I know this is splitting hairs, but the mortality rate across all known subvarieties of Ebola is more like 68%, according to various sources, including Wikipedia's article about it, which means Ebola probably isn't even in the top 10 for highest mortality. Number one is probably rabies, where there is no record of anyone having survived, ever, without medical treatment, and once symptoms of the disease have started appearing, even with the best modern medical care, less than a half-dozen people have survived.
We should be demanding progress on that mosquito vaporizing laser demonstrated at TED by Paul Allen's company. It seemed remarkably free of "side effects" and would not put a dent in the overall mosquito population (at least not until the "Star Wars" space based global anti-mosquito laser netwok is set up). They claimed that they could manufacture it (in quantity) for $50.
I had Dengue fever last year and I'd buy one for ten times the price (I know that won't work for the developing world but hey, what can I say? I'm selfish and one less food source available for mosquitos the better for everyone).
Anyone out there know how to get this thing "kickstarted"? How much would Paul Allen ask for the rights?
It's through Nathan Myhrvold's Intellectual Ventures company, with info here and I'm told they're actively looking for people to build the things -- but bear in mind that IV's whole business plan is to come up with great ideas and make money off licensing them, so it might not be cheap.
Seems to, but kids don't hold books as close to their eyes as cellphones and tablets. Kids that read all the time appear to have somewhat more problems with convergence excess -- too easy to cross the eyes, not easy enough to focus on distant objects -- than kids who don't read much. However, it's a parallax issue and the closer you get the worse the problem is. Again, this isn't the case with every kid, by a long shot, but it does tend to exacerbate existing issues. (And again I'm not an expert: I'm just listening to what my wife and her coworkers talk about when they're out in public: "oh, look at that kid: we'll be seeing HIM in our office, no doubt!")
Exactly. Other things to consider: the number has been shrinking over time. Before 1492, the communication bandwidth between 1/8 of the world and the rest of it could be expressed in bits per century, and even now the degrees-of-separation between Australian aborigines and someone living in the slums of Buenos Aires could be more like 20 steps. (Not to mention the 60-some tribes living in Brazil that have never been contacted by Western civilization.)
Plus, the 5-or-6 estimate that Milgram came up with is a crowdsourced number, and like a crowdsourced best-Rubik's-cube solution, is merely the upper bound for the god solution: if you're living in Nebraska and thinking of who might know someone in New York you're going to think of someone who is connected to New York. It's possible your ex-girlfriend, now living in Ireland, is working with the contact's son, so there's actually only 3 degrees of separation, but without omniscience nobody will ever know that and the link will never be found.
The thing that really frustrates me about the Facebook approach is that while it approaches the god solution, because it *is* omniscient about the revealed relationships on Facebook (although it also doesn't guarantee that it captures unknown links) it has a very different view of connections. The connections mapped in Milgram's experiment were people actually known to other people, not necessarily friends, while the connections mapped on Facebook are almost exactly the opposite. They're measuring a whole different sort of network. (I'm "friends" of a sort with Randall Munroe of XKCD on G+. Doesn't mean he has any idea of who I am, or that I really have any idea of him aside from what I see in his comics, and it gives me zero insight into who he'd be friends with if I were trying to get a package from me to some other person.) As such, it seems to me like this research is measuring a different kind of connection, in a different kind of network, using a different kind of definition for connection, but using the same terminology as the original research.
If anyone wants to read a good analysis of the *original* six-degrees-of-separation study, Malcolm Gladwell wrote about it in The New Yorker about ten years ago. (You may wish to skip ahead to part 3.) The researchers -- and this was Stanley Milgram, of the infamous Milgram Experiment involving people's willingness to torture other people -- gave people envelopes addressed to a specific person, and told them to write their names on the envelopes then give them to someone they thought might know the addressee. When all the envelopes came through, they analyzed both the number of hops and the route. (The average was somewhere between 5 and 6 hops, with some being higher. There is no assurance this is the shortest route, but their initial estimates were 100 hops, not five.) The most interesting part was that of the envelopes that reached their destination, more than half came through just three people. It's the discussion of those people, the ones who know people in various different close-knit communities, that matters: they're the connection points.
My wife is a vision therapist. She is seeing lots of kids come in who have trouble reading, playing sports that involve catching, and similar issues. When you're young your brain is really malleable but what it sees is what it expects as normal. When kids' eyes are constantly focussing on small high-resolution screens just a short distance from their faces, their brains consider that normal and adapt to that, meaning the kids have trouble, later on, with easily getting distant images to fuse. It's not like their eyes are crossed, but they do have to work harder, in some cases a lot harder, to maintain distance vision. She gets lots of kids who are considered slow learners or who "just hate reading" and after 12 weeks or so of visual training, suddenly they can catch balls easily, are reading at their grade level, and are enjoying reading.
