I do not miss my days in the NOC. Not a bit.
Between graphic design and marketing, I think āprobabilityā is one click over from "baffle 'em with bullshit.
Alternately, itās just āIāll have that ready for you tomorrowā (probably).
Itās why I play 50 times with 50 friends every 50 days.
ps. if I win the lottery, I am buying BoingBoing.
Programming:
if (true == true) {
throw Exception("This should never happen");
}
Itās all good, itās just that with a philosophy/math degree and several courses in the philosophy of math, I donāt want anyone to feel bad about not understanding probability. You have to answer a dozen unanswerable ontological questions before you can even start talking about it.
They had a great episode of Radiolab about misunderstanding probabilities. The thing that stuck with me was the section on streaks. Basically, the chance of flipping a coin and getting heads 10 times in a row is as likely as any other single outcome. What I thought they didnāt go into enough, following that bombshell, was the way that we mentally group results. Single results arenāt interesting, itās that basically the results feel like either a streak (1 of the possible outcomes) vs. non-streak (all the rest of the outcomes.)
So itās not that screwing up is more likely, itās just that so many more outcomes are categorized as āfailureā while only a small handful qualify as āsuccessā¦ā
Or maybe you just shouldnāt have left that malted milkshake on top of that server rackā¦
Well, probability in my field comes into play in two ways.
First, my work environment is the planning fallacy incarnate. Multiple times I day I have to predict, to the day, when I will be done with different tasks, without knowing how many or how urgent additional tasks will come in in that time frame or how long many of them will take. And most people, internal and external, expect that to be within a few days or less.
Second, a big chunk of what I do is long-term technology and market forecasting. Here we are a lot more realistic. We go into meetings and explicitly say, āYou and I both know my exact numbers are wrong, the odds of being right 10 years out are too small to consider. So Iāll talk about the data we used, the assumptions we made, and the way we calculated everything. Where you disagree, we can see how our model would have turned out with your assumptions instead.ā
Thereās a (apocryphal?) story that the first bomb dropped on London in WW2 killed the elephant in the zoo. I know a stats professor who had his students calculate the odds of that happening (with some unrealistic assumptions to make this doable). Then heād tell them it did happen, and say, āThe bomb had to land somewhere.ā
Another good one was what a friend of mine used to say, āIf you want to experience a one-in-a-million event, they happen all the time. If you want a particular one-in-a-million event, youāll be waiting a looong time.ā
It was a little too thick and just needed a little heat to be perfect!! And yes, sometimes I do keep my artisnal beer near the AC condensors.
#Will @japhroaig screw up today?
- Is he awake?
- Is he typing?
0 voters
Two data centers in the same town?
Are your data-centers located in Second Life, or do you not experience Natural and UnNatural disasters (snowstorms, floods, earthquakes, Utility Co running a backhoe through the communityās main fiber line) ?
Extremely Improbable (10ā9 or less), Improbable (10ā5 or less), Probable (10ā5).
Well, I know what it says about Tyrone Slothrop, at any event.
bug repro rates:
5 out of 5 - I will fix this
1 out of 10 - I will fix this, but it will be annoying as hell
1 out of 100 - cannot reproduce
1 out of 1000 - I canāt reproduce it, but when we ship 10 million copies, this bug will ruin the experience of ten thousand players, on the first time they run the game alone.
this is why Iām deeply enamoured of any tool that detects shady code - clangās undefined behaviour sanitizer is rocking my world today
Wait, am I reading correctly that you think a 1% chance of death is high? Thatās good odds in my book, and if Iām wrong I wonāt be around to care.
Itās extremely dangerous compared to many things considered dangerous. Most extreme sports would have nothing on it.
That said, if you have zero other options, sure, itās better than it could beā¦
1% chance of death might be low or high, depending on a surgery. For the surgery I was getting, one to treat something that was not itself at all live threatening but rather inconveniencing, I think 1% is pretty extreme, and would equate to hundreds of needless deaths a year in Canada alone.
If you had to go into a room with 99 other people knowing that one of you would surely die in that room, how small of a reward would it take to get you into that room?
Do I get to pick the other 99?
This topic was automatically closed after 5 days. New replies are no longer allowed.