Quotulatiousness

August 10, 2012

For you, is no Singularity

Filed under: Science, Technology — Tags: , , , , — Nicholas @ 11:25

Charles Stross linked to this article which points out that we’re not likely to experience the Singularity/Rapture of the Nerds/etc., and for good reasons:

Given that you are tech-savvy, by that point you have almost certainly come across the idea of the Singularity [1] as defended by futurists like Ray Kurzweil and Vernor Vinge. As a reminder, it is the notion that, when we are at last able to compile a smarter-than-human artificial intelligence, this AI will in turn manage to improve its own design, and so on, resulting in an out-of control loop of “intelligence explosion” [2] with unpredictable technological consequences. (singularists go on to predict that after this happens we will merge with machines, live forever, upload our minds into computers, etc).

What’s more, this seemingly far-future revolution would happen within just a few decades (2040 is often mentioned), due to the “exponential” rate of progress of science. That this deadline would arrive just in time to save the proponents of the Singularity from old age is just a weird coincidence that ought to be ignored.

Objection, your honor. As a scientist, I find the claim that scientific progress is exponential to be extremely dubious. If I look at my own field, or at any field that I am vaguely familiar with, I observe roughly linear progress — a rate that has typically been going on since as far back as the field’s foundation. “Exponential progress” claims are usually supported by the most bogus metrics, such as the number of US patents filled per year [3] (essentially a fashion utterly decorrelated from scientific progress).

And as somebody who does AI research, I find the notion of “intelligence explosion” to make exactly zero sense, for reasons reaching back to the very definition of intelligence. But I am not going to argue about that right now, as isn’t even necessary to invalidate the notion of the Singularity.

August 2, 2012

Charles Stross: Where Moore’s Law and Koomey’s Law interact

Filed under: Science, Technology — Tags: , , — Nicholas @ 09:32

On his blog, Charles Stross explores the long-term implications of Moore’s Law (the doubling of computer circuits every two years) and Koomey’s Law (the energy efficiency of computers doubles every eighteen months):

A couple of basic physical rules underly the dizzying progress in electronics that we have seen over the past fifty years. Moore’s Law, attributed to Intel co-founder Gordon Moore, postulates that the number of transistors that can be placed on an integrated circuit of constant size doubles approximately every two years. Originally coined in 1965, Moore’s law has run more or less constantly ever since. It can’t continue indefinitely, if only because we’re getting close to the atomic scale; a silicon atom has a Van der Waals radius of around 200 picometres, and to build circuits that mediate electron transport we need discrete atomic-scale structures. It is not obvious that we can build electronics (or other molecular structures) with a resolution below one nanometre. So it’s possible that Moore’s law will expire within another decade.

Having said that, predictions of the imminent demise of Moore’s Law within a decade go back to the 1970s. And if we can’t increase the two-dimensional structure count on an integrated circuit, we may still be able to increase the number of structures by building vertically.

A newer, and more interesting formulation than mere circuit count is Koomey’s Law, proposed by Jonathan Koomey at Stanford University: that the energy efficiency of computers doubles every 18 months.

This efficiency improvement has held true for a long time; today’s high-end microprocessors require far less power per instruction than those of a decade ago, much less two or three decades ago. A regular ARM-powered smartphone, such as an iPhone 4S, is some 12-13 orders of magnitude more powerful as a computing device than a late 1970s-vintage Cray 1 supercomputer, but consumes milliwatts of power for computing (rather than radio) operations, rather than the 115 kilowatts of the Cray.

Taking them together, what do these two laws imply about the not-too-distant future?

