Slashdot Mirror
Why Programmers Need To Learn Statistics
David Gerard writes "Zed Shaw writes an impassioned plea to programmers: Programmers Need To Learn Statistics Or I Will Kill Them All. Quoting: 'I go insane when I hear programmers talking about statistics like they know s*** when it's clearly obvious they do not. I've been studying it for years and years and still don't think I know anything. ... I have taken a bunch of math classes, studied statistics in grad school, learned the R language, and read tons of books on the subject. Despite all of this I'm not at all confident in my understanding of such a vast topic. What I can do is apply the techniques to common problems I encounter at work. My favorite problem to attack with the statistics wolverine is performance measurement and tuning. All of this leads to a curse since none of my colleagues have any clue about what they don't understand. I'll propose a measurement technique and they'll scoff at it. I try to show them how to properly graph a run chart and they're indignant. I question their metrics and they try to back it up with lame attempts at statistical reasoning. I really can't blame them since they were probably told in college that logic and reason are superior to evidence and observation.'"
go suck some dick. Us heteros have shit to do.
Someone mod this up
"XML is like violence. If it doesn't solve your problem, use more." - Anonymous Coward
And still has the "ignorance is bliss" and unwarranted "know it all" attitude.
Statistics is a phony science and should be thrown on the garbage heap altogether.
I will qualify this. There are a few exceptions, primarily quantum mechanics, where uncertainty is part of nature.
Except for that there is very few cases where one can apply statistics.
You probably still think I am a lunatic, but hear me out.
With human calculators it was necessary to approximate all kinds of calculations because all you could do was 0.1 IPS (instruction per second) at best. Pretty much all science became statistics, from thermodynamics, to economics to geology and meteorology. As computers became faster, more and more could be modeled accurately, and we can actually model each individual human in most population models. The same with thermodynamics. For many systems, the solution can be solved numerically, and there is no uncertainty. Quantum dynamic properties carry over to macro systems sometimes, like Einstein condensates, superconductivity etc. but most often there is no "built-in" uncertainty, and statistics is just a way to excuse incompetence, laziness or worse.
Real world example is the "medical advice" you will get before performing a procedure such as amniocentesis. The doctor will tell you there is a 0.1% chance it will have catastrophic consequences. This might mean that the hospital has one problem every 50 years, and when you dig in the data you find out that the problem happened when the hospital caught fire during the procedure. That is a manageable risk, not a probability.
Failure modes is also modeled statistically. It should not; Bridges that fail, fail predictably. It is usually just a question of collecting some data. The same with foreclosures. Some properties in my neighborhood are in foreclosure, and in all the cases I looked into, it is not hard to see why. (Like: "I told my loan officer that I could not pay the mortgage after the low teaser rate ended. He just told me to refinance again, and get a new low teaser rate, so I signed up, and a year later he told me he could not refinance." Duh!)
So programmer, throw statistics away, and stop using that sliderule.
Have fun.
don't cut it off www.mgmbill.org
Geulogies to NBSD's
...or their favorite skill in CS programming know jack s*** about Pure and Applied Mathematics, across several fields of Physics, Engineering and more. There is a reason most Engineers have taught themselves several programming languages. They have a ton of subject areas to cover that may or may not be better served in one language over another. Yet, with all the necessary Numerical Analysis, Heat Transfer, Fracture Mechanics, Creep, Stress/Strain across non-linear boundaries you still see the bulls*** coming from Biology majors who go on to Grad school, take a few more statistical classes and act as if they are one up on Engineers in Pure and Applied Mathematics. Drag their lame asses down to the labs to explain the Fluid Dynamics going on in an axial flow fan and they quickly realize they don't know jack about jack. CS major are the same way. They learn a damn programming language and walk around as if it's the answer to the Universe and not just a tool amongst thousands of tools to get work done.
Hell, most grad students in mathematics aren't running around bragging about their God knowledge of Statistics.
Just pick a field you love and live it
No, you dumb fuck.
There are 4 possibilities where one of the two is heads.
Either flip 1 is heads, or flip 2 is heads.
If flip 1 is heads, there are two options:
Flip 2 can be heads, or flip 2 can be tails
If flip 2 is heads there are two options:
Flip 1 can be heads, or flip 1 can be tails
The 4 options are therefore:
Heads-Heads (flip 1 is known to be heads)
Heads-Heads (flip 2 is known to be heads)
Heads-Tails (flip 1 is known to be heads)
Tails-Heads (flip 2 is known to be heads)
Therefore, if one of the flips is known to be heads, there is a 50% chance of the other being heads.
Still dont believe me?
Videotape yourself flipping 2 coins as many times as you like (at least 1000 times). For every time that one of the flips of the two is heads, and the other is tails, I will pay you $1
For every time that one of the flips is heads, and the other is tails, you will pay me $1.25. Any time it is tails-tails, no one pays anyone anything.
If you are right, you should profit an average of $18.75 per 1000 flips. Since I'm right, let know how much you lose, let me know youve got the unedited video, and Ill let you know where to send the certified check.
Of course I think that you are that stupid. You think that when you flip 2 coins, and one of them is heads, there is 1/3 chance that the other will be heads.