TFA explains how to set a bit (from 0 to 1), but not how to clear the bit.
The actual paper in inaccessible to me (without $$$), even from a university, as it is pre-publication. (I think it will become available on the 17th.)
Yeast are not bacteria, they're fungi and are more closely related to you than to any bacteria (or, for that matter, any plant.)
My understanding is that so long as there is sugar around, yeasts will metabolize it to alcohol so as to poison competitors for the food source, and later metabolize the alcohol once the sugar runs out. However, I'm not sure I got this from a reliable source, and I couldn't find confirmation in a quick web search. In any case, I think it is one of those rare evolutionary innovations and so is an exceptional case.
Scientists have been sending up experiments on sounding rockets for many decades (and more recently Pegasus has been available also). So there really is only a gain for experiments that require human intervention to run. And can be done in 5 minutes. And don't contain hazardous substances. And are small enough to take on one of these launches. What does that leave?
(The original article is inaccessible to me - sorry if these questions are answered there.)
Neither the BBC article nor the abstract* of the original paper mention 'life on Earth may have come from outer space'. They say that the nitrogen may have come from outer space. From the abstract "we speculate that [ammonia rich comets] delivery to the early Earth could have fostered prebiotic molecular evolution" (emphasis mine).
* Alas, my institution only has free access to PNAS articles older than 6 months, so I haven't seen the paper. I could probably get up and read it in the library, but reading a paper off paper just seems morally wrong. Won't somebody please think of the trees?
From your link: "Alvin Wong believes the program would be less expensive today because trainers could use flight simulators to train the birds rather than taking them up in real helicopters."
Only if we could make flight sims with sufficient visual quality to fool a pigeon. Remember, our RGB monitors are adapted to our colour perception, they are not a good reproduction of reality.
My understanding* is that at install time, an Android app has to list what permissions it wants to be able to operate. If I was installing some new wallpaper and it demanded internet access, I'd abort instantly. So does this attack only work against naive users?
* I don't have, and have not used, an Android or other smart phone
Trains are much tougher than planes, and can take more damage without catastrophic failure. Nothing terrible is likely to happen to you on a high speed train unless it derails. To derail it from inside, a terrorist needs to penetrate the floor and damage the wheels. They need to achieve this with equipment you carried on, in a crowded environment. By comparison, if a terrorist want to derail the train from the outside, they have thousands of km of track to target without a train floor in the way, no onlookers, and potentially a truck's worth of equipment. (Indeed, a truck as the equipment would work pretty well.)
Security screening the passengers really doesn't make sense. So, yes, probably you'll wind up going through the same security as you would at an airport.
Surveys such as this tend to find lots of large planets close to their stars. It is worth pointing out that this is at least partly because such planets are easier to detect, and does not necessarily mean they are a high proportion of planets in the galaxy.
Kepler detects changes in stellar brightness due to transiting planets. The closer a planet is to its star, the less precise the alignment has to be for us to observe a transit. Also, the closer it is, the faster it orbits, and the more likely we observe a transit in the limited time we're observing that star. This second factor will become less restrictive as the Kepler mission runs for a longer time. (I presume they need at least two, possibly more, transits before they claim a detection.) Large planets will also give a larger, easier to detect change in brightness.
The other major way of detecting planets is spectroscopically: the planet wobbles the star slightly, and we observe the Doppler shift. This favours massive planets (they wobble the star more) and close planets (they wobble the star faster.)
There have I think also been a few cases where clever interferometry has allowed direct imaging of extrasolar planets. I don't know what the selection effects on this are - further away means easier to separate from the star (good) but less bright (bad.)
All the Planck units involve powers of G, the universal gravitational constant. We have measured G to an accuracy of 10^-4. We want a definition of mass which is accurate to about 3x10^-8. Can you spot the problem?
I'm not at all sure you are serious, but enough people seriously hold this opinion that it is worth responding.
A good system of units needs: 1) Base units which are well defined and independently reconstructible (i.e. a suitably equipped lab can calibrate their equipment purely from the definition of the units.) 2) Logically constructed compound units (e.g. units of force are derived from the units of mass, time and distance.) 3) Logically constructed convenience units (e.g. kilometres for use for distances which would be an inconveniently large number of metres.) 4) To be widely used.
The initial choice of your base units is largely arbitrary - whether it was a from a not-very-accurate measure of a king's foot size or from a not-very-accurate measure of the Earth's circumference. Item (1) can be satisfied equally well (or, in the case of mass, badly) by the metric or imperial systems. The definition of the metre has long since changed from the size of the Earth to quantities measurable in a lab (as has the definition of the foot.)
