Belligerati

« December 2006 | Main | February 2007 »

January 31, 2007

Strange joys

I guess you have found the right field when you go to invert a nasty matrix think it is another horrible partitioned matrix inversion, but they turn out to just be a 2x2 matrix of scalars, and you get a rush of joy.

Posted by OneEyedMan at 2:30 AM | Comments (0)

January 30, 2007

I find your lack of faith disturbing

A New Sith, or Revenge of the Hope
Reconsidering Star Wars IV in the light of I-III

Very enjoyable and interesting.

Posted by OneEyedMan at 3:09 PM | Comments (0)

January 28, 2007

Stop motion video of video games using candles

Posted by OneEyedMan at 2:36 PM | Comments (1)

January 25, 2007

So that's how they do that.

If you were wondering how you could check that your linear regression application works exactly as it should, it turns out the national institute of standards and technology can help you with that. They have data sets and the correct coefficient vectors to check if your program works as it should.

PS
Today I worked from 6AM to 10:30.

Posted by OneEyedMan at 2:22 AM | Comments (0)

January 24, 2007

What words keep popping up?

A list of all the state of the union addresses but instead of showing their full text it shows their 100 most common words excluding things like "to" and "your". It is cool how the themes change over time.

Posted by OneEyedMan at 2:08 AM | Comments (0)

January 23, 2007

Have singularities happened before?

A great interview with Joe Quirk, evolutionary psychologist and singularity theorist is over at 10 zen monkeys. The technological singularity is a theoretical point (or more accurately a brief interval) where changes in our lives from technologically become so vast that we cannot imagine what life will be like afterwards, much as you cannot observe what happens on the other side of a black hole (another singularity).

It raises all sorts of issues I've never considered before, including whether singularities have happened before.

RU: So what do you really think? Are you fundamentally a believer in “The Singularity” or are you a skeptic?

JQ: I'm a scared skeptic and a hopeful skeptic. Most people who hear about it think it's whacko, so I find myself defending it more often than criticizing it. And I think Kurzweil's actual arguments in his two most important books are more compelling than the counter-argument from Incredulity, which is just a knee-jerk reaction -- "C'mon, this is Rapture for the geeks." Every group makes up some kind of mythos, and this is a mythos for the geeks. I keep thinking of other examples of Singularities. I've never heard anyone talk about the Singularity that's already happened. Let's see if you guys can point it out.

RU: Language?

JQ: That's one, but I've never heard anyone talk about the Singularity of techneme -- the singularity of tools. Imagine a Homo habilis playing with his stone axe, and his buddy says to him, "Grok! These stone axes are not going to change for millions of years, because we're on the flat part of an exponential curve. But this has an abstract design within it, which means it contains information that can be passed down through the generations. And in another 3 million years, we're going to have a feedback loop of information, and pretty soon our tools are going to cover the world; they're going to be on our bodies; and we're going to go from a few thousand of us to a few billion of us. Everything we touch will be a tool. Our tool designs are going to inhabit matter and build our dreams around us. Everything we look at is going to be a manifestation, an embodiment of an idea."

Posted by OneEyedMan at 2:54 PM | Comments (0)

What do we have in common?

Donald Brown, professor of anthropology at UCSB wrote a book called Human Universals in which he discusses the 200 (or there abouts) attributes that all human societies share.
A few surprises from the list:
Poetic lines characterized by repetition and variation
Polysemy (one word has several meanings)
Disapproval of stinginess

Posted by OneEyedMan at 12:22 PM | Comments (0)

January 22, 2007

How to behave in a resturant

Augieland has a few bits on restaurant conduct.

Much of it you may already know, but one especially caught my eye:

When you have purchased a good bottle of wine, offer to share a taste to your server (rather then cooing about it to him).
The sommelier has probably already had it, but the server may not have. Either way, you are sharing an experience. If he has tasted it, your gesture will probably be declined, although appreciated. If he hasn’t, you will be adding a quill to his quiver of talents and he will appreciate it. For what will cost you about one of the ounces in your bottle, you'll open a dialogue that will allow you to benefit from his experiences. Discussion may then ensue on your common likes and dislikes, which will greatly help you value his recommendations which will be good to know once you are a regular.