July 19, 2012

Walter Olson: more red flags in the Regnerus study

Filed under: Health, Politics, USA — Tags: , , , , , — Nicholas @ 11:00

Some studies provide results that challenge common beliefs and understandings. Others reinforce them. But some studies are designed from the desired results backwards. The Regnerus study on gay parents’ influence on their children appears to be one of the latter. Walter Olson points out that even in its own terms, the study shows something different from what it is intended to:

By now almost everyone has had a whack at the recent Mark Regnerus (University of Texas) study claiming that young adults who report having a gay parent score worse on a range of life-success indicators than children from intact biological families. According to the study, these kids as young adults have lower educational attainment, are arrested more often, and have more trouble in their own relationships, among other problems. Critics have pointed out that the story is mostly one of collapsed heterosexual families, not “same-sex parenting”: The great majority of the kids were born to male-female couples, most of the presumedly gay dads and many of the moms didn’t get custody of their kids after their relationships dissolved, and few of the kids were actually raised through long periods by gay couples. LGBT advocates point out that sociologist Mark Regnerus accepted $695,000 from the anti-gay Witherspoon Institute to carry out the study.

But many critics have missed one of Regnerus’ most unexpected findings, one that may illuminate his study’s shortcomings. Specifically, and feeding into pretty much all the other problems, the study diagnoses children of gay parents as having a huge problem with poverty. Here’s Regnerus:

    Sixty-nine (69) percent of LMs [respondents with lesbian mothers] and 57% of GFs [those with gay fathers] reported that their family received public assistance at some point while growing up, compared with 17% of IBFs [those with intact two-parent biological families]; 38% of LMs said they are currently receiving some form of public assistance, compared with 10% of IBFs. Just under half of all IBFs reported being employed full-time at present, compared with 26% of LMs.

Those are big gaps. And of course they’re much at odds with the affluent image of gay families presented in both pro- and anti-gay-parenting literature as well as Modern Family-style popular entertainment. What do they signify?

Probably the biggest single reason is the one cited at the outset: This is mostly a survey of what happens when heterosexual families crack up. (Interestingly, if a married couple stayed together, they were counted as an “IBF,” no matter whether one or both partners pursued same-sex liaisons.) Decades of data indicate that children of family breakup do worse than children whose parents stay together, on many variables related to adult success. One reason, though not the only reason, is that they grow up significantly poorer.

Choice: re-evaluating the notion that too much choice is a bad thing

Filed under: Economics, Liberty, Science — Tags: , , , — Nicholas @ 09:37

There was a famous study several years ago that supposedly “proved” that providing too many choices to consumers was worse than providing fewer choices. At the time, I thought there must have been something wrong with the study.

The study used free jam samples in a supermarket, varying between offering 24 samples and only six, to test whether people were more likely to purchase the products (they were given a discount coupon in both variants). The result was that people who sampled from the smaller selection were more likely to actually buy the jam than those who had the wider selection to choose from. This was taken to prove that too many choices were a bad thing (and became a regular part of anti-consumer-choice advocacy campaigns).

Tim Harford explores more recent attempts to reproduce the study’s outcome:

But a more fundamental objection to the “choice is bad” thesis is that the psychological effect may not actually exist at all. It is hard to find much evidence that retailers are ferociously simplifying their offerings in an effort to boost sales. Starbucks boasts about its “87,000 drink combinations”; supermarkets are packed with options. This suggests that “choice demotivates” is not a universal human truth, but an effect that emerges under special circumstances.

Benjamin Scheibehenne, a psychologist at the University of Basel, was thinking along these lines when he decided (with Peter Todd and, later, Rainer Greifeneder) to design a range of experiments to figure out when choice demotivates, and when it does not.

But a curious thing happened almost immediately. They began by trying to replicate some classic experiments – such as the jam study, and a similar one with luxury chocolates. They couldn’t find any sign of the “choice is bad” effect. Neither the original Lepper-Iyengar experiments nor the new study appears to be at fault: the results are just different and we don’t know why.

After designing 10 different experiments in which participants were asked to make a choice, and finding very little evidence that variety caused any problems, Scheibehenne and his colleagues tried to assemble all the studies, published and unpublished, of the effect.

July 18, 2012

What is the best way to demonstrate care for the future?

Filed under: Economics, Environment, Government — Tags: , , , — Nicholas @ 08:29

According to Steven Landsburg, the answer is to cut capital taxes, and he makes a good case:

There are only three things you and I can do to make the future world a better place. First, we can consume less, leaving more resources behind. Second, we can work harder, planting trees, building factories and writing poems that will live on after we’re gone. Third, we can innovate, advancing science and technology so that our children’s children’s children can make better use of the resources they inherit.