We don't know its strength per unit weight: a house is easily deformed if you moosh it with a bulldozer, after all. If stuff that only weighs a milligram can support a gram of weight, that's pretty good. That's why mechanical engineers talk about specific strength: the strength (in this case how much stress you can put on it before it permanently deforms) divided by its density.
Add to that, that the geometry of an item has a lot to do with how it reacts to stress. An I-beam holds weight almost as well as a solid block of the same material, but for a tenth the weight. So you might be able to do some really spectacular stuff with this material sandwiched between two layers of carbon fiber, because the resultant chunk of material would act like it was made of solid carbon fiber -- which is to say, ludicrously strong -- but have very little weight. That's a lot harder to predict without messing with the stuff.
1) take a large slab and wrap it in an airtight non-gas permeable membrane. Pump out the air. Voila! You now have a lighter than air structure that doesn't use expensive helium or flammable hydrogen. Let the new age of dirigibles (and floating in mid-air furniture) begin!
While I love the idea, I don't think this stuff will work for that. People have been talking about making vacuum-filled structures for years, and all the numbers I've seen (and I've done these calculations myself) say that using a skin made of the material with the best strength-to-weight ratio we have, no matter how large the internal enclosure (up to at least a couple of kilometers in diameter) the material's tensile strength will be exceeded well before you get the entire structure's density below that of air. It'll crush like a pop can. (Awesome demo: put a liter of water in a 55 gallon steel drum, heat the bottom up until the water boils for a minute, stick the cap back in and tighten down firmly, and give a lecture about vacuum and density. When that drum cools and collapses it makes a sound like nothing I've ever heard, and scares the daylights out of everyone in the lecture hall.) They don't seem to be claiming this stuff has higher tensile strength than carbon fiber, unfortunately. Also hey cool Wikipedia's article on specific strength lists the length of cable that could be made without breaking under its own weight out of various materials, which data I've never seen gathered in one place before.
The lattice is constructed through several steps, Carter said. First, lasers beam ultraviolet light into a reservoir of a resin that forms polymer fibers when the light hits it. The fibers follow the path the light takes, and using multiple beams creates multiple interconnected fibers.
Next, the rest of the resin is washed away, the polymer fibers are coated with a very thin layer of nickel, and the polymer fibers are then dissolved, leaving only the metal lattice.
The dimensions of the lattice can be adjusted by changing the properties of a perforated mask through which the ultraviolet line is beamed, the paper said.
That's difficult. UV lasers are a pain because air absorbs UV somewhat well, and UV-transparent lenses aren't cheap. (A lot of intermolecular bonds have energies in the same range as UV photons, so they're not only opaque but also break apart when hit by UV -- hence DNA damage, for instance.) All the UV lasers I've worked with used fluorine gas, which was always exciting to work with, although there are some very fussy, expensive, low-efficiency frequency-doubling systems that can produce UV.
The other drag is coating extremely convoluted surfaces with a thin film of nickel. Physical vapor deposition, where you just boil the nickel and let it drift in and stick, isn't going to get all the inner recesses, and the outer areas where the vapor hits first is going to be much more heavily plated than inner areas, so they're probably doing chemical vapor deposition or the like -- or some strange nickel tetracarbonyl complex -- to infiltrate the stuff and then crash out the nickel on the surfaces. Totally cool but not stuff you want to play with at home unless you're an expert or enjoy having nickel-plated lungs.
I expect we simultaneously look at word shape, the leading and closing letters, the length, and the middle letters along with some "predictive" matching based on context cues so we can narrow down likely candidate words that "fit" the sentence.
We appear to do a lot more recognition of the first letter than the last: it's extremely easy to think of five words beginning with 'p' but significantly harder to think of five words ending in 'p'. I've seen research saying we're roughly 5 times faster at first than last (but waaaaay faster at last than at second-to-last.)
Rocky Mountain Arsenal, bordering the city limits of Denver, tried disposing of liquid waste by injecting it 12,000 feet below the ground. The result was a series of damaging earthquakes in Denver, up to 5.0 - 5.5 magnitude. USGS wrote a report in 1990.
The Victorian warehouse at 1000 Bannock still shows steel L-braces affixed to the exterior to hold the brick building together from the 1967 earthquake damage -- notice also the long crack running clear through from the back wall diagonally up to the roof.
