food is eaten, the jars are the perfect size to store countless small items. We recycle instead of making more items that biodegrade. We recycle instead of not manufacturing disposable items. Think about the last one, if items weren't manufactured as disposable and marketed that way, we wouldn't throw them away unless they actually broke. We only throw away glasses when they are dropped and don't survive the sudden stop. What if we didn't make disposable grocery bags? Then everyone would simply buy and use the reusable ones, what other choice would they have? And we wouldn't throw away 500 billion to 1 trillion plastic grocery bags every year worldwide. A simple solution to a problem. But instead we have instituted programs to recycle plastic grocery bags.
What am I suggesting? Here it is plain as day:
Steps to Solving Any Problem Well:1) Analyze the problem, don't assume you understand the problem, think about it first.2) Clearly state what you are trying to accomplish, your goal is not the first solution to come to mind. The goal isn't "recycling" the goal is "to eliminate waste."3) Brainstorm solutions.4) Any tacked on or excessively complicated solutions should be eliminated as poor.5) Try to work all the angles of whatever solutions are left, if they have multiple benefits and few if any costs, it's a good solution. It's the perfect solution if, once implemented, it will sustain itself with no need for later reinforcement.6) Implement the solution.
There you go, simple as can be. I'm suggesting that solutions should be systemic. Meaning they work as a system with other systems and are self-reinforcing. Programmatic solutions, meaning solutions that involve adding programs to work against the systems already in place, should be avoided when at all possible.
BTW, don't bother critiquing the recycling thing; I came up with that on the fly. Little to no thought went into it. If it's really bad, no thought went into it, but it's still an example of thinking systemically rather than programmatically.
complexity, programs, system, systems thinking39 Responses to "Keep It Simple, Stupid"
Janene says: November 19th, 2005 at 10:55 amHey Ben --
Yes, yes, yes:-)
Another point that you hit sort of sideways... cycles are (generally) preferable to anything linear.
Back to recycling... that sounds cyclical, right? But the way we do it is not. We recycle materials into 'different' products. So instead of recycling paper into paper, we recycle it into 'recycled paper' with different qualities that are either 'prefered' (if you are liberally inclined) or 'crappy' (if you are conservatively inclined).
Jim and were talking -- and I believe this is being done in some cases in Japan -- what if electronics (appliances, computers etc) manufacturers made all of thier products so that they could be de-compiled with one simple tool. Then, they could take back 'old' products for a small returned deposit, and reuse the majority of the components directly with little or no additional modification. Recycling costs would plummet, production costs would plummet and the whole system would be 'simplier' 
Janene
JCamasto says: November 19th, 2005 at 2:34 pmIn terms of total system efficiency, energy, and complexity, the mantra has always been:
Reduce, Reuse, Recycle. In that order.
Remove a speck of "use" from the top of the production pyramid, and you eliminate the whole section of infrastructure underneath, supporting it.
Further, very little can be truly recycled without either: material degradation or large inputs of energy.
The earth takes care of reusing and recycling air/water/earth for us. While our culture's production economy looks upon this a "freebie" or externalized cost - we know better... We're stealing down the earth's capacity to provide these "free" services and using the resources to make, essentially, only human shit.
Systemic limitation (crisis) approaches...
qrswave says: November 19th, 2005 at 3:20 pmSimplicity is always the best solution.
But, you are assuming that "solution" is the goal when in fact it is consumption.
If left unchecked, greed consumes everything.
Benjamin Shender says: November 19th, 2005 at 3:58 pmBut, you are assuming that "solution" is the goal when in fact it is consumption.
Sorry, what does this mean?
qrswave says: November 19th, 2005 at 5:50 pmYou said: "...solutions should be systemic."
And, I was just pointing out that finding a solution is not everyone's goal.
Those who make money from more consumption will always want more and more of it, regardless of the consequences. Consumption is their goal.
Utter disregard of consequences is characteristic of the malignant form of selfish greed that pervades our current economy.
McKinley says: November 19th, 2005 at 6:50 pmLeaving aside that perhaps there should not even be grocery stores or cars...
You pull into the parking lot and park the car. You grab a shopping cart and head back to the truck of your car. You take out the familiar wooden crate with its equally divided partions. Each partition holds one or more glass jars. The jars look like something between a caraf and a Mason jar. Your crate has a one liter jar which fills an entire partition, two half-liter jars in another partition, and four quater-liter jars in yet another. In two of the empty partitions you have crammed a few canvas shopping bags. One of the partitions has a stack of 11 metal lids.
Placing the crate into the shopping cart you head into the store. A greeter smiles and takes your crate. He hands you back your canvas bags and a small token with 5 small dots (3 for the jars, 1 for the crate and one for the lids) on it. Then he places an empty crate in your cart. You ask for another. It is a big shopping day for you.
