parakkum

No, this post isn't about spontaneous human combustion (which isn't spontaneous at all, btw -- there's a great Discovery Channel show about that...), but rather, a research paper all about why people flame each other in online discussion forums:

Alonzo and Aiken's article

Apparently, being assertive draws flames. How surprising.

If you were to suggest to people that they celebrate by flinging very sharp knives blindly into the air directly above themselves, they'd look at you funny.

Somehow, this never stops people from shooting skyward in celebration.

Even when they shoot down an airplane.

Geez.

From this article:

http://securitysolutions.com/mag/security_stuck_security/

"Sticky foam is a hydrocarbon solution held in pressurized containers built into steel doors. Should attackers attempt to break through a sticky foam door, they will rupture the container and expose the solution to air. Air transforms the liquid solution into a solid, and the sticky foam explodes out of the door, expanding by a 40-to-1 ratio."

They even point out that the defender can deploy it by shooting a hole in the door with a rifle. I'd watch where I stood when I did that, tho'...

I was curious, on reading the news today, how LA and Baghdad compare in terms of homicides (counting insurgency-caused deaths as homicides, here).

The short answer is, yeah, Baghdad's still way more dangerous, just based on the numbers I could find so far (this stat is harder to look up). However, some more interesting info from this page about DC's murder rate:

Highest murder rate (counted in terms of murders:population): Washington, D.C. with 45.8 per 100,000 residents in 2002
Lowest murder rate in thirty-two cities surveyed: 2.0 per 100,000 residents

The top five (high to low): Washington, D.C.; Detroit, Baltimore, Memphis, Chicago (Los Angeles is number nine)
The bottom five (from the low to high): Honolulu, El Paso, San Jose, Austin, San Diego (SF is number twenty-four...)

Well, it's nice to see the San Jose stat being so low. This makes sense in light of the massive police presence whenever they expect trouble (e.g. every Friday and Saturday night in downtown San Jose).

So, if you look at this map, it certainly looks as if Bangladesh (the gap between India and Burma on that map) should have caught some trauma. Is there some feature of the continental shelf in the Bay of Bengal that kept Bangladesh and Burma safe and relatively unharmed? Anyone with good geography knowledge out there?

Not that I'm complaining.

The recent disaster in the Indian ocean is enormous, to the point of being difficult to visualize. It is clearly a disaster, though, and this clarity comes from how very abrupt it was. By comparison, it's a given that a couple million people will die each year from malaria. There, the toll is bigger, especially considering that it happens every year. But this regularity makes it part of the scenery, just like car crashes and smoking in the United States.

Consider the slow-moving disaster that is HIV in Africa. Nigeria has a prevalence of 5.4%. That's at least 3.6 million people with HIV or HIV/AIDS. Remember that this condition is incurable. With the right drugs, you can drag it out. With aggressive antiretroviral therapy, you can drag it out for quite a while - though such extensive treatment isn't available to the poor. So that's one in twenty people in Nigeria who will die premature deaths in the next half decade or so.

But Nigeria's HIV rate is far from the worst. After all, its population is still increasing.

Over a third of the population of Botswana has HIV. That's one in three dying within the next few years, or a little longer with effective medical care. Botswana's population is contracting, with a growth rate of 0.89%, which is an exceptional thing to see in a third-world nation. Perhaps more telling, the average life expectancy at birth is just shy of 31 years.

That's a disaster, too.

All the numbers were pulled from the CIA World Factbook.

For a possibly interesting look at how papers (can) come out of a research lab, here are all the research (not review or methods) papers to come out of my old lab since its inception, by year. They're marked by first author (all the papers have at least two authors). P means the principal investigator, or prof in charge of the lab. All the letters denote graduate students. I'm E.

1997 - P
1998
1999 - A
1999 - A
2000 - B
2000 - A
2000 - C
2001 - D
2001 - A
2001 - D
2001 - A
2002 - B
2003
2004 - E
2005 - E

Another way to look at it:

1997 - P
1998
1999 - AA
2000 - ABC
2001 - AADD
2002 - B
2003
2004 - E
2005 - E

Mirapoint and Radicati recently carried out a survey that found that spam is totally worth it, sadly.

The BBC article

One in three people has clicked on an imbedded link and one in ten has bought a product advertised in unsolicited commercial email.

"This preliminary data is surprising and somewhat shocking to us," said Marcel Nienhaus, market analyst at the Radicati Group.