As it happens, there’s one key policy variable that drives all three of these things, and that’s the tax rate on capital income (which includes interest, dividends, corporate income and capital gains). Capital taxes are a disincentive to save, and when people don’t save they consume instead. Capital taxes are a disincentive to work and a disincentive to innovate.

This is not a plea for lowering taxes in general, and it’s not a plea for making the tax system either more or less progressive. (If you want to soak the rich, there are plenty of things to tax besides capital.) As a matter of fact, this isn’t even a plea for lowering taxes on capital. It’s simply an observation that if your goal is to leave a better world for our descendants, then your best bet is to support lower capital taxes.

H/T to Tim Harford for the link.

July 9, 2012

The F-35 is “unaffordable and simply unacceptable”

Filed under: Military, Technology, USA — Tags: , , , , — Nicholas @ 12:46

Winslow Wheeler on the near-doubling of the F-35 price (so far):

On June 14 — Flag Day, of all days — the Government Accountability Office released a new oversight report on the F-35: Joint Strike Fighter: DOD Actions Needed to Further Enhance Restructuring and Address Affordability Risks. As usual, it contained some important information on growing costs and other problems. Also as usual, the press covered the new report, albeit a bit sparsely.

Fresh bad news on the F-35 has apparently become so routine that the fundamental problems in the program are plowed right over. One gets the impression, especially from GAO’s own title to its report, that we should expect the bad news, make some minor adjustments, and then move on. But a deeper dive into the report offers more profound, and disturbing, bottom line.

Notorious for burying its more important findings in the body of a report — I know; I worked there for nearly a decade — GAO understates its own results on acquisition cost growth in its one-page summary, which — sadly — is probably what most read to get what they think is the bottom line.

[. . .]

Set in 2001, the total acquisition cost of the F-35 was to be $233.0 billion. Compare that to the current estimate of $395.7 billion: cost growth has been $162.7 billion, or 70%: a lot more than what GAO stated in its summary.

However, the original $233 billion was supposed to buy 2,866 aircraft, not the 2,457 currently planned: making it $162 billion, or 70%, more for 409, or 14%, fewer aircraft. Adjusting for the shrinkage in the fleet, I calculate the cost growth for a fleet of 2,457 aircraft to be $190.8 billion, or 93%.

The cost of the program has almost doubled over the original baseline; it is not an increase of 42%.

July 3, 2012

US Army’s UCP camouflage pattern “makes soldiers more visible, not less”

Filed under: Military, Technology, USA — Tags: , , , — Nicholas @ 09:06

As I mentioned briefly last week, the US Army is abandoning their most recent camouflage patterned combat uniforms:

The United States military is abandoning its recently-adopted pixelated camouflage uniforms, according to articles this week in The Daily as well as Stars and Stripes.

The drab grey digital pattern, known as the Universal Camouflage Pattern (UCP), will be discarded after only eight years following mounting evidence that the colour scheme makes soldiers more visible, not less.

The articles pull few punches in their appraisal of the move to adopt the pattern in 2004.

“Army brass interfered in the selection process, choosing looks and politics over science,” reports Stars and Stripes, the official newspaper of the United States armed forces.

And while the Pentagon spent $5 Billion on the much-heralded uniforms, some of the earliest attempts to conceal soldiers on the battlefield were considerably less expensive.

The This is War blog has a discussion of the development of camouflage over the last century and a half.

June 29, 2012

US Army reluctantly admits USMC did better

Filed under: Military, Technology, USA — Tags: , , , , , — Nicholas @ 09:06

In developing camouflage, that is:

The U.S. Army has decided to scrap its digital pattern camouflage combat uniforms for the more effective, but more expensive, MultiCam. In the last decade, both the army and marines adopted new, digital, camouflage pattern field uniforms. But in Afghanistan, U.S. soldiers noted that the marine digital uniforms (called MARPAT, for Marine Pattern) were superior to the army UCP (Universal Camouflage Pattern). There’s been growing dissatisfaction with UCP, and it has become a major issue because all the infantry have access to the Internet, where the constant clamor for something better than UCP forced the army to do something. This is ironic because UCP is a variant of MARPAT, but a poor one, at least according to soldiers who have encountered marines wearing MARPAT. Even more ironic is that MARPAT is based on research originally done by the army. Thus some of the resistance to copying MARPAT is admitting the marines took the same research on digital camouflage, and produced a superior pattern for combat uniforms.