It wasn't just Denver. I was living in Leadville, CO at the time and some friends had a hobby mine that went into an old fault line. Gold concentrates where there are breaks in the rock because that's where the water moves. When those earthquakes started, the latter third of their mine collapsed (because it was into looser rock adjacent to the fault line). The first time they were like what? and dug it back out and started shoring it, and then the second one hit, and then the third... and they were completely freaked out because the earthquakes were happening on a very regular basis, since the deep well injection dumps were being done on like the third friday of the month, and the earthquakes were happening like half a day later, so they were having earthquakes on the third saturday of each month. Pretty weird. That ended up with them abandoning that mine, although once the Rocky Mountain News started writing articles about the connection between the Arsenal and the earthquakes, they stopped pumping crap down through the water table into the underlying rock.
Trusted foundry is not cheap. It is not feasible to manufacture all electronics used by the federal government in the United States. This has been a well known problem for a long time. Here is an excerpt from a 2005 report.
http://www.acq.osd.mil/dsb/reports/ADA435563.pdf
"Most leading edge wafer production facilities (foundries), with the exception so far of IBM and possibly Texas Instruments, are controlled and located outside the United States. The driving forces behind the “alienation” of foundry business from the United States to other countries include the lower cost of capital available in developing countries, through foreign nations’ tax, market access requirements, subsidized infrastructure and financing incentives (including ownership), and the worldwide portability of technical skills, equipment and process know-how."
I keep reading this, and find it confusing and at odds with my experience. National Semiconductor doesn't/didn't have any fabs in China. Neither does Texas Instruments. Nor Linear Technologies. Nor Analog Devices, nor (as noted) IBM. Even when you move to the likes of Broadcomm, Qualcomm, and LSI, who don't have their own fabs, all their silicon is being made in contract fabs in Taiwan, not People's Republic Of China, and Taiwan contains the three largest foundries in the world (IIRC.) I think the critical phrase in here is "leading edge wafer production facilities": if you need 12 inch wafers with sub-45 micron features, it's possible all the manufacturing capacity is in the PRC. But it's not been my experience that much of production military stuff is using leading-edge electronics: they're using chips that have been run through military and industrial qualifications, and that's not brand-new stuff. Either I'm missing something, or sourcing parts securely isn't as big an issue as it's being made out to be. (Or the problem is actually that they want the cheapest stuff, and aren't willing to pay for parts made through non-PRC fabs.)
What I'm working on was inspired by the sculptures Joshua Harris has been building/leaving in NYC that suddenly explode into full size when a subway goes by and then collapse again to look like a pile of debris. What I've finished: electret microphone amplifier, that tells an arduino when someone's within two meters, the arduino then closes a big monster relay, and I've built a couple of monsters using the instructions on destructables.org for inflating sculptures. (Basically cut up a stuffed animal and blow up the pattern design.) Mine are humanoid shapes about 2 meters tall with arms that stick out and a bunch of tentacle-like things sticking out of their faces.
What I'm working on is the inflation system. My original thought was to use a piece of 6" PVC tubing, about a meter and change long, with about 5 atmospheres of pressure in it, and a lawn sprinkler valve that the arduino triggered, so I have a high pressure lowish volume inflation system. I'm finding that's really loud and not fast enough to get the movement I want, which is to have the monster go from a pile of invisible black rubbish to 2 meters tall in under a second -- really, jumping out at people. So I'm playing with high volume low pressure: having the arduino turn on a shopvac with its exhaust inflating the humanocthulhuoid figure.
Unfortunately I've only got a couple days left to get it all working, and I also have a wedding ring to make before I get married in two weeks, and guess which one is being given higher priority by other involved people? So maybe I'll get lucky and get it done, but most likely it'll be next year. Then I'll have time to add strobe lights on the ground pointing at it.
I've also made most of a soliton gun, using a piece of 15" diameter, 2 meter long cardboard concrete form with a constricted front, the intent being that I can blast big puffs of air from a significant distance and hit people with them: just walking along and suddenly wham a big blast of air from nowhere. It might be interesting, especially if I can time it so it hits people at the same moment as the jumpy monster jumps.
I suspect loss of people to cancer alone has a productivity cost of greater than $157,000/person, not to mention the money saved in treatment of cancer through surgery and radiation/chemotherapy.
why don't some of you idiots understand this? why do you persist in this complete insanity that an unregulated marketplace is somehow fair and equal and somehow it is the government screws it up? the government is the only tool we have to keep it regulated, policed, and therefore fair, where the large are prevented from using their entrenched position to cheat off the backs of the small
I think they understand that perfectly, and you're the one who is missing the plan. A couple years back there was a Dilbert strip, in which Dogbert was campaigning for strict gun laws and anti-weapons laws, because if nobody else had any weapons he could take over the world armed with nothing more than a butter knife. People who think they're entirely self-made, that they're where they are purely because of their internal drive for success, are going to favor a regulatory environment that they think will allow them to become the large of whom you speak. They *want* an environment that allows and encourages wealth and power into the hands of a few, precisely because they think they're the few. It's a completely consistent philosophy. Given their premises, I think their conclusions, and the policies they favor, are quite rational. I think their premises are completely wrong, but I don't think their worldview is wide enough for them to be able to analyze their premises and understand why I think they're wrong.
