## Monday, August 11, 2014

### Why your blog sucks

Math isn't just fancy calculations that cause people to have violent seizures upon reading. Calculations are a recent thing in math, like enlightenment age recent. Story goes, Leibniz was hella good at calculations and people just wanted to be like him. His rival, Newton, based at least one of his books on geometry. The statistics that people are so fond of using in their blog posts to back up a claim are based on calculations; some simple that are extensions of ratios that have been used since the Greeks, and some hard $\mathbf{F}\left(y\right)=P\left(X distribution type calculations. If you ask me a year ago, I would have claimed you shouldn't discuss these stats if you don't understand math. That was based on a pet peeve of mine where people claimed to dislike math then quoted stats at me to back up arguments. The idea you need to understand something to use it is ridiculous, everyone reading this is using a computer, but only a small handful of you understand how it works. Instead of taking a Prop and Stats class just so you can post a number backing up your article on gun violence, I will talk about something that's normally glossed over in beginning math: Logic and reasoning.
I'm going to start off by stealing and referencing John Allen Paulos. Mathematicians can do this, because there is pressure to be original, but work is built off of others, to a degree. And not referencing someone with more knowledge and experience would be hours of work that I will never have. He wrote a book called "Logic of Stories" and it's great. It's short, like less then 200 pages, and it's written in a way that flows and keep you turning pages. He also condensed it for a New York time article and a TedX talk. He has a much less crass way of stating my thesis: "In listening to stories we tend to suspend disbelief in order to be entertained, whereas in evaluating statistics we generally have an opposite inclination to suspend belief in order not to be beguiled." When dealing with the numbers, the first important rule is "Correlation DOES NOT equal causation." (The second rule is something about how the rules must have recursion.) If I told you that gun violence rose 80% in ten years, and then said live panda births dropped by 80% during the same time, I would hope that your next question would be "So?" That's a ridiculous enough of an example that someone would hopefully see right through. However, I could pick something less ridiculous like "average education" and "chances of bicycle being primary mode of transportation" instead. As the average education of a population drops, the use of bicycles increases, so getting rid of bike lanes will help with education. See any problems there? A town where the education isn't great on average may also indicate that income of the population is lower, so bicycles are used more because people can't afford cars. Bicycles don't cause people to be stupid (I hope not at least). In this case, and in a lot of cases I see, it's not the numbers that are problem but the logic.
Numbers are like words and letters. They are completely meaningless symbols until people give them the rules and context to make them powerful. ΔX >= ΔY isn't a something that a lot of people know, and even if you do know what those symbols mean you're aware of the fact it's gibberish unless I define it. The concept is similar to how you use words in your posts to make a point. Words aren't evil, but they do convey certain meaning to certain individuals; dropping an f-bomb is fine when I'm talking to 20-somethings at a bar, but I'll clean up the language a bit around a mix group of children and grandparents. "7 Dirty Words" is the definitive example in my mind about using words and how words meaning changes on context and how words change over time. That George Carlin is a scholar. With that knowledge, men like Churchill and Martin Luther King were able to move large groups of people. Numbers can be used to the same affect. You can put them in their like the wanna-be intellectual who uses big words without understanding their meaning, or you can use logic and reasoning, the basic building block of good mathematics, to change that paper into a post that cuts with surgical precision.
While I'm on the subject of words, you need to understand that different groups of people use words differently. Average does not hold the same meaning to the layman that it does to a mathematician. My second favorite statistic is "The average American makes about \$90,000 a year, but the majority of Americans make less than that." (When reading that last sentence, keep in mind my favorite statistic is "87 % of statistics are made up on the spot".) Given a group of numbers {2, 3, 3, 3, 20, 5}, you can see that the common number is 3, but you can't see that the average is 6. Thinking back to 3rd grade math, you remember that you add 2+3+3+3+20+5, then divide by 6, since that's how many numbers you have. That 20 really screws up the average. So beware the phrase "On average. . ."
I really, really want to just talk for hours on prepositional logic. It's what made me fall in love with math, so it's like that lover or child (not a child-lover, you sicko) that I won't shut up about. I have to keep it short and really just keep to the surface. And with that, I'm just resigned to saying "Watch out for Non-sequiturs". Basic logic that has been around since Aristotle falls under the pattern of "If statement A is equal to statement B, and statement B is equal to statement C, then A is equal to C." If that last statement was "then A is a banana" then it would be a non-sequitur and just plain surreal. Aristotle, the father of logic, did it though. His were not surreal, but see the Russell quote at the top of the page. A problem of his was that he would draw conclusions based on "common knowledge". Try reading some of his papers, and you'll see some great examples of logic, but you'll see more stuff about how weird ancient Greece culture was. Views on women, sex with boys, and how smaller penises are better spring to mind. This problem still occurs when people start using statistics. If you come at a number or a set of numbers or even some problem with some pre-established prejudices,  statements like "red states have lower IQs" or "black people only use government hand outs for drugs" start getting tossed around. Extreme examples in both cases, but ask yourself if a conclusion follows logically, or if the conclusion suffers from bias.
Using numbers in a post will help your writing standout more. Using numbers and logic will change it into something beautiful, something magnificent, something actually fucking coherent. And let's face it, there is always someone smarter than you. Einstein had Gödel, you know. Some of those guys want to keep you in the dark because it gives them the advantage. Knowledge is power.