/dev/mull

there's a new paper about treefrogs that have oddly electronic-sounding calls. or at least, that's the authors' hook. i'd say just a couple of them sound like new trek sound effects, but i will admit that boldly citing star trek has made the paper a bigger splash than not. as they say,

[W]e primarily aim to honor the focus on inspiring – even if not always accurate – science and nature conservation prevalent in numerous Star Trek episodes [...], and the spirit of discovery and scientific exploration that it fosters. It seems also fitting that finding these frogs sometimes requires considerable trekking; pursuing strange new calls[, seeking] out new frogs in new forests; [and] boldly going where no herpetologist has gone before[!]

i'm gonna skip the formal nomenclature, and just say that kirk's treefrog and picard's treefrog have what it takes to be good ST:NG SF/X. kirk's treefrog could be a tricorder indicating the presence of a plot element, and picard's treefrog sounds like a communicator's incoming call sound followed by a double-beep confirming call pick-up. the rest just sound like birds chirping to me. (.wav files; all of the sound samples, including those two, are with the supplemental data.)

a bunch of folks with the USGS claim to have evidence for early human habitation of what's now california. this is apparently an improvement over their previous claim about that. but it appears to be press-release science, not even a poster at a conference, much less a peer-reviewed paper with publically-accessible raw data (the current gold standard).

nevertheless, ( they have a nice visualization of what they claim. )

also in the new science imagery department, osiris-rex launches an escape pod ejects its sample return capsule:

( cut for moderately large gif )

the glare at the top is the sun, and the limb of the earth is just visible at the left.

NASA recently announced that the kepler mission had found a big whack of exo-solar planets. some of them are even the habitable zone™ of stars, even though we really don't know what we mean by that¹.

usually, these announcements are accompanied by pretty pictures showing earthlike planets with atmospheres, seas, and clouds. unfortunately for science, we don't actually know any of that². all we know about these planets is an estimate of their orbital period, which gives us an estimate of their distances from their stars, and a guess about how big they are. that's it. no air, much less seas and clouds and other features indicating people will eventually be able to start new lives in the off-world colonies.

so i'm insanely pleased that NASA is running a scientifically accurate picture of what we know:





and it's a very information-dense image too: not only does it show the planets' orbits and sizes, it provides comparisons of those to venus, earth, and mars; it tells us about the color of the stars they orbit; and even whether what we're looking at was just reported or old news. be still, my geeky heart!

it also makes some other important points: even assuming any of these worlds had air we could breathe, they're probably not very habitable. most of them are much larger than earth, which tends to suggest much higher surface gravity. they're also -- again, making the assumption of air we can breathe -- either too hot or too cold for us.

more subtly, the light on many of these worlds is very dim to our eyes. the green band indicating the handwaved habitable zone goes a lot further out for the cooler stars. that's because more of their light is in the infrared. that's great for keeping the planet warm, but not so much for seeing by. that's good to know, but the PR folks at NASA don't like to play that up. it gets in the way of telling a good story.

1: loosely, the habitable zone™ is defined where water could be liquid -- anywhere from just below boiling to just above freezing -- making some hefty assumptions about the conditions on the planet. notably, we have to assume an atmosphere (reasonable, given these planets' likely masses), assume an atmospheric composition (entirely unknown), assume a surface albedo (reflectivity, also entirely unknown) and ignore a bunch of ugly feedback effects among the assumptions. notably, it's easy to find combinations that lead to run-away greenhouse effects, like venus has, or run-away freezing, leading to "snowball" planets in perpetual ice ages.

but if you ignore all that, it's great science. :)

2: the next generation of orbital telescopes will start to be able to look for atmospheres around exo-planets. some of them will be able to tell us about the atmospheric compositions. but you'll need to wait a few years for the earliest of those results.

UPI is running a story discussing a paper about how the genetic distances between walnut forests in asia correlate with the linguistic distances of the peoples who live near those forests. which is to say, walnuts were probably passed along trade routes, resulting in a common origin for linguistic and genetic similarities.

i'm kinda pleased by this paper for a bunch of reasons.

* for a change, the statistics look right. be still, my beating heart! if the stats are right, there's a chance, however small, that the rest of the paper is right. genetic distances were measured by a standard method (in addition to other methods), so they're probably correct, or at least no more wrong than any other estimate. i have no clue about the linguistic distances, but probably some of my readers do.