The SI system (based on metric measures) beats the imperial system hands down on items 2 and 3, and because of this now has a large advantage also on item 4.
Item 2: In Imperial you might measure (heat) energy in BTU and mechanical energy in some mixture of foot-pounds-seconds, but then you need a conversion factor to compare the two. Such conversion factors are never needed in SI.
Item 3: Imperial also messes up the convenience units by having lots of weird conversion factors (e.g. an acre is (I think) a furlong by a chain. How many square feet is that? How many ounces in a ton?*) Metric uses convenience units constructed from base units via consistently named factors of 10 or 1000.
You can't use the current problems with the kilogram as a reason to prefer imperial to metric, as imperial will be just as prone to exactly the same problems. The (UK) Imperial pound is similarly defined by the mass of a unique artifact. In the US, it is defined relative to the kilogram. Mass is the last base unit which doesn't satisfy requirement (1), and the efforts to fix this are what has triggered this entire debate.
One could go a step further, and define your fundamental units in terms of fundamental physical constants (i.e. the Plank mass, Plank time and Plank distance, charge on an electron, etc.) In such a system of units, the speed of light is 1, the formula for the energy of a photon doesn't need a constant in it etc. In practice, we can't use such a system, because we can't measure (in particular) the universal gravitational constant G with sufficient accuracy. Every time we got a better measure of G, our entire system of units would need to be updated. (I.e. with current technology, this system can't satisfy requirement (1) above.)
* And how many different sorts of ounces and tons are there? It is quite a few.
Because it is really hard to measure exactly a mole of hydrogen. It is much easier to measure a mole of a crystalline solid because you know very accurately the density of the atoms. This is precisely what one of the two methods does - trying to define the kilogram as the mass of a precisely sized sphere of pure crystalline silicon. I think silicon was chosen because we know how to make it exceptionally pure, and in a single crystal of sufficient size.
I've met a number of magnetic-sticks-and-balls toys, and Geomag is by far the best (of the ones I've tried.)
You can get really cheap stick-and-ball toys where an icosohedron is a really difficult and delicate thing to make.
The next step up was Magnetix. The magnets were much better than no-brand, and you can get panels (squares, pentagons etc.) with magnets at each vertex. However, the magnets can fall out of the plastic bits (which in turn makes them a swollowing hazard), the stick lengths are not sufficiently accurate, panels are hard to come by, and the magnets still could be better.
Geomag uses much stronger magnets than Magnetix, and built with much better precision. They also do panels in a much better way: there are cheap plastic panels which clip in between the sticks and balls to transform (e.g.) a 5-cycle of sticks and balls into a rigid pentagon. Although they sell sets without panels, to my mind you're missing at least half the point if you don't have them.
For example this is strong enough to pick up. By comparison the much smaller snub dodecahedron is extremely delicate and hard to make with Magnetix.
I do worry about this. While lithium isn't consumed by the batteries, electrifying a substantial portion of the world's road transport would require a huge increase in world wide lithium production.
World production was about 25,400 tons in 2008, and reserves about 10,000,000 tons, so we have about 400 years of production at current rates.
If an electric car typically contains a 200kg battery which is 5% lithium by weight, that is 10kg per car. (These numbers are slightly better than guess work, but not by much.) If we manufacture say 200 million new cars per year, that is about 2 million tons of lithium per year, or about 20% of world reserves per year. (200 million cars with 10 year average lifespan would give 2 billion cars worldwide.)
So it does indeed look like it could be a big problem. Eventually there will be a significant contribution from recycling the previous generation of electric cars, but that is some way away. There may also be major reserves yet to be found, as it has not previously been worth looking hard for them.
In any case, even if the reserves exist, we're likely to need on the order of an 80-fold increase in mining capacity if we're to convert the world to battery cars.
The NZ market is significantly different from the US market in that cable TV has only ever been a very minor player in the market. (I only some suburbs of Wellington, and only in the last 10 years or so, but I haven't tried to keep current on this.) Subscription TV comes by satellite to decoder boxes. This means that currently cable modem is not an option, but I'm guessing that fiber-to-the-home will get used for cable TV service once it is installed.
I'd count dating historical observations of solar eclipses as forensic astronomy, and I think was done well before 30 years ago. Here are some examples.
There are also celestial alignments of pyramids and stone circles - although it would have to be a stellar alignment to count, as the sun doesn't change its path over historical times.
Tor lets you collect your porn anonymously, but at a heavy bandwidth price. The three letter agencies are (we guess) providing Tor nodes with lots of bandwidth so as to be able to sniff the exit traffic.
