elliptic

And yet Ed Thorpe, from 1969-88, had 227 months where he made money, and 3 where he lost. Over that time frame, he had a mean return of well north of 15%, with a standard deviation of 4%. What precisely did you prove? Edit - your arguments are incoherent - I, or most others who claim that some people can consistently beat the market, are not relying upon some mythical, eternal method. You're correct in that any such method will become public and therefore worthless. I imagine consistently winning takes constant work and innovation.

Furthermore, no is claiming that [i]anyone[/i] can "make huge sums of money systematically without risk," or that successful hedge funds that can turn 1 dollar into 1+r dollars can always turn 1e9 dollars into (1+r)e9 dollars, or even that there exists a single financial genius that can entirely eliminate risk.

What a collection of straw men you have succeeded in knocking down!

It doesn't bother me that he feels like this. I kinda want the director of national intelligence to be a super-secretive, paranoid psycho. What's disturbing is that his elected handlers apparently feel the same way.

There's an amusing though mean-spirited article of Paul Samuelson's purporting to debunk the use of the Kelly Criterion - [url]www-stat.wharton.upenn.edu/~steele/Courses/434F2005/Context/Kelly%20Resources/Samuelson1979.pdf[/url] (pdf). All the words are one syllable. I find the title rather poetic - "Why we should not make mean log of wealth big though years to act are long".

Well, that's fair. I suppose I don't in general think much of stories that are essentially philosophical inquiries, with a little bit of "plot" added on. I'm just not emotionally affected by the "Wow" at the end, like I was in Nightfall, or how I was affected by the dead child aspect of "The Dead Past."

Also, I think that "The Dead Past" really demonstrates the apex of Asimov's technical writing ability in a way that the format of "The Last Question" doesn't really allow.

Odd that this is Asimov's favorite - it seems so inferior to Nightfall, or The Dead Past, or probably some others that I am forgetting.

Well, in that case the analogy is virtually meaningless, since "witch hunts" have always occurred.

Witch hunts were primarily a phenomenon of the early modern period, not the middle ages, during which both the political authorities and the Church repeatedly cautioned against blaming e.g poor crops on supposed witches.

Also, I believe about half of witchcraft trials ended in execution - I'm not sure what the comparable rate is for Gitmo (although obviously there are no trials).

Jesus, this sounds like my nightmare. Just lead me to my cube and let me get to work.

First I've heard anyone referring to James Simons as a physicist.

And now Continental's just as bad. Almost makes me want to move to Singapore.

I believe Newton and Galileo worked alone for much of their lives. Kepler was far more intellectually generous.

I don't agree with your point or its opposite, but the example you give seems like it is at odds with your observations on brevity and symbol usage.

Quick note - Nutt is incorrect. The Elysian Fields were a Greek domain of the dead. He probably means the city of Eleusis, which was the location of the famous mysteries, during which some (1) speculate that the participants did in fact use psychoactive substances, although I believe ergot is the most frequently suggested culprit.

1 - it's by no means universally accepted. E.g, I don't believe Walter Burkert thinks much of the idea.

Probably not for useful definitions of "generous," which should take into account the actual utility (in money or time), given away. Someone who donates half of their life savings of 50k is (I would say, at least) more generous than someone who donates 90% of 1 billion.

Its query planner may be good (I don't know about this), in the sense that it can typically choose the best of a bad lot. However, the query executor (of stock MySQL) is awful - no hash joins? No merge joins? It is good compared to what?

Let's keep in mind that Exxon and Apple are the world's most valuable publicly traded companies. There are a number of state-owned 0&G companies that are more valuable than both. I believe I read that the market cap, if it were to go public, of Saudi Aramco would be in the 5-10 trillion dollar range.

If the author was named Henry, the probability that this post would have been written would be far lower. For better or for worse.

Oh, that's good to know. What if a security guard had pulled a weapon on him? Would anyone have been injured then? Or what if one of the other customers had rushed him?

I used to think that a man (it's almost always a man) brings a loaded gun to a crime for one reason - to murder someone who got in his way. I've since mollified my stance on that, but still, the fact that no-one was injured isn't in itself really exculpatory.

To anyone who is really interested in the subject, I recommend a book called "Night Falls Fast." It's available as an e-book.

< here is nothing inherent to any of the languages I mentioned that makes them slow.

I don't agree. The inability to explicitly allocate on the stack in Java and OCaml? The lack of threads in OCaml? Java GC pauses (by the time you're storing things off-heap you might as well be writing C++)? I don't know as much about Lisp, but when looking at a task X, I tend to give the benefit of the doubt to languages (or frameworks, or hardware, or people, for that matter) that have actually achieved a measure of success at X.

An analytic database like Vertica? A computer game with cutting-edge graphics? Etc.

Completely false. Like much of the rest of Bill of Rights, the second amendment has been incorporated (see Chicago vs McDonald), and thus held to apply to the states.

It slightly irritates me when people say this, because usually those people have no problem accepting that other amendments that they hold to be more important (e.g, the 1st) apply to the states.

I prefer to solve problems like Hercule Poirot

I'd recommend Xenophon's Anabasis.

Accutane is pretty miraculous, if you want to risk the side effects and can get it somewhere it's available. It worked for me (~ 10 years ago).

Is this situation similar to that of Louis de Branges & the Riemann Hypothesis a few years back? I.e, a well-respected mathematician (de Branges had settled the Bieberbach conjecture in the 80s) releases a proof of an important unsolved problem using his own poorly understood mathematical technology?

Edit - lest this sound too negative, one should realize that the Bieberbach proof took a long time to be accepted.

Surely the burden is squarely on the author to show what the generalization (in this context) offers that the specific concept of sufficient statistics does not? In other words, how else can a statistician or ML expert use this foreign "monoid" concept to improve or better understand parameter estimation?

Have you a link for that?

NaN is incredibly useful, but I'm not so sure that 1/"a" = NaN is a great idea.

I believe the title should be: From the Times, Advice on Writing Well