A digital camouflage pattern uses “pixels” (little square or round spots of color, like you will find on your computer monitor if you look very closely), instead of just splotches of different colors. Naturally, this was called “digital camouflage.” This pattern proved considerably more effective at hiding troops than older methods. For example, in tests, it was found that soldiers wearing digital pattern uniforms were 50 percent more likely to escape detection by other troops, than if they were wearing standard green uniforms. What made the digital pattern work was the way the human brain processed information. The small “pixels” of color on the cloth makes the human brain see vegetation and terrain, not people. One could provide a more technical explanation, but the “brain processing” one pretty much says it all.

June 5, 2012

The US military’s SF research emporium

Filed under: Media, Military, Science, Technology, Weapons — Tags: , , , , , , — Nicholas @ 08:18

John Turner sent me a link to this amusing little survey of what the US military’s R&D organization is willing to admit they’re working on and how it might be helpful in case of an alien invasion:

As summer blockbuster season kicks into high gear, big-budget action movies like The Avengers, Battleship, and Prometheus remind us that there’s one thing that unites Americans: Our shared fear of an alien attack. They also remind us that when the invading space fleet arrives, humanity is not going to surrender without a fight to our intergalactic invaders. Instead, we will band together to fight off their incredibly advanced weaponry with our … well, with what, exactly? Are we really ready to battle our would-be alien overlords?

Luckily, the Pentagon’s Defense Advanced Research Projects Agency, better known as DARPA, as well as some of the world’s largest weapons manufacturers, are dreaming up the weapons of the future today. With the help of everything from lasers on jets to hypersonic planes to invisibility cloaks, we just might be able to make the battle for Earth a fair fight. You may think we’re joking, but why else would NASA be uploading The Avengers to the International Space Station if not as a training manual? Here’s a look at some of the most space-worthy inventions being cooked up now.

An issue for any unmanned, armed vehicle (whether land, sea or air) is the security of communications from the controller to the vehicle. Recent use of such devices has almost always been in combat against relatively low-tech opponents who did not have jamming or hacking capabilities (although the UAV forced down in Iran may signal the end of the easy period for combat UAVs). Earlier discussions of benefits and drawbacks to unmanned fighters are here, here, and here.

April 17, 2012

Buckyballs: the silver bullet for aging?

Filed under: Health, Science — Tags: , , — Nicholas @ 09:34

Well, it’s been shown to almost double normal lifespan … in rats. But in only one study so far:

In the current study researchers fed the molecule dissolved in olive oil to rats and compared outcomes to a control group of rats who got plain olive oil.

The main question they wanted to answer was whether chronic C60 administration had any toxicity, what they discovered actually surprised them.

“Here we show that oral administration of C60 dissolved in olive oil (0.8 mg/ml) at reiterated doses (1.7 mg/kg of body weight) to rats not only does not entail chronic toxicity,” they write “but it almost doubles their lifespan.”

“The estimated median lifespan (EML) for the C60-treated rats was 42 months while the EMLs for control rats and olive oil-treated rats were 22 and 26 months, respectively,” they write.

Using a toxicity model the researchers demonstrated that the effect on lifespan seems to be mediated by “attenuation of age-associated increases in oxidative stress”

March 28, 2012

Speaking of science, here’s a new development in “thermal cloaking”

Filed under: France, Science, Technology — Tags: , — Nicholas @ 09:42

Lewis Page at The Register walks us through the high points of a new development from France:

Top boffins in France have come up with a radical new take on the “cloaking” and invisible-shed physics breakthroughs of recent years. They have designed a technology which instead of bending microwaves or light can shield an object from heat — or concentrate heat upon it.

