They must have let Goobs measure...again

[Left Seater]

https://www.google.com/amp/s/amp.busine ... ng-2017-10

How the hell do you fuck this up? Twice?

For the second year in a row The Milwaukee Marathon course has the wrong distance. Last year it was too long, this year too short. :doh:

[Mikey]

On average they’re pretty close.

[Left Seater]

Can you imagine training for this then finding out after you didn’t run a marathon?

Or worse, thinking you qualified for Boston or NYC and then be told your run didn’t count.

[Mikey]

I heard about a pilot who had to abort a landing because he came in too high and would have overshot the runway. So he came around for another try but over corrected and piled his plane into the ground. He was pretty close on average though.

[Left Seater]

I guess if he isn’t dead.

There are these tech gadgets that can help measure or land a plane. Problem is they still require humans to operate them.

[Goober McTuber]

I'm supposed to care?

[Mikey]

I guess if he isn’t dead.

There are these tech gadgets that can help measure or land a plane. Problem is they still require humans to operate them.

Actually I just made that up to illustrate a point.

Lately I've had to put together a lot of sampling plans, which are used to do statistical sampling of the power (kW) draw of some lighting products to determine an "average" power draw with a 90% certainty that the result is within 10% of the actual average value of the population.

Sounds like a lot of BS, right?

The size of the sample that's required to meet this criteria depends on how much the individual values vary around the mean, which can be expressed as the Coefficient of Variation, or the Standard Deviation. With a large variation you might have to measure 50 or 60 samples, with a small variation maybe 4 or 5, even to get the same "average" value.

I hope that's all perfectly clear.