That said, his bit on tipping (third article linked to above) just strikes me as deeply flawed.

I agree with what he says about tipping on things that you get comped. And of course, tip huge if you want to. Especially if you like to return to restaurants and be treated especially well. But don't pretend as though enormous tipping is in any way morally required. While it may be more difficult to serve in a more expensive restaurant, it certainly doesn't scale the way the price does. And if we are debating the restaurant business, tell me why the waiters of fancy restaurants deserve huge salaries while the kitchen staff, who have a much harder job and have at least as much impact on your overall experience, deserve a salary that scales very little with restaurant price compared with the waitstaff. It doesn't have a moral explanation, it is just the anthropology of the restaurant business.

My experience in discussing whether to tip on the pre- or post-tax bill with current and former servers is there certainly is dispute. Many consider an 18% tip on a pretax amount an good tip. But big tipper beware, as tipping is an arms race for a limited supply of special attention.
Tipping simply has major problems. First, why should your staff treat customers differently depending on how well your customers bribe the staff? Second, Why should we provide such a convenient method of tax evasion to waiters? Third, if we agree that wait service is something worth paying for, why allow people to shirk and avoid paying it at all? I think that we can all agree that tipping is a stupid, sub-optimal way of paying for restaurant staff. We'd all be better off with staff wages built into the menu price. Chez Panisse, a fancy restaurant in Northern California shifted to this model because they thought a 15 percent service charge was more fair because it both created more money to pay support staff and provided more even service to customers.

We got to this situation from rich Americans in the late 19th century throwing their money around. And certainly that was their right, as I said before, if you like it, and the restaurant allows it, feel free to tip (if they don't, it simply is a bribe). But why does it remain? It may have to do with tax treatment differences. Mandatory service charges are subject to sales tax and tips are not, but both are subect to income tax. Net margins in the restaurant business are about 3.5%. Median US sales tax rates are 6.4%. If the average restaurant tip was 18.7% for 2006. That means that a restaurant shifting to a service charge of 18.7% from an average tip of 18.7% is facing giving an additional 1.2% of the gross to the tax man. That's a third of net margins and for many restaurants, the difference between profitability and failure.

Posted by OneEyedMan at 12:26 PM | Comments (1)

January 21, 2007

Choosing the correct null hypothesis

"[T]he idea that abortion is at the root of women’s psychological ills is not supported by the bulk of the research. Instead, the scientific evidence strongly shows that abortion does not increase the risk of depression, drug abuse or any other psychological problem any more than having an unwanted pregnancy or giving birth."
NY Times

That doesn't sound like the right question to test against. We should not treat the test for psychological wellbeing after getting an abortion against actually having a baby, even one you don't want. Oh, I understand the appeal of an argument like that, after all if you don't get an abortion, you are probably going to end up a mother, but that's not the way real women deal with abortions, at least in my experience. In any case, the answer to that question isn't enough.

Women mostly get abortions in order to avoid the life changing maelstrom that is motherhood. These women hope that everything is going to be as before they got pregnant, not that it will merely be better than being pregnant. It would seem to be more appropriate to compare the mental well being of women who get abortions with those that are otherwise identical but don't get pregnant. Telling women that are thinking about having an abortion that they won't be any more messed up than real mothers is something, I suppose. But suppose women who get abortions were substantially more unhappy than women who never get pregnant in the first place. Many women would make use of that. Perhaps some would say "Oh, well since I cannot go back to the way things are, and to some extent I'm going to be more unhappy now anyway, maybe I should see if I can find the energy and resources to figure out if I can have this baby." That would make abortions less attractive on the margin. On the other hand, if you could get an abortion and you psychological profile became identical to women who had never been pregnant, then getting an abortion does put Humpty back together again, and you'd expect more frequent abortions.

Posted by OneEyedMan at 12:49 PM | Comments (0)

January 20, 2007

How nasty can you make a Jello shot?