“Our key goal with this research was to control the way heat diffuses in a manner similar to those that have already been achieved for waves, such as light waves or sound waves,” says Sebastien Guenneau, of Aix-Marseille uni and France’s Centre National de la Recherche Scientifique (CRNS).

“Heat isn’t a wave — it simply diffuses from hot to cold regions,” Guenneau adds. “The mathematics and physics at play are much different.”

For now designed only in two dimensions, Guenneau and his colleagues’ approach involved shaping isotherms — lines showing density of heat flux transferring from point — so as to make heat travel around a given area rather than into it.

“We can design a cloak so that heat diffuses around an invisibility region, which is then protected from heat. Or we can force heat to concentrate in a small volume, which will then heat up very rapidly,” Guenneau says.

March 10, 2012

Some diseases may be caused by “endogenous” retrovirii

Filed under: Health, Science — Tags: , , , , — Nicholas @ 12:17

Matt Ridley on some recent discoveries in genetics and medicine that may help to explain certain diseases like multiple sclerosis:

The virus implicated in multiple sclerosis is called HERV-Fc1, a bizarre beast called an “endogenous” retrovirus. What this means is that its genes are part of the human genome. For millions of years, they have been integrated into our own DNA and passed on by normal heredity. It was one of the shocks of genomic science to find that the human genome contains more retroviral than “human” genes: some 5% to 8% of the entire genome.

Normally, the genes of endogenous retroviruses remain dormant, but — a bit like a computer virus that springs into action on a trigger — something wakes them up sometimes, and actual viruses are made from them, which then infect other cells in the body. The Danish scientists suggest that this is what happens in multiple sclerosis. Bjørn Nexø of Aarhus University writes that “retroviral infections often develop into running battles between the immune system and virus, with the virus mutating repeatedly to avoid the immune system, and the immune system repeatedly catching up. One can see the episodic nature of multiple sclerosis as such a running battle.”

The possibility that you can inherit the genes of a virus blurs the distinction between a genetic and an infectious disease. The HERV-Fc1 genes lie on the X chromosome. Since women have twice as many X chromosomes as men, this might explain why some forms of MS are more common in women. Dr. Nexø concludes hopefully: “The finding that a disease is caused by an infectious agent is an encouraging one. These are the diseases which we know best how to treat.

The research also appears to show a link between cat ownership and schizophrenia:

Human beings can also catch toxoplasma from cats, and it’s known to affect behavior: altering personalities, slowing reaction times and increasing the risk of car accidents. More than 20 studies have now found an association between schizophrenia and toxoplasma. Schizophrenia is more common among those who had pet cats in their childhood homes (but not in those who had pet dogs).

Indeed, some scientists think that schizophrenia only became common, around 1870, when keeping cats as indoor pets became fashionable. The parasite has genes for dopamine, a neurochemical found in excess in schizophrenics.

February 25, 2012

Tim Harford: The problem of “interdisciplinary problems”

Filed under: Economics, Science, Technology — Tags: , , , — Nicholas @ 10:25

Tim Harford recently visited Oxford Martin School to discuss the phenomenon of problems that are seen as intractable when viewed from within a “silo” or single discipline, but which yield solutions when approached in co-operation with multiple disciplines:

In academia, the challenge of encouraging interdisciplinary research is at least recognised as a problem. The advancing frontier of scientific knowledge forces most researchers to specialise in ever narrower fields and, as a result, collaboration between these silos is essential. I recently visited the Oxford Martin School, a seven-year-old initiative designed to foster cross-disciplinary projects at the University of Oxford. I talked to the school’s director, Ian Goldin, about the challenges of breaking down academic silos.

He thinks these silos are mostly artificial. Academic journals are largely specialised rather than interdisciplinary and official funding bodies shy away from interdisciplinary projects. The result is that academics with interdisciplinary interests have few ways to fund the research and few credible outlets for publishing the results. The Martin School has funding, but most of the researchers are either junior, with some freedom to experiment, or professors so senior they no longer need to worry about their publication record. The mid-career academics are missing. It is nice to hear the tenure system sometimes produces the hoped-for courage and independence, but not so nice that there is no career track for interdisciplinary researchers.