>Throwing money at solar and wind is well and good, but it's a luxury that a rich country ('rich' being relative these days) like the United States can afford; it's a joke to imagine that India or Indonesia or China are going to handicap their economies when they've only just lately (to varying degrees) got round to having economies in the first place.
For the record, I am involved with designing LED drivers for street lights. About 70% of our revenue is coming from sales in southeast Asia, because they can't afford to build new power plants at the rate they need so instead they're buying lighting that uses less power. As you say, that's not pursuing green energy for its own sake, but the result is the same (and very gratifying to us.)
Totally agree with your reasoning, and add: square is nearly the only shape that, given a lip so it can't drop into the hole, can be turned such that it will. If "can't drop into the hole" were the only reason, you could go with triangles or anything with more than about 5 sides, and the major diameter - minor diameter width would be exceeded by the lip so they wouldn't either. In other words, not only does history roll its eyes at this question, so does basic geometry. I've been asked this question in an interview and got a surprisingly positive reception when I gave my "this is a stupid question and here's why" response, so keep it in your bag of interview tricks.
Since you seem to know something about it, maybe you could answer this: why is this suddenly such a big deal? I thought the original article was a big deal because he'd made synthetic aperture radar out of stuff he picked up on ebay -- it was a really amazing DIY hack, in other words. But militaries have had SAR for decades, right? So how is what he's doing different than what's been available for many years? Obviously it is, because this is making news all over the place, but I feel like I'm missing some important part of the research/development.
Small warm-blooded animals have a tough time keeping warm, particularly in rain. Some people who study hibernation have theorized that wings and feathers both came from the need to have something like a rainjacket, that could deflect rain, but could also be opened up to vent excess heat during exercise, based on the huge primary feathers of many waterfowl, that cover their whole backs and sides.
FWIW we've printed our own LEGO bricks on a Stratasys and they've worked pretty well. We needed a rapid mechanical prototyping system and LEGO bricks are perfect, so we built some custom baseplates that are basically LEGO-covered c-clamps and they've done an awesome job.
When I was a kid I lived in a town that had the first LEGO manufacturing facility outside Denmark (brief franchise with Samsonite) and I knew some of the people doing the plastic injection molding there. They said the LEGO manufacturing tolerances were higher than American companies making medical equipment: it took forever to meet LEGO specs even though they were using the LEGO molds.
Plug, wire, etc. maintenance is still easy (except in designs that crowd the engine bay). It's a design that's known to work fairly reliably, and when it doesn't, it can be fixed. Try replacing the engine in something like a Subaru sometime... or even getting at the plugs.
On the other hand, I'd love to see more inline engines. They have a lot of the same benefits.
Interesting you say that. The last time I did an engine swap on my Subaru Legacy, it did take *quite* a while, but that's because it's a 4WD, and the only reason it was more difficult than when I pulled the engine out of my Jeep is because the Subaru engine takes up a lot more of the engine compartment than that inline six in the Jeep. The engine was also much easier to deal with because it was so much lighter than the Jeep engine/transmission/transfer case. It was also much easier than pulling the engine in the 1967 Oldsmobile, which has about the same amount of engine compartment clearance as the Subaru, simply because it was so much lighter.
And I'm really surprised by the claim about sparkplugs. It takes me like 30 minutes to change the plugs on the Soob -- a little longer than on the inline six in the Jeep, because the Jeep plugs are so easy to access, but *far* easier than changing the 8 plugs on the Olds, because I have to loosen and tilt the power brake booster to get to the furtherest-back sparkplug on the Olds.
I'd far rather work on Subarus with their nice little flat boxer engines. They fit so nicely in the hood space. Everything's easily accessible.
Now where that doesn't work is, say, drugs.... It takes... like a billion dollars to get a new drug on the market, so you either need to find something that is going to sell a billion dollars in six months or you need long-term patent protection.
I wonder if that's true. I know that existing pharmaceutical companies that operate in the existing patent environment spend billions of dollars looking for new drugs and so, to sustain themselves, they need long-term patent protection.... I wonder, though, what sorts of drugs would be invented if the patent system were overhauled to shorten the monopoly? Maybe the next multi-billion-dollar penis pill for 70-year-old men won't be invented and, instead, we'll cure something. It'd be nice if it worked out that way, huh?