Result? The NSA is subsidising your anonymous porn collection!
You don't have to care about encryption so long as you don't mind if the NSA has sniffed your porn before you do.
I saw this on the BBC website too, but neither article tells me how it is to the advantage of the hackers to give random people big telephone bills. Do the hackers own some little phone company which the calls are going through? Do they have some overpriced premium number connecting to a computer in Scott Base which recites astrology readings in a synthetic voice?
More seriously: why should the phone OS allow a game to initiate phone calls? (I really hope the answer is 'the OS has a bug' rather than 'that's how they designed it.')
not (as it might have been, for all I knew) a new compiler, for example.
Slashdot summaries have this annoying habbit of assuming that because the author is thoroughly familiar with the technology, everyone else is too and you can just reel of project names with no further explaination.
You should see what it is like in Australia. Here's an extract from one ISP's web page, I think this is typical.
"Monthly usage allowance means combined upload and download data transfer. 1 GB = 1000MB. Unused usage allowance is forfeited each month. Additional usage charged at $0.15/MB and is capped at $300 from the first bill cycle on or after 29th of November 2009."
That's $150/GB for excess. And they short-change us on the GB's they provide in the cap.
Having just skimmed the paper, it doesn't look like they account for CO2 removed from the atmosphere by the growing trees. Here's a quick calculation of that: Assumption 1: all plant matter which does not make its way into paper is burned, or otherwise releases its carbon as CO2, hence is neutral for this analysis. (It could net contribute to greenhouse if it releases as methane instead.) Assumption 2: paper is 100% cellulose.
Cellulose is a polymer of (C5 H10 O5), which means that it is 4/9 carbon by weight. One unit of carbon burns to produce 11/3 = 3.667 units of CO2. So one unit of paper would burn to produce 44/27 = 1.630 units of CO2, and conversely, 1.63 tonnes of CO2 were removed from the atmosphere to make that paper.
So we're still behind on CO2. And, of course, there are all sorts of other environmental costs.
Slashdot Mirror
TFA explains how to set a bit (from 0 to 1), but not how to clear the bit.
The actual paper in inaccessible to me (without $$$), even from a university, as it is pre-publication. (I think it will become available on the 17th.)
Yeast are not bacteria, they're fungi and are more closely related to you than to any bacteria (or, for that matter, any plant.)
My understanding is that so long as there is sugar around, yeasts will metabolize it to alcohol so as to poison competitors for the food source, and later metabolize the alcohol once the sugar runs out. However, I'm not sure I got this from a reliable source, and I couldn't find confirmation in a quick web search. In any case, I think it is one of those rare evolutionary innovations and so is an exceptional case.
Scientists have been sending up experiments on sounding rockets for many decades (and more recently Pegasus has been available also). So there really is only a gain for experiments that require human intervention to run. And can be done in 5 minutes. And don't contain hazardous substances. And are small enough to take on one of these launches. What does that leave?
(The original article is inaccessible to me - sorry if these questions are answered there.)
Neither the BBC article nor the abstract* of the original paper mention 'life on Earth may have come from outer space'. They say that the nitrogen may have come from outer space. From the abstract "we speculate that [ammonia rich comets] delivery to the early Earth could have fostered prebiotic molecular evolution" (emphasis mine).
* Alas, my institution only has free access to PNAS articles older than 6 months, so I haven't seen the paper. I could probably get up and read it in the library, but reading a paper off paper just seems morally wrong. Won't somebody please think of the trees?
From your link:
"Alvin Wong believes the program would be less expensive today because trainers could use flight simulators to train the birds rather than taking them up in real helicopters."
Only if we could make flight sims with sufficient visual quality to fool a pigeon. Remember, our RGB monitors are adapted to our colour perception, they are not a good reproduction of reality.
My understanding* is that at install time, an Android app has to list what permissions it wants to be able to operate. If I was installing some new wallpaper and it demanded internet access, I'd abort instantly. So does this attack only work against naive users?
* I don't have, and have not used, an Android or other smart phone
Thank you - that is much more informative than the fluff-piece pointed to by the summary.
Trains are much tougher than planes, and can take more damage without catastrophic failure. Nothing terrible is likely to happen to you on a high speed train unless it derails. To derail it from inside, a terrorist needs to penetrate the floor and damage the wheels. They need to achieve this with equipment you carried on, in a crowded environment. By comparison, if a terrorist want to derail the train from the outside, they have thousands of km of track to target without a train floor in the way, no onlookers, and potentially a truck's worth of equipment. (Indeed, a truck as the equipment would work pretty well.)