Over at My Science Project, they've attempted to make the most powerful possible Jello shots that people can stand. It turns out that when you add pure Everclear to gelatin it devolves and you get a disgusting mess like this:

Posted by OneEyedMan at 9:22 PM | Comments (0)

January 19, 2007

New York Restaurants at Night

New York Restaurants at Night is a map of almost 24,000 NYC restaurants. Click on a neighborhood, and then use Google maps to pick one. You can also filter by food type.

Posted by OneEyedMan at 8:23 PM | Comments (0)

January 18, 2007

A friend asks

J Writes

I thought of you when I read this article on the cost of the Iraq war.

As a self-professed libertarian, can you support the decision to invade Iraq when you look at these cost numbers? I know that you supported it at the time, but do you still do so? Would you have supported it at the time if you knew how much it would cost?

My Response:
Much like dropping the atomic bomb on Japan, knowing the right facts with certainty before the fact in Iraq (that they did not have WMD) would have driven me to advocate alternative behavior. However, I still think that given the information at the time I think it was the correct decision.

Does that require explanation? As a libertarian, I actively decry the head-in-the-sand wing of the libertarian movement who pretends that if we dismantle our most of our military and close all our embassies that all our enemies will forget they hate us. That sort of unilateral disarmament makes no sense. The costs of doing nothing are not free, you just don't pay now.The NY times said the other day that the real estate alone in NYC was worth $802.4 billion. If Saddam could get a nuke and use it against us, $1.2 trillion could seem cheap. Every expenditure comes with a touch of sadness of the forgone alternatives, although my understanding is that these cost estimates are too high.

The tremendous waste that is the carnage in Iraq of great sadness to me. I feel we should hold a referendum post-haste (perhaps after a post-surge improvement in order) where we ask the Iraqis if they want us to stay.If not, I say we pack up and leave. If so, we owe them and ourselves a thorough attempt at establishing peace and order.

What do you think of that?

Posted by OneEyedMan at 2:04 AM | Comments (0)

January 16, 2007

Last night's speaker

Last night I saw Michael Dukakis (1988 presidential candidate) speak at The Institute of the Americas. While the topic veered wildly, it did manage to cover the basic topic of the evening, which was international relations and development of Central and South America. Governor Dukakis was a speaker of great entertainment. He held our attention with great stories and clever turns of phrase. But his basic policy prescriptions were totally uncreative: more aid, more multilateralism, more summits, more international governing bodies, higher minimum wages, and higher labor and environmental standards for our trading partners as solution to the hemisphere's problems. His solution to the immigration problem was something new: greater minimum wages and stricter enforcement of labor laws in general to destroy the demand side of the illegal labor. As a pair that certainly won't work, but he's probably right money spent on checking citizenship information at workplaces is money far better spent than on fences.

All and all, he was good for an evening's entertainment but I'm glad that Bush Senior beat the man.

Posted by OneEyedMan at 11:39 AM | Comments (0)

January 15, 2007

W Screws the country

Never lose your sense of humor:

Thanks to Adrants for the tip.

Posted by OneEyedMan at 5:30 PM | Comments (0)

January 14, 2007

The Mainstream media discovers circumcision does something about AIDS

We spoke about how circumcision decreases AIDS transmission rates a month ago and again six months ago.

The NY Times magazine claims that it might be effective enough to for us to consider it our vaccine. The recommend widespread free circumcisions as an aids control technique.

For years, AIDS researchers have observed that many African tribes that circumcise boys or young men had lower AIDS rates than those that don’t, and that Africa’s Muslim nations, where circumcision is near universal, had far fewer AIDS cases than predominantly Christian ones. The first research proof came in 2005, when a study in South Africa was stopped early in the face of evidence that the men who had been randomly assigned to be circumcised were getting 60 percent fewer H.I.V. infections than the men assigned to the control group. Last month, ethics boards halted two similar studies, in Uganda and Kenya, when they found similar results. In both, the circumcised men caught the AIDS virus half as often as the uncircumcised control group.