You procede to shop placing produce into your canavas bags. You grab a liter of milk, a half liter of butter, a quater liter of sour cream and a quater liter of yogurt. Each comes in the same shape of jar. You know which is which by the color of the metal lids on each jar.
After placing some bread into your canvas bags you grab a jar of jam. You grab a shallow eighth-liter jar of creamed honey with its familiar yellow and brown lid. Each time you take a jar, you place it into your crates. When the crates are nearly fully, you head to checkout.
You place your crates on the counter and the checker peeks at each partition making a count of all your jars. When she cannot make out the colors of a lid, she just lifts the jars on top a little. After all your groceries are counted your total is brought up. Your hand the checker some money and a few tokens including the token that you collected on the way into the store. She hands you your change and you exit the store.
Benjamin Shender says: November 19th, 2005 at 10:42 pmYou said: "...solutions should be systemic."
And, I was just pointing out that finding a solution is not everyone's goal.
Those who make money from more consumption will always want more and more of it, regardless of the consequences. Consumption is their goal.
Utter disregard of consequences is characteristic of the malignant form of selfish greed that pervades our current economy.
Ohhhhh, but still, they have problems that they need solutions for. Solutions to how to achieve one's goal, whatever that is.
qrswave says: November 19th, 2005 at 11:16 pmI see.
I thought you were talking about sustainable solutions.
michael says: November 20th, 2005 at 12:33 amI disagree with the broader implications of this post and the idea that simpler-is-better in general. Maybe this is just semantics or confusion on my part, but here goes...
1.) Healthy ecosystems are complex and must maintain their complexity. Stable and biologicaly rich ecosystems (especially those surrounding the grassland/forest ecotones where humans emerged and spent most of their time evolutionary speaking) are composed of an arguably countless number of species and, by extension, interactions between species and their surrounding hydro/geological resources. Simple ecosystems, i.e, those with fewer species and subsequently lower levels of biotic interaction are unstable and tend to either collapse or diversify as species from neighboring ecosystems move in. Of course some radically simplified ecosystems (monoculture) can be maintained if we engage in constant biological suppression through water diversions, the elimination of pests, weeds, etc.
The problem isn't complexity; the problem is humans who seek to simplify the incredible complexity of natural bio-geo-chemical cycles.
2.) and to speak directly to this quote:
"The universe runs on very simple. Evolution, very simple. Gravity, very simple. Light, very simple. Every time a scientist comes up with a complicated way to explain something, a better scientist comes along later and explains it better and simpler."
While "evolution" as a *concept* could be described as simple the actual process is not. The map is not the terrain in much the same way that our concepts/words are not the things they represent. Our concepts, especially in the scientific realm, are simplified abstractions that generally do not do justice to the processes they represent.
Also, our concepts of gravity and light have dramaticaly increased in complexity, and while we (arguably) have come closer to an understading of these phenomenoa we still can't explain their tendencies to act differenly at macro and sub-atomic levels.
Overall, I think that our scientific explanations have become more complex since E=mc^2, and much more so since the days of a geocentric "solar" system.
Benjamin Shender says: November 20th, 2005 at 2:16 amSystemic solutions are sustainable.
Michael:
1) I don't really think of ecosystems as being that complicated. Stop trying to think of how everything can work together. Accept that it does it all makes sense. Simple sense. A forest. I add more water. The water table rises, the growth of certain plants increases, while other plants have problems. The animals related to these plants have problems, or an advantage. The animals related to those animals. etc, etc, etc. It's not complicated, simply connected. Civilized people think of it as being complicated only because civilization teaches us to think only linearly. Which is silly, because nothing is really linear.
2) Light and Gravity are fairly simple too. We're in a complicated phase right now, but it won't last much longer. The last time this happened Maxwell brought all the relevant theories of the day together by adding a dimension on to the matrix. Simplfing everything, and it made for better predictions too.
Evolution is simple. If your genes suck you die. How much simpler can it be? If certain traits get you more tail, then your representation in the gene pool increases. It's not complicated until we try and make it complicated, and that usually only happens when we start trying to give evolution a goal, of which it has none. Or almost none, your genes have a goal though: "have sex, will travel."
LucidWanderer says: November 20th, 2005 at 10:21 amVery nice! I love this one, Ben. I've been doing a lot of thought on the subject of solutions myself, and might even post something. Here's some thoughts:
Complexity (a dynamic nonlinear system) can exist as a result of simple actions and forces, but the contrivance of a complex solution will not be as effective as an understanding of the system's function with a simple solution:
Think globally, act locally. Put a cog in a huge machine to stop it from working, as its a big system, but once you find the critical point, you can affect it with a relatively simple solution if your goal is to stop it from working.