Security screening the passengers really doesn't make sense. So, yes, probably you'll wind up going through the same security as you would at an airport.
Surveys such as this tend to find lots of large planets close to their stars. It is worth pointing out that this is at least partly because such planets are easier to detect, and does not necessarily mean they are a high proportion of planets in the galaxy.
Kepler detects changes in stellar brightness due to transiting planets. The closer a planet is to its star, the less precise the alignment has to be for us to observe a transit. Also, the closer it is, the faster it orbits, and the more likely we observe a transit in the limited time we're observing that star. This second factor will become less restrictive as the Kepler mission runs for a longer time. (I presume they need at least two, possibly more, transits before they claim a detection.) Large planets will also give a larger, easier to detect change in brightness.
The other major way of detecting planets is spectroscopically: the planet wobbles the star slightly, and we observe the Doppler shift. This favours massive planets (they wobble the star more) and close planets (they wobble the star faster.)
There have I think also been a few cases where clever interferometry has allowed direct imaging of extrasolar planets. I don't know what the selection effects on this are - further away means easier to separate from the star (good) but less bright (bad.)
All the Planck units involve powers of G, the universal gravitational constant. We have measured G to an accuracy of 10^-4. We want a definition of mass which is accurate to about 3x10^-8. Can you spot the problem?
I'm not at all sure you are serious, but enough people seriously hold this opinion that it is worth responding.
A good system of units needs:
1) Base units which are well defined and independently reconstructible (i.e. a suitably equipped lab can calibrate their equipment purely from the definition of the units.)
2) Logically constructed compound units (e.g. units of force are derived from the units of mass, time and distance.)
3) Logically constructed convenience units (e.g. kilometres for use for distances which would be an inconveniently large number of metres.)
4) To be widely used.
The initial choice of your base units is largely arbitrary - whether it was a from a not-very-accurate measure of a king's foot size or from a not-very-accurate measure of the Earth's circumference. Item (1) can be satisfied equally well (or, in the case of mass, badly) by the metric or imperial systems. The definition of the metre has long since changed from the size of the Earth to quantities measurable in a lab (as has the definition of the foot.)
The SI system (based on metric measures) beats the imperial system hands down on items 2 and 3, and because of this now has a large advantage also on item 4.
Item 2: In Imperial you might measure (heat) energy in BTU and mechanical energy in some mixture of foot-pounds-seconds, but then you need a conversion factor to compare the two. Such conversion factors are never needed in SI.
Item 3: Imperial also messes up the convenience units by having lots of weird conversion factors (e.g. an acre is (I think) a furlong by a chain. How many square feet is that? How many ounces in a ton?*) Metric uses convenience units constructed from base units via consistently named factors of 10 or 1000.
You can't use the current problems with the kilogram as a reason to prefer imperial to metric, as imperial will be just as prone to exactly the same problems. The (UK) Imperial pound is similarly defined by the mass of a unique artifact. In the US, it is defined relative to the kilogram. Mass is the last base unit which doesn't satisfy requirement (1), and the efforts to fix this are what has triggered this entire debate.
One could go a step further, and define your fundamental units in terms of fundamental physical constants (i.e. the Plank mass, Plank time and Plank distance, charge on an electron, etc.) In such a system of units, the speed of light is 1, the formula for the energy of a photon doesn't need a constant in it etc. In practice, we can't use such a system, because we can't measure (in particular) the universal gravitational constant G with sufficient accuracy. Every time we got a better measure of G, our entire system of units would need to be updated. (I.e. with current technology, this system can't satisfy requirement (1) above.)
* And how many different sorts of ounces and tons are there? It is quite a few.
Because it is really hard to measure exactly a mole of hydrogen. It is much easier to measure a mole of a crystalline solid because you know very accurately the density of the atoms. This is precisely what one of the two methods does - trying to define the kilogram as the mass of a precisely sized sphere of pure crystalline silicon. I think silicon was chosen because we know how to make it exceptionally pure, and in a single crystal of sufficient size.
I've met a number of magnetic-sticks-and-balls toys, and Geomag is by far the best (of the ones I've tried.)
You can get really cheap stick-and-ball toys where an icosohedron is a really difficult and delicate thing to make.
The next step up was Magnetix. The magnets were much better than no-brand, and you can get panels (squares, pentagons etc.) with magnets at each vertex. However, the magnets can fall out of the plastic bits (which in turn makes them a swollowing hazard), the stick lengths are not sufficiently accurate, panels are hard to come by, and the magnets still could be better.