Circumcision would be given more weight if the world recognized that it is, in fact, the real-world equivalent of an AIDS vaccine....

Many vaccines provide nearly 100 percent protection — after my daughters finish their two doses of the measles, mumps and rubella vaccine, for example, they won’t have to think about those diseases again. But that’s not on the horizon for AIDS. “Fifty to 60 percent efficacy is what people would feel really good about,” says Frances Priddy, the director of efficacy trials with the AIDS vaccine initiative. The best candidates in the vaccine pipeline right now — which won’t be ready until 2013 at the earliest — wouldn’t keep you from getting H.I.V. They instead would seek to change your body’s response to the virus so that if you did get infected, the disease would progress more slowly — or not at all — and you would be less infectious to others.

An efficacy rate of 50 to 60 percent is actually a lot better than it sounds, because of herd immunity. We get AIDS from one another. Every time a person is rendered less infectious, the chance of an uninfected person catching H.I.V. from each sexual contact drops, and in a virtuous circle, the whole community becomes progressively safer. A vaccine of 50 to 60 percent efficacy might come close to wiping out the epidemic in places with low AIDS rates. In high-prevalence areas, it could reduce the epidemic and save millions of lives.

Posted by OneEyedMan at 1:15 PM | Comments (1)

January 12, 2007

Too clean for your own good

American Scientist has a detailed explanation of why we develop food allergies and suggests why it has gotten worse in the last few decades. It turns out we really are too clean, filth really does help us be healthier, at least at low levels. Reminds me of something my dad says, your kids will eat a pound of dirt a year, there isn't anything you can do to stop it.

Posted by OneEyedMan at 2:08 PM | Comments (0)

Jobs for Iraqis

Back in July I discussed a new plan to employ Afghans to discourage them from picking up arms. In today's WSJ, Giuliani and Gingrich are calling for a similar plan for Iraq. They are proposing using an annual wage of the "pre-war median annual income equivalent to $700". That strikes me as dangerous, because given the current economic calamity that's a good wage, maybe too good. It could attract the very people that are already employed, harming the floundering economic efforts of a post-Saddam economy. Better to make it less than the median wage, but not so much less that it attracts no takers. I hope if they do a policy like this, it includes relocation. They should target families with unemployed young men living in Baghdad, and then move them out to rural factories. That does several important things. It keeps people out of the resistance, it moves them out of harms way (between 50 and 73 percent of deaths are in Baghdad), it gives their life some purpose, it acts as a relatively incorrupting method of distributing aid, and maybe even helps the economy a bit.

Posted by OneEyedMan at 1:35 PM | Comments (0)

January 11, 2007

What's he building in there? Let's go look it up online

The BlueEyedGirl and I watched The Smartest Guys in the Room, a rather misleading and unhelpful look at the fall of Enron that everyone seemed to love but me. I read an awful lot about what happened in the Enron story, and I haven't found anything as insightful and lucid at Malcolm Gladwell's piece in The New Yorker, Open Secrets.

Enron's S.P.E.s were, by any measure, evidence of extraordinary recklessness and incompetence. But you can't blame Enron for covering up the existence of its side deals. It didn't; it disclosed them. The argument against the company, then, is more accurately that it didn't tell its investors enough about its S.P.E.s. But what is enough? Enron had some three thousand S.P.E.s, and the paperwork for each one probably ran in excess of a thousand pages. It scarcely would have helped investors if Enron had made all three million pages public. What about an edited version of each deal? Steven Schwarcz, a professor at Duke Law School, recently examined a random sample of twenty S.P.E. disclosure statements from various corporations--that is, summaries of the deals put together for interested parties--and found that on average they ran to forty single-spaced pages. So a summary of Enron's S.P.E.s would have come to a hundred and twenty thousand single-spaced pages. What about a summary of all those summaries? That's what the bankruptcy examiner in the Enron case put together, and it took up a thousand pages. Well, then, what about a summary of the summary of the summaries? That’s what the Powers Committee put together. The committee looked only at the "substance of the most significant transactions," and its accounting still ran to two hundred numbingly complicated pages and, as Schwarcz points out, that was "with the benefit of hindsight and with the assistance of some of the finest legal talent in the nation."

