I came here to post a similar sentiment. I think it is a terrible idea to just blow ahead every time an assertion is too confusing. Getting the big picture and developing mathematical intuition is great but it doesn't mean that you'll actually be able to do math. For that, practice.

I actually advise the opposite, to bang your head against a problem over and over until it breaks (the problem, that is). I don't think that we as a society or a species or whatever deal with confusion very well, and tend to take it as some sort of personal deficiency. We've also done a great deal of dumbing math down, so that when someone tries to make the jump from, say, AP calculus to real analysis, minds get blown and souls shattered. It's probably not that mathematicians enjoy crushing students, but rather that higher levels of math are just plain confusing for most people. They're based on abstractions that are pretty far removed from the human experience. None of this is to say that people who are good at math are better somehow, but it usually means that they put in a lot of time. I suspect that a lot of people who are math-phobic would get over it if you locked them in a room with nothing but math books to keep them busy.

One that is clear, however, is that most mathematicians have no fscking clue what the word "obvious" means. There are some brilliant, dead authors that I would love to punch in the face.

A few folk here have commented using incomplete or inaccurate definitions of p-values. A p-value is the probability of finding new data as or more extreme as data you observed assuming a null hypothesis is true. A couple of salient criticisms not mentioned in the article are a) why should more extreme data be lumped in with what was observed and b) what if "new" data can't sensibly be obtained.

In a less technical sense, what the article didn't get into so much is that there is a strong publication bias towards results that are significant (i.e. small p-values), to the point where you need <0.05 to even consider submitting. Some key reading: http://www.stanford.edu/~neilm/qjps.pdf. The short version is to not believe it when the news says that "recent research shows...".

Personally, I wait for evidence to accumulate before, say, changing my diet. And if you really want to get it right, dig through the literature yourself. Some of my saddest moments have come from statistics consulting where mostly people come to you looking for permission to run an inappropriate analysis, not understand their data or fit the "right" model. They want to get published, and that's just how things are done.

Hi, statistician here. That's not what "regression to the mean" is.

Regression to the mean applies when you have a pair of comparable measurements, the first one of which is high (or low) when compared to similar quantities. So a father and sons' heights, for example. High measurements tend to be a combination of nature + luck. Looking at the other of the pair, the luck component isn't likely to be as good, so on average you're left with just nature. Hence, really big regresses back to big.

You are, however, likely correct that the hype is what is at play. Or that there was pent up demand. Or that people love sales. Or whatever. Kind of hard to imagine a good experiment on introducing Steam to Linux over and over again that we could analyze.

HELP! HELP! I'm being repressed!

Sorry Holmes, but "social justice" does have some meaning. Maybe as it is bandied about now it has no bite, but the term has roots in Catholic, specifically Jesuit tradition. Not that I remember my indoctrination days all that well, but quick perusal of Wikipedia associates it with "life and dignity of the human person", and "preferential option for the poor and vulnerable".

Ethicists long ago figured out how to make decisions that respect such high-minded and abstract principles, so long as you admit that there are tradeoffs involved. You can even imagine a society that operates with those as some of its core values, although that might be bordering on crazy-talk.

From Think Progress, April this year:

Fifty-six percent of domestic terrorist attacks and plots in the U.S. since 1995 have been perpetrated by right-wing extremists, as compared to 30 percent by ecoterrorists and 12 percent by Islamic extremists.

Maybe in air travel incidents you're correct (dunno, haven't looked into it), but the basic insinuation that all terrorism originates from Islamic extremists is misleading.

I was just finishing up my undergrad when Facebook came to my school - one of the first handful of institutions if I recall. Were we being "delusional" when we posted those all those photos, or was what we hoped Facebook would become entirely reasonable? Is it really impossible to believe that someone could make a website on which you could share things with your direct friends and possibly, but no more than, their friends? Aside from the creepy image of Mark in the old banners for the site, there weren't many indications that it would end up like it did.

I suppose you can live by the maxim that everything online will necessarily turn to crap and everyone is just waiting for the right moment to sell you out, but I'm not ready to swallow that pill yet. And you'll have to forgive me if I keep on hoping for better.

That being said, I'll admit that I approached posting to Facebook in its early stages as a risky decision and did my best to consider the potential outcomes and their likelihoods. However, not everyone who was crawling all over the thing at the time spent the 90s plugged in, and belittling them for not making an informed decision without any prior information seems a little extreme.

Studying these folk would be incredibly difficult, as we would end up with a lot of data on the kind of people who are wealthy enough, sick enough, and desperate enough to undertake the procedure, with no real basis of comparison. Scientifically, they're not of much use.

Brownian Motion is a mathematical construct, which, among other things, is nowhere differentiable (almost surely). You can pin a BM down into sets with high probability, but no, you can't really predict it. It is merely used to *model* the movement of a particle in a fluid, it is not actually the process by which the molecules move. Indeed, "such a path represents the motion of a particle that in its wanderings back and forth travels an infinite distance in finite time. [BM] does not in its fine structure represent physical reality." (Billingsley, "Probability and Measure"). At least the science is interesting.