Geomag uses much stronger magnets than Magnetix, and built with much better precision. They also do panels in a much better way: there are cheap plastic panels which clip in between the sticks and balls to transform (e.g.) a 5-cycle of sticks and balls into a rigid pentagon. Although they sell sets without panels, to my mind you're missing at least half the point if you don't have them.
For example this is strong enough to pick up. By comparison the much smaller snub dodecahedron is extremely delicate and hard to make with Magnetix.
I do worry about this. While lithium isn't consumed by the batteries, electrifying a substantial portion of the world's road transport would require a huge increase in world wide lithium production.
[... goes away and searches ...]
http://minerals.usgs.gov/minerals/pubs/commodity/lithium/mcs-2010-lithi.pdf
World production was about 25,400 tons in 2008, and reserves about 10,000,000 tons, so we have about 400 years of production at current rates.
If an electric car typically contains a 200kg battery which is 5% lithium by weight, that is 10kg per car. (These numbers are slightly better than guess work, but not by much.) If we manufacture say 200 million new cars per year, that is about 2 million tons of lithium per year, or about 20% of world reserves per year. (200 million cars with 10 year average lifespan would give 2 billion cars worldwide.)
So it does indeed look like it could be a big problem. Eventually there will be a significant contribution from recycling the previous generation of electric cars, but that is some way away. There may also be major reserves yet to be found, as it has not previously been worth looking hard for them.
In any case, even if the reserves exist, we're likely to need on the order of an 80-fold increase in mining capacity if we're to convert the world to battery cars.
At least the market capitalization of HP went down by about ten billion dollars for this departure.
The NZ market is significantly different from the US market in that cable TV has only ever been a very minor player in the market. (I only some suburbs of Wellington, and only in the last 10 years or so, but I haven't tried to keep current on this.) Subscription TV comes by satellite to decoder boxes. This means that currently cable modem is not an option, but I'm guessing that fiber-to-the-home will get used for cable TV service once it is installed.
Even this gets 134 rods to the hogshead.
I'd count dating historical observations of solar eclipses as forensic astronomy, and I think was done well before 30 years ago. Here are some examples.
There are also celestial alignments of pyramids and stone circles - although it would have to be a stellar alignment to count, as the sun doesn't change its path over historical times.
Tor lets you collect your porn anonymously, but at a heavy bandwidth price. The three letter agencies are (we guess) providing Tor nodes with lots of bandwidth so as to be able to sniff the exit traffic.
Result? The NSA is subsidising your anonymous porn collection!
You don't have to care about encryption so long as you don't mind if the NSA has sniffed your porn before you do.
I saw this on the BBC website too, but neither article tells me how it is to the advantage of the hackers to give random people big telephone bills. Do the hackers own some little phone company which the calls are going through? Do they have some overpriced premium number connecting to a computer in Scott Base which recites astrology readings in a synthetic voice?
More seriously: why should the phone OS allow a game to initiate phone calls? (I really hope the answer is 'the OS has a bug' rather than 'that's how they designed it.')
The point is that I just saved several hundred people from having to look it up.
not (as it might have been, for all I knew) a new compiler, for example.
Slashdot summaries have this annoying habbit of assuming that because the author is thoroughly familiar with the technology, everyone else is too and you can just reel of project names with no further explaination.
You should see what it is like in Australia. Here's an extract from one ISP's web page, I think this is typical.
"Monthly usage allowance means combined upload and download data transfer. 1 GB = 1000MB. Unused usage allowance is forfeited each month. Additional usage charged at $0.15/MB and is capped at $300 from the first bill cycle on or after 29th of November 2009."
That's $150/GB for excess. And they short-change us on the GB's they provide in the cap.
"Online chats always display each character as its typed."
This was completely standard 25 years ago, when the closest thing to IM was the Unix 'talk' command.
Having just skimmed the paper, it doesn't look like they account for CO2 removed from the atmosphere by the growing trees. Here's a quick calculation of that:
Assumption 1: all plant matter which does not make its way into paper is burned, or otherwise releases its carbon as CO2, hence is neutral for this analysis. (It could net contribute to greenhouse if it releases as methane instead.)
Assumption 2: paper is 100% cellulose.
Cellulose is a polymer of (C5 H10 O5), which means that it is 4/9 carbon by weight. One unit of carbon burns to produce 11/3 = 3.667 units of CO2. So one unit of paper would burn to produce 44/27 = 1.630 units of CO2, and conversely, 1.63 tonnes of CO2 were removed from the atmosphere to make that paper.
So we're still behind on CO2. And, of course, there are all sorts of other environmental costs.