Posted by OneEyedMan at 9:43 PM | Comments (0)

Take a look at this job posting

Where was this job when I was in high school?

Employer: Position Title: Dungeon & Dragon Master Posting Date: January 5, 2007 Expiration Date: February 15, 2007 Division: N/A Organization Description: Family Description : We are looking for a Dungeon & Dragon Master to host a D&D game for 6-8 kids (ages 10-14) once or twice a week. Their old DM left for college in LA. Qualification: Must have prior DM experience and a fun and outgoing personality. Job Function: Education/Teaching Position Type: Part Time, Paid Duration: Indefinite Approximate Hours Per Week: 5 Desired Start Date: January 12, 2007 Salary Level: $10.00

Posted by OneEyedMan at 8:16 PM | Comments (0)

The ethics of blogging

Joel on Software has a great blog piece "Bribing Bloggers" which I thought laid out nicely the ethical issues surrounding publicity inducing gifts / bribes that bloggers seem to be getting more and more of.

His general position is to accept nothing, but does offer up the NY Times policy as an acceptable alternative.

Staff members may keep for their own collections -- but may not sell or copy -- books, recordings, tapes, compact discs and computer programs sent to them for review. Such submissions are considered press releases.

I like the NY Times policy better, but since I've been offered a book, and nothing more valuable, I guess that's not a serious moral consideration for me. It is also noteworthy that many product technology and car magazine (excepting of course consumer reports and cooks illustrated), don't buy their own stuff, they just send it back when they are done. There is something to be said that if your business is reviewing things and everyone gives you stuff to try, that while you might not be impartial, that's not the same as auctioning off reviews to the highest bidder. Further, argument can be made that the ranking intensive, brief reviews for non-experts provided by the fully objective magazines I mentioned above are a very different product than those the obsessed aficionado targets, lengthy and technical reviews provided by the magazines that do take lender products.

If you are going to take lender products, the measure of your objectivity depends a lot on what percentage of your revenue is represented by the gifts/bribes, the level of separation of the ad and content departments (for example, since I use Ad sense they are totally detached), and perhaps whether the primary reason that people read what you write is for the entertainment or for the information. That said, I think we can all agree that a straight up bribe of $3000 worth of stuff to write an article about something is probably not ethical.

Posted by OneEyedMan at 2:56 PM | Comments (0)

January 10, 2007

Where have I been on vacation?

create your own visited countries map

Posted by OneEyedMan at 4:35 PM | Comments (2)

Words for buttocks

Callipygian: Having shapely buttocks
Dasypygal: Having hairy buttocks

Posted by OneEyedMan at 2:38 AM | Comments (0)

January 9, 2007

Economics and non-economists

Consider the extent to which we live in a world where economics shapes our world view. Four of the ten best selling books on Amazon are built around economic thinking. Humor website over heard in NY can make jokes like this (note the title):

His Archenemy Is a Severely Progressive Tax Structure
Older brother teaching the finer points of comic books: Yeah, Batman's really cool. Best thing about him -- he doesn't have superpowers, so he's really an ordinary guy.
Younger brother: Wow, no superpowers?
Older brother: Well, apart from being super rich.
--F train to Queens

So perhaps it isn't a surprise that the Stern Report on Global Warming prepared for the British government, a long, arcane, and technical estimate of the costs of global climate change has attracted much popular interest. But who would have thought that discount rates and the elasticity of marginal utility of consumption would become topical discussion points for non-economists?

Jane Galt has a fascinating criticism of some of the parameter estimates that underlie the Stern report, even as she largely supports the document's policy prescriptions anyway. I wonder why -- ascetics?

Meanwhile, The economist blog, in their piece, Right-to-discount, discusses how the assumption of a zero-discount-rate by the Stern report creates serious problems in justifying an open abortion and birth control policy if we insist that discount rates be used consistently.

Posted by OneEyedMan at 2:16 PM | Comments (0)

January 8, 2007

Classes

Today was the first day of the new semester. Monday's and Wednesdays are my longest days, with classes between 8:00 and 3:20.