Ian
The problem for drug companies is that like 1% of the cost of getting a drug to market is actually doing the chemistry to produce it -- and that's the part that someone else can copy, avoiding the other 99% of the cost. The reason for that is because something like 20% of the get-drug-to-market cost is advertising (I'm not claiming that's a positive thing) and the lion's share of it is paying off A: all the stuff that didn't work and B: all the testing and clinical trials involving thousands of patients and hundreds of doctors and statisticians to prove to the FDA that it is 1: effective and 2: not horribly dangerous. Pharmaceutical companies produce thousands of potential drugs, of which dozens get as far as actual trials in animals, of which a very small handful get to clinical trials in humans, of which maybe one actually goes to the market, and that's a *huge* amount of money spent, all of which has to be recovered with the one that got to market.
Now before everyone goes awwww poor pharmaceutical companies, this isn't a bad thing from their viewpoint, because having an extremely expensive certification process means small players can't get in the game. This is a gatekeeping function that the pharmaceutical companies have had a hand in putting in place, because only a multibillion dollar company can afford a billion dollar certification process. However, one side-effect is that they need patent protection to pay for their gate. I'm not saying that's a good idea, but that's the way it is, and if we're going to keep getting new & improved drugs, slashing patent protection alone is not going to work (and trying to fix two separate major regulatory problems is roughly four times as hard as just trying to fix one, I suspect.)
Both time figures were somewhat unrealistic, but the point is copying something is usually a lot cheaper and quicker than developing something from scratch. Certainly I think something could be ripped off after release long before an inventor would see his profit.
That, and some "simple" things come out of long periods of trying to solve a problem. The end solution might be a simple widget, but coming up with that widget as a solution to the problem may have involved significant resources. If someone can then just start producting that widget with no money going to the people who came up with it, you'll see people a lot less willing to spend money inventing.
I'm not sure this answers your question: I work in an area -- integrated circuit design for consumer markets -- where we can patent things until we're blue in the face, and someone else can still come up with another implementation that does the same thing we do, and get it to market. (Despite what so many people claim, the US patent office *does* reject things for being too obvious, like, oh, say, building an LED driver that can work with wall dimmers, so if you want to build those you have to build a specific implementation of a dimmer driver, and patent that, and then someone else will just find another clever way of doing it.) So we do the research, build a product, and everyone else copies it. But we still make money, and the way we do that is by choosing markets where we know it'll take long enough for the copies to hit the market that we'll have already repaid our investment and made a profit. I'm not saying that's the best way of doing things, but one nice side-benefit is that instead of just one solution to the problem, there are a half-dozen, and that's a huge amount of information being dumped into the public pool for future inventors to use. We almost don't need patents at all, because they don't actually help us -- but they do help the world, by requiring people to publish their implementations to problems. We keep investing on innovation because we make money, and the world gets a lot of value from that, as a very nice side-effect.
Now where that doesn't work is, say, drugs. It costs us like I dunno let's say a million dollars to bring a new IC to market. We can recoup that in six months. It takes more like a billion dollars to get a new drug on the market, so you either need to find something that is going to sell a billion dollars in six months or you need long-term patent protection. Maybe that means we should have variable patent lengths, maybe patents should expire in 5 years unless you reapply for continued protection. I don't know. But patents have some obvious and huge benefits over trade secrets. We have trade secrets we've kept for 30 years, and some of them appear to me to be exactly the sort of stuff that'd really help a lot of other designers. I don't think that's a great idea either.
I'll ask around and see what I can find. Work has a lab to specifically measure human eye sensitivity across the spectrum, but unfortunately it's in Illinois. However, I'm wondering if you couldn't build a DIY spectrophotometer using a DVD as a diffraction grating, and see what your sensitivity/wavelength/intensity graph would be.
Slashdot Mirror
I know this is splitting hairs, but the mortality rate across all known subvarieties of Ebola is more like 68%, according to various sources, including Wikipedia's article about it, which means Ebola probably isn't even in the top 10 for highest mortality. Number one is probably rabies, where there is no record of anyone having survived, ever, without medical treatment, and once symptoms of the disease have started appearing, even with the best modern medical care, less than a half-dozen people have survived.
We should be demanding progress on that mosquito vaporizing laser demonstrated at TED by Paul Allen's company. It seemed remarkably free of "side effects" and would not put a dent in the overall mosquito population (at least not until the "Star Wars" space based global anti-mosquito laser netwok is set up). They claimed that they could manufacture it (in quantity) for $50.
I had Dengue fever last year and I'd buy one for ten times the price (I know that won't work for the developing world but hey, what can I say? I'm selfish and one less food source available for mosquitos the better for everyone).
Anyone out there know how to get this thing "kickstarted"? How much would Paul Allen ask for the rights?
It's through Nathan Myhrvold's Intellectual Ventures company, with info here and I'm told they're actively looking for people to build the things -- but bear in mind that IV's whole business plan is to come up with great ideas and make money off licensing them, so it might not be cheap.