* trees that produce both edible fruit and useful wood¹ have struck me as just eminently reasonable things for our ancestors to have domesticated, but we don't know much about how they did it. now we know a little more about it.

* apparently, there's a common proto-word -- i have no idea what it technically is called -- for walnut that's common to languages from central asia to the levant (sogdian and tajik to aramaic). i knew the indo-european languages tended to have such, but it's an impressive example, and seems to argue that walnut (wood and fruit) has been extensively traded for a long, long time.

* i've been to some of the oddball places the paper talks about, like kashgar and the hunza valley. they're on (one of) the silk road(s).² turns out that the silk road actually matters for something. who knew?

1: there aren't that many trees which do both. my list looks like: almond, apple, cherry, chestnut, maple (sorta, if you count syrup), oak (were acorns eaten outside of north america?), peach, pear, and walnut. some kinds of pines produce edible nuts, but i'm not sure if they're the same as yield softwoods.

2: as the paper notes, the silk road is more a collective term for a set of trade routes than a specific path. kashgar, though, is on most of them, since its location makes it a good crossroads.

Researchers share MRI video of knuckle cracking

EDMONTON, Alberta, April 16 (UPI) -- University of Alberta researchers shared an MRI video showing what goes on inside the hand when someone cracks a knuckle.

The video, released as part of a study published Wednesday in journal PLOS ONE, shows the knuckle joint separating and creating a bubble of gas in the synovial fluid between the bones when the researchers used a cable to pull the man's finger and crack his knuckle.

...

250-year-old sex toy found in 18th-century Polish toilet

GDANSK, Poland, April 16 (UPI) -- Archaeologists examining the site of an 18th-century latrine said they discovered a 250-year-old sex toy apparently dropped into the toilet by its user.

The Regional Office for the Protection of Monuments in the Baltic city of Gdansk said the sex toy, consisting of an artificial penis and testicles, dates from the second half of the 1700s and was found "preserved in excellent condition."

...

i clearly went into the wrong area of science.

i mentioned a new paper discussing the cost-effectiveness of security measures to defend against attacks at airports, and its conclusion that none of them are cost-effective. a little bluebird of happiness did fly by and leave me a copy of the paper, and i've now read it.

the paper has both strengths and weaknesses. it's strong in the math side of the reasoning, which means their methods are defensible and subject to analysis. but it's weak in the empirical data side of the reasoning, because well, there haven't been all that many terrorist attacks in the civilized parts of the world, and most of them have been unsuccessful. and while bodies are easy to count, many costs aren't, especially so-called indirect costs, such as lost income from reduced tourism.

still, the authors' math provides a framework to look at the problem, and their conclusions can be neatly summed up in two paragraphs, a chart, and a graph:

The decision problem [about what security measures to use] can be recast as a break-even analysis. The minimum attack probability for security measures to be cost effective is selected such that there is 50% probability that the benefit will equal the cost (see Table 7). As expected, break-even probabilities are less than observed attack probabilities of 0.2-0.5% only for adding skycaps, check-in personnel, and more TSA lines (security measure 2), enhanced training of airport police to SWAT standards (3), and add curbside blast deflection and shatterproof glass (5). All other security measures require considerably higher attack probabilities than those currently being observed for them to be cost-effective. However, a decision-maker may wish the likelihood of cost-effectiveness to be higher before investing millions of dollars in security measures e to say 90% so there is more certainty about a net benefit and small likelihood of a net loss. Table 7 also shows the minimum attack probabilities needed for there to be a 90% chance that security measures are cost-effective. In this case, the threshold attack probabilities more than double when compared to the break-even analysis. The results are not overly sensitive to the probabilistic models used.

Clearly, due to the uncertainties inherent in such an analysis, a sensitivity analysis is recommended. Doubling the cost of physical damages or loss of life has a negligible effect on NPV or BCR [Net Present Value or Benefit/Cost Ratio], which illustrates that in this situation the expected losses are dominated by indirect losses. Many of the assessed security measures would only begin to be cost-effective if the current rate of attack at airports in the U.S., Europe, and the Asia-Pacific increases by a factor of 10-20. Thus, input parameters [how many people killed and how much stuff is blown up every year] can be doubled or halved and this would not change the fundamental findings herein that many airport security measures fail a cost-benefit assessment.