Janene says: November 20th, 2005 at 10:26 amHey --
Michael, I think the 'simple attribute' that Ben is trying to get at is that the more complex, rich, varied something is, the more likely it is comprised of many simple components.
Take ecosystems. It is not 'complexity' that beans secrete a sugary substance in thier roots. Nor is it 'complex' that certain bacterias like to eat those substances... but when the two work together, you end up with nitrogen building which enables the entire plant community to excel including both the plants themselves and insects, birds, mammals, reptiles amphibians... in other words, all life benefits from the simple sugars produced by a chick pea.
On physics... I can't prove this, but I suspect, down in my bones, that one day they will discover (or not, depending on future history) that C is not a constant... rather, that it is a simple equation that they have treated as a constant. And so all of the complex relationships that they have tried to account for can be better, and more simply explained by that slight variability of C. So is this more complex or less? It is all a matter of perspective, really.
Nested Sets are very complex phenomena... but they are created through the application of very simple equations (or commands, biological and otherwise). Consider the tree form. Very complex to our eye -- looking at a tree in our back yard. But the list of equations that create that complexity -- very simple.
And so on and so forth....
Janene
michael says: November 21st, 2005 at 3:51 amwell, one big part of my critique of civilization is that it seeks simplicity and standardization rather than complexity and diversity. Stable ecosystems are more complex than cornfields and that isn't just the "linear thinking" taking . Ecosystems are much more complex than the nitrogen fixing properties of legumes; there are literally thousands of overlaping (and overlaping to various degrees) cycles impacting and being impacted by all the other cycles around them. The whole is greater than the sum of its parts!
Civilization is uncomfortable with this; we (the civilized) need standardization (solid pavement and rows of corn). Our built environment is standardized and simplified to the point that natural processes are supressed or excluded....To make a long theory short: A lack of experiential knowledge of one's place in the larger complexity of an ecosystem damages the human mind and implants the narrative of suppression onto ones psyche, i.e., "I must supress the organic complexity of myself and perform tedious tasks from 9-5 to be a 'productive member of society'"
I wish I had more time to continue discussing this, but I'll check back later.
michael says: November 21st, 2005 at 4:18 amok, just one more point
quote:"Very complex to our eye -- looking at a tree in our back yard. But the list of equations that create that complexity -- very simple."
the list of equations that create that tree would be enormous: what created that tree is billions of years of evolution and co-evolution/adaptation to predators and diseases that were also adapting to it, as well as climatic shifts and changes in soil composition. Even looking at the individual tree and it's life span without taking into account its evolution, your list of equations would have to include everything from the structure of atoms to the solar system. Even without taking into account chemisty and astronomy, the list of all the biological inputs and interactions that are occouring and influencing the tree would be anything but simple.
In fact, many of our environmental problems are a result of our lack of ability to acknowledge just how complex the natural world is.I'll list some concrete examples when i get more time...
Janene says: November 21st, 2005 at 9:27 amHey Michael --
Hmmm.. I don't think I'm explaining myself well.
OF COURSE the natural world is very complex... much more so than we will ever be able to diagram, recreate or fully explain.
But all of that complexity is created throught the interaction of billions of 'simple' principals and organizing 'rules'
I'm not disagreeing with you that this complexity exists, I'm saying that there are different kinds of complexity and differenet definitions in use in this discussion. You are looking at the macro, top down, 'wow is that complex' side of this circle... and that is totally valid and I dig what you are saying. But I'm looking at micro, the bottoms up, 'what did each step along the way look like' side of the circle and each of those bits and peices is simple. Its elegant.
So of course an ecosystem is many millions of time more complex than the sugary excretion of a legume plant. But it is complex because of simple things: typified by that simple sugar.
When I mentioned the tree, I was actually refering specifically to its form -- its shape. That shape is recreated over and over and over in nature... and the rules that govern the making of that shape are very simple. Yet it LOOKS complex.
So I think Ben is suggesting that we mimic nature by creating simple structures that have the ability to produce complex results
Look at the example I offered in the first response above. Replacing the rivets and glue and soder in a computer with removable screws is a simple thing -- in practice, all it requires is a change in materials that already exist. So implementing it is as close to cost free as it gets (sure, yes, some restructuring of purchasing, retraining of personnel, perhaps a little retooling, but relatively minor) -- but the result... 90%+ reuse of components, reducing the energy needs of 'initial production' by adding a tiny bit of energy to take apart and sort returned PC parts, decreasing the amount of raw material needed for each subsequent generation, removing all of these materials from landfills (perhaps eventually making 'recycling' and 'landfill management' obsolete if similar ideas can be adopted everywhere) and so on and so forth. Fundamentally, you can look at the energy cost, and that cost would plummet.