Posted by OneEyedMan at 9:33 PM | Comments (0)

A neat new tool for bloggers

I make use of Furl as a tool to manage the many articles referenced in this blog so that I revisit them later if the hosting website takes them down. I've often found it frustrating that some web pages, either thorough an instable linking system or just in using very long articles, make it difficult to show readers where you've quoted from. That's why I was excited to learn about Cite Bite. SCite bite is similar to Tiny URL, in that it gives you a new URL that posts to the same content as before, but not only does this shorten the URL, it also allows you to point it to a specific quote in the linked piece, even changing its color.

Posted by OneEyedMan at 6:50 PM | Comments (0)

January 5, 2007

Do we need to start hiring foreign troups?

In Does America need a Foreign Legion?, Colby Cosh discusses the reasons why the US might want an army of foreigners to help protect it.

The French Foreign Legion is a force of foreign volunteers which in exchange for a combination of money, an opportunity to fight France's enemies, and a shot at French citizenship has attracted fighting men from around the world. I heard a few years ago that joining got very difficult in the first few years after the fall of the USSR because elite Soviet troups tried to join by the, well, legion.

Max Boot, over at the Council on Foreign Relations has another bit on the same idea of letting some fight for their US citizenship. He mentions two strengths. "...it would make it easier for the U.S. armed forces to fill their ranks with high-quality volunteers. Second, it would increase the armed forces' knowledge of foreign languages and customs."

But a serious question is if we are willing to make the military into Another job Americans just won't do. Is the solution to just raise pay levels? Pay is a fairly small componant of a budget of $300,000 per soldier.

Posted by OneEyedMan at 7:17 PM | Comments (0)

Humans and Computers as Compliments Complements

Luis von Ahn is a professor of Computer Science at Carnegie Mellon University. He has an amazing talk, which is long but fascinating, about how to create computer games to get humans to assist computers in doing tasks that computers do poorly at with specific examples in image processing and common sense facts about the world.

Long but fascinating.

Posted by OneEyedMan at 10:35 AM | Comments (0)

January 4, 2007

What sorts of movies make money?

I once spent a semester doing econometric analysis of what made some movies profitable and others not as a part time job. I liked it a lot, and if they'd been hiring when I graduated my life might be quiet different. One of the things at the time that struck me was that R rated movies seem to do worse than the more youth friendly pictures. Others seem to have found this fact with great robustness. A recent study in the book Holywood Economics found that not only do G rated movies they make more money on average, but that altering the product mix would result in higher studio profits.

A few questions come to mind:

1) Are good movie scripts around (unproduced) in all ratings? Maybe all the potentially good (that is not certain to be bad) G and R rated movies are made, so there just there aren't more G movies to make profitably?

2) Since so many G rated movies are animated, is there something about the popularity or means of production of animated movies that limits how well it can be scaled?

Another finding is that cartoons seem to get gentler ratings than live action movies with the same behaviors. The ratings were far less stable than I would have guessed.

Here is a link to the top grossing movies by MPAA Rating.

Posted by OneEyedMan at 2:01 PM | Comments (0)

January 3, 2007

$100 PC

I had thought that the new $100 computers that were developed as first computers for the children of the developing world were going to just use a light-footprint linux setup. I was mistaken. It seems that actually they will run an entirely new GUI designed by Novell for their needs.

Posted by OneEyedMan at 9:54 AM | Comments (0)

Youtube Nuget

An amazing piece of amateur stop motion video with live actors. I laughed aloud several times. How do they hover like that -- green boxes used as platforms?

Posted by OneEyedMan at 8:15 AM | Comments (1)

It isn't a one way march to tyrany

I just learned that France has ended their 30 year ban on supermarket advertising on TV. I predict a combination of lower prices, better service, and wider selection to result.

Later that same day I learn Sau Paulo just the opposite, but worse!
Sao Paulo is banning all outdoor advertising.

Posted by OneEyedMan at 8:05 AM | Comments (0)