(my emphasis)

their table 7:



and their figure 1, which plots their conclusions as continuous curves:



the authors further go on to note that these conclusions don't include costs to passengers. if those are included, adding more workers to shorten lines (and thus reduce the number of people gathered in one place) is valuable, since people want that; armoring the front windows of airports and adding some jersey barriers along curbs is neutral or close to it¹; and better training of cops is only a slight cost, once one includes the possibility that they'll over-react (which, of course, can be completely discounted in the US).

1: obviously, if the fortifications become too elaborate and/or numerous, they're a pain in the ass, but even at the over-secured airports i travel thru, they're not.

there are higher-order effects the authors note but can't do anything with. for example, they point out that obviously harder defenses probably reduces the probabilities of attacks taking place (because terrorists conclude they can't kill many people) and reduce the damage they cause. both of these paradoxically make defenses harder to justify, since the observed attack probabilities should then fall.

regardless, their conclusion is inescapable: by and large, we waste money and people's time on pretending to make them safer, and the most justifiable method of doing so -- reducing lines by adding staff -- is in entirely the opposite direction than the government and airlines have taken.

since i've been sick, i haven't posted a few interesting things...

1) in the least surprising of the NSA revelations, the spooks are trying to attack TOR. real shocker there; in fact, if they weren't trying, i'd have the folks who run the place fired.

also no surprise that they're using traffic analysis and TOR nodes they control to undermine the silly thing. those are the obvious routes of attack for any national government. as of the secret report's date (june 2012), the NSA didn't control enough TOR nodes to do anything significant to TOR, but that's just a matter of getting more nodes on it, which is mostly a HUMINT/social engineering problem. (technical means will work somewhat, but why bother with those if you can just convince people to trust your pet relay?)

2) take with salt:



from Nightmares and bedtime stories, in the economist, which relies on data from "real clear politics"'s Public Approval of Health Care Law. which in turn relies on how the pollsters asked their questions and surveyed the country, so i'd salt this to taste.

3) ...and on the fifth day during the cambrian explosion, random forces created a whole shipload of living things:

a simplified chart of when the major groups of living things appeared:



the red box indicates the period the authors were discussing, which ends just after the cambrian explosion.

and a spaghetti plot showing the complexity of what actually happened (current best guesses):



(more legible but much larger version)

both from Causes of the Cambrian Explosion, by m. paul smith and david a. t. harper.

in short, the authors write that geological changes (in red boxes in the chart) ultimately caused the origin of life as we know it today. sea level rise begat erosion; erosion begat calcium pouring into the seas; lots of calcium in the water begat critters trying to get rid of it; and having calcium to ditch led to biomineralization (bones and shells). and likewise for the other child of sea level rise: shallow seas increasing the world's habitable volume... and so on. so, all the fancy biology we're used to -- complex genomes, new body plans (bilateral ones, like us), and macrophagy (eating multicellular prey, also like us) -- seems to have started with weathering rocks.

it was discovered this past summer in a croatian cave. it's quite small; the mouth of its shell is a hair under 1.5mm (about the thickness of a nickel; i have no idea what 1.5mm is in amuricin).

the paper has a few more pictures of the critter, and a diagram showing the depths the discoverers went to find this thing:

( cut for large image )

Launched over 35 years ago, Voyagers 1 and 2 are on an epic journey outward from the Sun to reach the boundary between the solar plasma and the much cooler interstellar medium. The boundary, called the heliopause, is expected to be marked by a large increase in plasma density, from about 0.002 cm−3 in the outer heliosphere, to about 0.1 cm−3 in the interstellar medium. On 9 April 2013, the Voyager 1 plasma wave instrument began detecting locally generated electron plasma oscillations at a frequency of about 2.6 kHz. This oscillation frequency corresponds to an electron density of about 0.08 cm−3, very close to the value expected in the interstellar medium. These and other observations provide strong evidence that Voyager 1 has crossed the heliopause into the nearby interstellar plasma.

In Situ Observations of Interstellar Plasma With Voyager 1; D. A. Gurnett, W. S Kurth, L. F. Burlaga, N. F. Ness

today interstellar space, tomorrow eventually the stars!

the kepler planet-hunting mission, that is. or rather the people who want to popularize its discoveries make me crazy, because they seem too willing to mix fiction with their science.

( the good, the bad, and the ugly )