[. . .]

If problems are one focal point for collaboration, tools can be another. An example: systems needed to deal with the gigantic data sets generated in finance, astronomy and oceanography. Such tools naturally bring together computer scientists and the statisticians, economists and scientists who might use the data. Goldin points to “crowdsourcing” as a second example of a cross-disciplinary tool, complexity science as a third and (optimistically, I feel) practical ethics as a fourth.

Perhaps the real lesson is that promoting cross-disciplinary research need not require a mysterious blend of social-networking tools and funky collaborative architectural spaces. All that is sometimes required is a shared problem, or a shared set of tools, and, above all, the money to pay for the job to be done.

February 14, 2012

Envisioning the all-online university

Filed under: Media, Technology — Tags: , , , — Nicholas @ 09:56

Megan McArdle on the recent announcement that MIT will be offering online programs (at lower cost than regular courses) and if this is a sign of the future (as it almost certainly is) what changes will occur in the realm of higher education:

I can see all sorts of factors that might combine to preserve the status quo, from signaling and status and networking, to the desire of college students for a four-year debt-financed semi-vacation. On the other hand, disruption never looks inevitable until it suddenly is — if you’d told someone in 1955 that GM was going to have its lunch eaten by some Japanese upstart, they would have laughed until the tears came. So it’s interesting and maybe even useful to contemplate what the college system would look like if this sort of distance learning becomes the norm.

1. Education will end up being dominated by a few huge incumbents. [. . .]

2. Online education will kill the liberal arts degree. [. . .]

3. Professors (course developers) will be selected for teaching instead of research brilliance. [. . .]

4. 95% of tenure-track professors will lose their jobs. [. . .]

5. The corollary of #4 is the end of universities as research centers. [. . .]

6. Young job-seekers will need new ways to signal diligence. [. . .]

7. The economics of graduate school will change substantially. [. . .]

8. Civil society will have to substitute for the intense friend networks that are built at college. [. . .]

9. The role of schooling in upward mobility will change. [. . .]

10. The young will have a much lower financial burden in their 20s. [. . .]

11. The tutoring industry will boom. [. . .]

12. If the credentials become valuable, cheating will be a problem. [. . .]

February 11, 2012

Yet another theory on the solar effect on the Earth’s climate

Filed under: Environment, Science, Space — Tags: , , — Nicholas @ 12:14

You’d automatically assume that the sun is a major factor in the climate, but this theory is non-intuitive:

That man is the Danish physicist Henrik Svensmark, who seems to have discovered the most important factor that actually regulates Earth’s climate, and who is quietly in the process of proving it.

[. . .]

Let me briefly sum up Svensmark’s theory. The temperature of the Earth, he argues, is regulated by the intensity of solar radiation, but not in the obvious way. It is not that the increase is solar radiation heats the Earth directly. (It does, of course, but not to a sufficient degree to explain climate variations.) Rather, an increase in solar radiation extends the Sun’s magnetic field, which shields Earth from cosmic rays (highly energetic, fast-moving charged particles that come from deep space). How does this affect the climate? Here is the crux of Svensmark’s argument. When cosmic rays hit the atmosphere, he argues, their impact on air molecules creates nucleation sites for the condensation of water vapor, leading to an increase in cloud-formation. Since clouds tend to bounce solar radiation back into space, increased cloud cover cools the Earth, while decreased cloud cover makes the Earth warmer.

So if Svensmark is right, lower solar radiation means more cosmic rays, more clouds, and a cooler Earth, while higher solar radiation means fewer cosmic rays, fewer clouds, and a warmer Earth.

Those who have followed the global warming controversy over the years may recall that cloud-formation is one of the major gaps in the computerized climate “models” used by the consensus scientists to predict global warming. They have never had a theory to explain how and why clouds form or to account accurately for their effect on the climate. Svensmark has smashed through this glaring gap in their theory.

« Newer Posts — Older Posts »

Powered by WordPress