Seems to, but kids don't hold books as close to their eyes as cellphones and tablets. Kids that read all the time appear to have somewhat more problems with convergence excess -- too easy to cross the eyes, not easy enough to focus on distant objects -- than kids who don't read much. However, it's a parallax issue and the closer you get the worse the problem is. Again, this isn't the case with every kid, by a long shot, but it does tend to exacerbate existing issues. (And again I'm not an expert: I'm just listening to what my wife and her coworkers talk about when they're out in public: "oh, look at that kid: we'll be seeing HIM in our office, no doubt!")
Exactly. Other things to consider: the number has been shrinking over time. Before 1492, the communication bandwidth between 1/8 of the world and the rest of it could be expressed in bits per century, and even now the degrees-of-separation between Australian aborigines and someone living in the slums of Buenos Aires could be more like 20 steps. (Not to mention the 60-some tribes living in Brazil that have never been contacted by Western civilization.)
Plus, the 5-or-6 estimate that Milgram came up with is a crowdsourced number, and like a crowdsourced best-Rubik's-cube solution, is merely the upper bound for the god solution: if you're living in Nebraska and thinking of who might know someone in New York you're going to think of someone who is connected to New York. It's possible your ex-girlfriend, now living in Ireland, is working with the contact's son, so there's actually only 3 degrees of separation, but without omniscience nobody will ever know that and the link will never be found.
The thing that really frustrates me about the Facebook approach is that while it approaches the god solution, because it *is* omniscient about the revealed relationships on Facebook (although it also doesn't guarantee that it captures unknown links) it has a very different view of connections. The connections mapped in Milgram's experiment were people actually known to other people, not necessarily friends, while the connections mapped on Facebook are almost exactly the opposite. They're measuring a whole different sort of network. (I'm "friends" of a sort with Randall Munroe of XKCD on G+. Doesn't mean he has any idea of who I am, or that I really have any idea of him aside from what I see in his comics, and it gives me zero insight into who he'd be friends with if I were trying to get a package from me to some other person.) As such, it seems to me like this research is measuring a different kind of connection, in a different kind of network, using a different kind of definition for connection, but using the same terminology as the original research.
If anyone wants to read a good analysis of the *original* six-degrees-of-separation study, Malcolm Gladwell wrote about it in The New Yorker about ten years ago. (You may wish to skip ahead to part 3.) The researchers -- and this was Stanley Milgram, of the infamous Milgram Experiment involving people's willingness to torture other people -- gave people envelopes addressed to a specific person, and told them to write their names on the envelopes then give them to someone they thought might know the addressee. When all the envelopes came through, they analyzed both the number of hops and the route. (The average was somewhere between 5 and 6 hops, with some being higher. There is no assurance this is the shortest route, but their initial estimates were 100 hops, not five.) The most interesting part was that of the envelopes that reached their destination, more than half came through just three people. It's the discussion of those people, the ones who know people in various different close-knit communities, that matters: they're the connection points.
My wife is a vision therapist. She is seeing lots of kids come in who have trouble reading, playing sports that involve catching, and similar issues. When you're young your brain is really malleable but what it sees is what it expects as normal. When kids' eyes are constantly focussing on small high-resolution screens just a short distance from their faces, their brains consider that normal and adapt to that, meaning the kids have trouble, later on, with easily getting distant images to fuse. It's not like their eyes are crossed, but they do have to work harder, in some cases a lot harder, to maintain distance vision. She gets lots of kids who are considered slow learners or who "just hate reading" and after 12 weeks or so of visual training, suddenly they can catch balls easily, are reading at their grade level, and are enjoying reading.
We don't know its strength per unit weight: a house is easily deformed if you moosh it with a bulldozer, after all. If stuff that only weighs a milligram can support a gram of weight, that's pretty good. That's why mechanical engineers talk about specific strength: the strength (in this case how much stress you can put on it before it permanently deforms) divided by its density.
Add to that, that the geometry of an item has a lot to do with how it reacts to stress. An I-beam holds weight almost as well as a solid block of the same material, but for a tenth the weight. So you might be able to do some really spectacular stuff with this material sandwiched between two layers of carbon fiber, because the resultant chunk of material would act like it was made of solid carbon fiber -- which is to say, ludicrously strong -- but have very little weight. That's a lot harder to predict without messing with the stuff.
1) take a large slab and wrap it in an airtight non-gas permeable membrane. Pump out the air. Voila! You now have a lighter than air structure that doesn't use expensive helium or flammable hydrogen. Let the new age of dirigibles (and floating in mid-air furniture) begin!
