Clarification. Chebyshev's inequality is not going to help you with distributions that have no mean or standard deviation. Note also that the standard deviation mentioned in Chebyshev's inequality is the *population* standard deviation, and NOT the *sample* standard deviation.

Careful. Chebyshev's inequality doesn't help you if you are sampling from a physical process with a Cauchy distribution. Be careful not to confuse the *sample* standard deviation with the *population* standard deviation. The former always exists. The latter is what you use with Chebyshev's inequality... *if* it exists. In the case of a Cauchy distribution, your sample standard deviation would mislead you into thinking that the probability to fall outside N sample standard deviations had some particular bound that it did not have.

You might be interested in this: http://en.wikipedia.org/wiki/Chebyshev's_inequality The standard deviation tells you something important about *any* distribution.

I hope he has more examples than just the temperature (no, I didn't RTFA). For the temperature in a day, most people are satisfied with the minimum and maximum, and don't need any more complicated measure. The MAD would actually be LESS informative for temperatures within a day...

He's rather requesting the people start using a different statistical measure of spread, the mean *absolute* deviation, rather than the square root of the mean *squared* deviation (the standard deviation). I'm not familiar enough with it's particular characteristics to say whether or not this would be an improvement in any rigorous sense, but I'd be surprised if it were. So "Get bent." is probably still the right attitude.

I know several people who have left high energy physics to become data scientists. Nobody in HEP calls themselves a "data scientist", but that's (some of) what we do anyway. It's just analysis of very large data sets. Unlike in the life sciences, both HEP and many commercial / industrial environments have sufficiently large data sets that very complex questions can be asked and answered. You can never have "enough data" -- if you think you have "enough data", then you aren't asking hard enough questions.

Means for a government to systemtically supress or harass certain people or groups of people are worrisome to me.

Wasn't 3 Mbps "high-speed" ten years ago?

The interplanetary medium can carry sound waves. Of course, it is moving faster than the local speed of sound outward from the sun (the solar wind). So if you shouted really loud from the ISS, someone in the asteroid belt might be able to hear you. But not the other way around.

It was pretty much one of these: http://en.wikipedia.org/wiki/Explosively_pumped_flux_compression_generator

eq?, eqv?, or equal? ?

Possible, although unlikely. The -CDM model does an astonishingly good job of modeling the observed universe. But, that doesn't mean it is right.

In the case of aether, people didn't stop investigating it until a) experiments that should have observed no matter what saw no evidence of it and b) another theory that agreed with this new data came along.

People who trot out the tired old "dark matter is just like aether!" line typically do so while patting themselves on the back for their cleverness, while neglecting the above.

If there isn't WIMP dark matter, or even isn't dark matter at all, then we'll find out. That's how science works.

Weak.

You sound like Ebenezer Scrooge: http://youtu.be/QIcoa0oI5Cc?t=10s

It's not a favorable comparison.

This is exactly the right answer. Never write code that someone else has already written. If you can compose standard operations to do your calculations, then do so in a high-level language. Spend more time thinking and less time coding. OTOH, if you need to code up something custom and you're REALLY sure that you can't use standard operations to do it, then think again about whether or not you can do it with standard operations. You probably can. But, if you can't, then go with FORTRAN. Or maybe C or even C++. But probably FORTRAN. But even then, code as little as you can in FORTRAN. Don't write the whole thing in FORTRAN. Create small operations, and compose them in a high-level language as if they were the standard operations.