I'm surprised as well. I figured there must be a reasonably high hit rate or spam wouldn't be so popular, but the one in ten figure is pretty high (even if it means individuals that have done it as little as once, which I'm assuming it does). People must receive more interesting spam than I do.

I wonder how spam compares with broadcast media and junk mail advertising in terms of effectiveness. The ability to imbed a path directly to product purchasing seems as if it might lower the barrier to buying for some people.

I once read an essay that urged biologists to quantify whenever possible. I've found in working with folks in all disciplines that quantitative analysis does not come naturally. In some cases there's a reluctance to quantify -- after all, it's math and some are inclined to think that any manipulation of numbers is hard. But more than that, I think it simply doesn't occur to people to quantify.

I've been guilty of this as well, of course.

This comes up all the time, especially in understanding how things actually work versus your impression of how they work. It may astound you that someone wouldn't be able to figure out a search engine like Google, but the people at Google could show you empirical evaluations that demonstrate just how people can't figure it out. As my friend Tim once said, "Why don't you go out and measure that shit?"

But people don't, and I don't think it's so much because they don't want to as because it doesn't occur to them.

In the extended: A simple quantitative estimate replaces a WAG

A BBC article with the misleading title 'Teleporting' over the internet (and yes, I saw the quotes) is actually about developing a telepresence system that would combine motion capture on the user end with nanites that can order themselves and move to mimic the motion-captured person.

Why? Well, Dr. Todd Mowry says:

"It's very artificial to talk to somebody through a glass wall, which is effectively what you have when you have a screen," he added.

"You want to forget the fact that you're in different rooms."

Which will, of course, be easy as I wear my mocap suit and interact with the assembled nanite simulacrum representing my conversation partner. Given that you don't really touch other people in meetings, I'd think an immersive HD screen environment would work for this task.

...and for uses for the nanite simulacra, what about hazardous work environments? Deep-sea welding? Bomb disposal? Recon? (Though in that case, I'd rather have individual nanites, for better detection avoidance.)

NOAA's updated Atlantic Hurricane Outlook predicts for the remainder of the season 11-14 tropical storms of which 7-9 will become hurricanes, 3-5 these being major hurricanes.

Notable quote: These very high levels of activity are comparable to those seen during August-November 2003 and 2004.

We're watching Avatar: the Last Airbender right now. In the current episode, The Fortuneteller, Sokka is trying to convince some very superstitious villagers that their fortuneteller is wrong and the volcano is, indeed, going to destroy their town.

Sokka: "Look! Can your fortunetelling explain THAT?" (pointing at exploding volcano)

Villager: "Tsch. Can your 'science' explains why it rains?"

Sokka: "YES! Yes it can!"

I feel that way sometimes.

What percentage of commentators who refer to political "litmus tests" know what an actual litmus test is, and what it tests?

In a survey article in Nature, reported on in this Al Jazeera story and by the BBC, Wikipedia has been found to be roughly as accurate as the Encyclopedia Britannica in reporting on science content.

Experts who reviewed the articles found that the average scientific entry in Wikipedia contained four errors or omissions, while Britannica had three.

Such errors appear to be the exception rather than the rule, Nature said. Of eight "serious errors" the reviewers found - including misinterpretations of important concepts - four came from each source, the journal reported.

In bad news, Hwang Woo-suk has admitted that 9 of the 11 cloned stem cell lines reported in his group's breakthrough cloning article in Science were fake, as reported by the BBC. Here's a bit of the article that has some head-shake-inducing things for me:

A doctor who co-authored the paper says the credibility of the other two lines is also in doubt.

Many commentators said it was unpatriotic to challenge someone who had given the country a lead in such a promising new area.

Not at all my field, but both important and fascinating nonetheless:

Documenting Endangered Languages (DEL)

This multi-year funding partnership between the National Science Foundation (NSF) and the National Endowment for the Humanities (NEH) supports projects to develop and advance knowledge concerning endangered human languages. Made urgent by the imminent death of an estimated half of the 6000-7000 currently used human languages, this effort aims also to exploit advances in information technology. Funding will support fieldwork and other activities relevant to recording, documenting, and archiving endangered languages, including the preparation of lexicons, grammars, text samples, and databases. Funding will be available in the form of one- to three-year project grants as well as fellowships for up to twelve months. At least half the available funding will be awarded to projects involving fieldwork.

The Smithsonian Institution's National Museum of Natural History (NMNH) will participate in the partnership as a research host, a non-funding role.

If you're interested and it's relevant, here's the link. The proposal deadline is September 15, 2007.