While I love the idea, I don't think this stuff will work for that. People have been talking about making vacuum-filled structures for years, and all the numbers I've seen (and I've done these calculations myself) say that using a skin made of the material with the best strength-to-weight ratio we have, no matter how large the internal enclosure (up to at least a couple of kilometers in diameter) the material's tensile strength will be exceeded well before you get the entire structure's density below that of air. It'll crush like a pop can. (Awesome demo: put a liter of water in a 55 gallon steel drum, heat the bottom up until the water boils for a minute, stick the cap back in and tighten down firmly, and give a lecture about vacuum and density. When that drum cools and collapses it makes a sound like nothing I've ever heard, and scares the daylights out of everyone in the lecture hall.) They don't seem to be claiming this stuff has higher tensile strength than carbon fiber, unfortunately. Also hey cool Wikipedia's article on specific strength lists the length of cable that could be made without breaking under its own weight out of various materials, which data I've never seen gathered in one place before.
Someone linked a cnet article with more information including how it is produced.
From reading it it sounds like it will be easier to produce, but I really don't know a damn thing on this subject.
What's your take?
http://news.cnet.com/8301-30685_3-57327382-264/breakthrough-material-is-barely-more-than-air/?part=rss&subj=crave&tag=title
The lattice is constructed through several steps, Carter said. First, lasers beam ultraviolet light into a reservoir of a resin that forms polymer fibers when the light hits it. The fibers follow the path the light takes, and using multiple beams creates multiple interconnected fibers.
Next, the rest of the resin is washed away, the polymer fibers are coated with a very thin layer of nickel, and the polymer fibers are then dissolved, leaving only the metal lattice.
The dimensions of the lattice can be adjusted by changing the properties of a perforated mask through which the ultraviolet line is beamed, the paper said.
That's difficult. UV lasers are a pain because air absorbs UV somewhat well, and UV-transparent lenses aren't cheap. (A lot of intermolecular bonds have energies in the same range as UV photons, so they're not only opaque but also break apart when hit by UV -- hence DNA damage, for instance.) All the UV lasers I've worked with used fluorine gas, which was always exciting to work with, although there are some very fussy, expensive, low-efficiency frequency-doubling systems that can produce UV.
The other drag is coating extremely convoluted surfaces with a thin film of nickel. Physical vapor deposition, where you just boil the nickel and let it drift in and stick, isn't going to get all the inner recesses, and the outer areas where the vapor hits first is going to be much more heavily plated than inner areas, so they're probably doing chemical vapor deposition or the like -- or some strange nickel tetracarbonyl complex -- to infiltrate the stuff and then crash out the nickel on the surfaces. Totally cool but not stuff you want to play with at home unless you're an expert or enjoy having nickel-plated lungs.
I expect we simultaneously look at word shape, the leading and closing letters, the length, and the middle letters along with some "predictive" matching based on context cues so we can narrow down likely candidate words that "fit" the sentence.
We appear to do a lot more recognition of the first letter than the last: it's extremely easy to think of five words beginning with 'p' but significantly harder to think of five words ending in 'p'. I've seen research saying we're roughly 5 times faster at first than last (but waaaaay faster at last than at second-to-last.)
Rocky Mountain Arsenal, bordering the city limits of Denver, tried disposing of liquid waste by injecting it 12,000 feet below the ground. The result was a series of damaging earthquakes in Denver, up to 5.0 - 5.5 magnitude. USGS wrote a report in 1990.
The Victorian warehouse at 1000 Bannock still shows steel L-braces affixed to the exterior to hold the brick building together from the 1967 earthquake damage -- notice also the long crack running clear through from the back wall diagonally up to the roof.
It wasn't just Denver. I was living in Leadville, CO at the time and some friends had a hobby mine that went into an old fault line. Gold concentrates where there are breaks in the rock because that's where the water moves. When those earthquakes started, the latter third of their mine collapsed (because it was into looser rock adjacent to the fault line). The first time they were like what? and dug it back out and started shoring it, and then the second one hit, and then the third... and they were completely freaked out because the earthquakes were happening on a very regular basis, since the deep well injection dumps were being done on like the third friday of the month, and the earthquakes were happening like half a day later, so they were having earthquakes on the third saturday of each month. Pretty weird. That ended up with them abandoning that mine, although once the Rocky Mountain News started writing articles about the connection between the Arsenal and the earthquakes, they stopped pumping crap down through the water table into the underlying rock.
...an uninformed, knee-jerk comment.
Trusted foundry is not cheap. It is not feasible to manufacture all electronics used by the federal government in the United States. This has been a well known problem for a long time. Here is an excerpt from a 2005 report.
http://www.acq.osd.mil/dsb/reports/ADA435563.pdf
"Most leading edge wafer production facilities (foundries), with the exception so far of IBM and possibly Texas Instruments, are controlled and located outside the United States. The driving forces behind the “alienation” of foundry business from the United States to other countries include the lower cost of capital available in developing countries, through foreign nations’ tax, market access requirements, subsidized infrastructure and financing incentives (including ownership), and the worldwide portability of technical skills, equipment and process know-how."