Janene
Jason Godesky says: November 21st, 2005 at 9:48 amComplexity is the result of simplicity, just like predictability is the result of chaos. The result of cultural complexity is ecological simplicity (like monoculture). The result of cultural simplicity is ecological complexity.
It's an undeniable fact that civilizations are more complex than tribes. There are most social roles, because social roles exist; there are more kinds of material artifacts; there are whole new fields of thought that need to be expressed. For example, the whole vocabulary of theodicy is completely irrelevant to animist thought. Civilizations are far more complex than tribes. To support that level of complexity, they need a certain absolute amount of energy running through, and a complex ecology that nurtures many kinds of life can't provide that: a simple, monoculture field dedicated solely to us, can.
Benjamin Shender says: November 21st, 2005 at 11:31 amOk, I'm back in. Let me try this one more time. I'm say two things here.
1) Nature creates simple sturctures. A tree may seem complicated to us, because we're very stupid. Much stupider than the gods. But believe it or not, that tree is in it's simplest state. Every thing is. Nature doesn't work by making things excessively complicated. Everything in nature moves towards it's lowest energy level. From atoms to stars. Trees are setup in the simplest way that still works. And all the rules that made it that way are also simple. Evolution is simply a fancy way of saying that anything that doesn't work due to excessive complexity or excessive simplicity is recycled into the raw materials to try again. Most traits in the middle stay for a very long time, just in case. Dawn syndrom has been in the human genome for a long time, because it doesn't cause the human in question to not function. But in a couple centuries the situation might change to make dawn syndrum a huge advantage, and everyone else gets it. A tree is in the simplest way it can be. Humans too. A star. Hydrogen. Hell, take a good look at chemestry. Electrons in an atom get bumped up an energy level all the time, and they immediately go back down releasing energy. Nature knocks anything excessively complicated down. Nature knocks anything excessively simple down. A tree may look complicated. But I triple dog dare you to find a more efficent way to collect sunlight than the pattern of leaves has already come up with.
2) Cultures are natural and subject to the same resrictions. Excessively complicated things get knocked off, in a universe with entropy they're too large a waste. Excessively simple things can't compete. Really simple. Now lets think about how this interacts with problem solving, another natural function. (I'm going to stop saying that and just say everything that exists is natural by defintion. Supernatural is a null word.) And so the best solution would be the simplest one that solves the problem. Which, if follow the logic, would ultimately have to be a systemic solution. Don't tack something on to "fix the problem," the extra complexity is a death wish. Add or change something that alters the entire meaning of the system. Look back at "Systems Thinking and the Food Race." The bottom part is the way it is for all speices. We added something that changed the whole meaning of the system from one of balance to one of perpetual growth. And it is a systemic solution, just not a very good one as it turns out. Which is why it's so hard to fix, it's a part of the system, not contrary to the system. Adding on recycling and encuraging people to do so is hard, getting rid of it is easy. If recycling disappeared from the US, most people wouldn't notice. It's not a part of the system, it's added on. Birth control is added on. Etc. Etc. Etc. The way to solve a problem to insure a fix is to deal with the system in question at a metalevel (nothing as simple as food = people, that ignores that +people = +required food) and try and understand what can be added, changed, or removed to alter the system to the way you need it to be. In the food race the simplest way would be to knock out "more food," which would mean that, after a short delay, there would be fewer people until the balance is reacheived. This is not the most...appetizing solution perhaps, because it would require much in the way of death, but it is a solution. Daniel Quinn's solution wasn't bad either, slow down your response of "grow more food" until you've reintroduced a balance. And then slowly bring it down. The only problem with this solution is psychological. Even if they aren't people will scream about starvation. The first famine the year after would be blamed on this new project, the fact that the famine was right on schedule and would have happened anyway (probably worse too) would be utterly irrevelvant to a population that can't grasp something as obvious as evolution. I believe a third systemic solution would be useful.
michael says: November 21st, 2005 at 8:11 pmokay, thanks for the claification Janene, I have a bit of trouble with your pharase, "...the rules that govern the making of that shape are very simple. Yet it LOOKS complex.", because this can be said about anthing, say nuclear weapons, not just trees. I think ultimatly our disagreement on where to place things within our little simpicity v. complexity dichotomy is largely due to our definitions of said words and our differing viewpoints ...for example, Jason says : "It's an undeniable fact that civilizations are more complex than tribes."I agree with this (except for the "undeniable fact" part) on many levels, but at the same time there are many aspects of civilization that are grossly simplfied compaired to the tribes that once occupied the same geographical area..