I keep reading this, and find it confusing and at odds with my experience. National Semiconductor doesn't/didn't have any fabs in China. Neither does Texas Instruments. Nor Linear Technologies. Nor Analog Devices, nor (as noted) IBM. Even when you move to the likes of Broadcomm, Qualcomm, and LSI, who don't have their own fabs, all their silicon is being made in contract fabs in Taiwan, not People's Republic Of China, and Taiwan contains the three largest foundries in the world (IIRC.) I think the critical phrase in here is "leading edge wafer production facilities": if you need 12 inch wafers with sub-45 micron features, it's possible all the manufacturing capacity is in the PRC. But it's not been my experience that much of production military stuff is using leading-edge electronics: they're using chips that have been run through military and industrial qualifications, and that's not brand-new stuff. Either I'm missing something, or sourcing parts securely isn't as big an issue as it's being made out to be. (Or the problem is actually that they want the cheapest stuff, and aren't willing to pay for parts made through non-PRC fabs.)
What I'm working on was inspired by the sculptures Joshua Harris has been building/leaving in NYC that suddenly explode into full size when a subway goes by and then collapse again to look like a pile of debris. What I've finished: electret microphone amplifier, that tells an arduino when someone's within two meters, the arduino then closes a big monster relay, and I've built a couple of monsters using the instructions on destructables.org for inflating sculptures. (Basically cut up a stuffed animal and blow up the pattern design.) Mine are humanoid shapes about 2 meters tall with arms that stick out and a bunch of tentacle-like things sticking out of their faces.
What I'm working on is the inflation system. My original thought was to use a piece of 6" PVC tubing, about a meter and change long, with about 5 atmospheres of pressure in it, and a lawn sprinkler valve that the arduino triggered, so I have a high pressure lowish volume inflation system. I'm finding that's really loud and not fast enough to get the movement I want, which is to have the monster go from a pile of invisible black rubbish to 2 meters tall in under a second -- really, jumping out at people. So I'm playing with high volume low pressure: having the arduino turn on a shopvac with its exhaust inflating the humanocthulhuoid figure.
Unfortunately I've only got a couple days left to get it all working, and I also have a wedding ring to make before I get married in two weeks, and guess which one is being given higher priority by other involved people? So maybe I'll get lucky and get it done, but most likely it'll be next year. Then I'll have time to add strobe lights on the ground pointing at it.
I've also made most of a soliton gun, using a piece of 15" diameter, 2 meter long cardboard concrete form with a constricted front, the intent being that I can blast big puffs of air from a significant distance and hit people with them: just walking along and suddenly wham a big blast of air from nowhere. It might be interesting, especially if I can time it so it hits people at the same moment as the jumpy monster jumps.
I suspect loss of people to cancer alone has a productivity cost of greater than $157,000/person, not to mention the money saved in treatment of cancer through surgery and radiation/chemotherapy.
why don't some of you idiots understand this? why do you persist in this complete insanity that an unregulated marketplace is somehow fair and equal and somehow it is the government screws it up? the government is the only tool we have to keep it regulated, policed, and therefore fair, where the large are prevented from using their entrenched position to cheat off the backs of the small
I think they understand that perfectly, and you're the one who is missing the plan. A couple years back there was a Dilbert strip, in which Dogbert was campaigning for strict gun laws and anti-weapons laws, because if nobody else had any weapons he could take over the world armed with nothing more than a butter knife. People who think they're entirely self-made, that they're where they are purely because of their internal drive for success, are going to favor a regulatory environment that they think will allow them to become the large of whom you speak. They *want* an environment that allows and encourages wealth and power into the hands of a few, precisely because they think they're the few. It's a completely consistent philosophy. Given their premises, I think their conclusions, and the policies they favor, are quite rational. I think their premises are completely wrong, but I don't think their worldview is wide enough for them to be able to analyze their premises and understand why I think they're wrong.
>Throwing money at solar and wind is well and good, but it's a luxury that a rich country ('rich' being relative these days) like the United States can afford; it's a joke to imagine that India or Indonesia or China are going to handicap their economies when they've only just lately (to varying degrees) got round to having economies in the first place.
For the record, I am involved with designing LED drivers for street lights. About 70% of our revenue is coming from sales in southeast Asia, because they can't afford to build new power plants at the rate they need so instead they're buying lighting that uses less power. As you say, that's not pursuing green energy for its own sake, but the result is the same (and very gratifying to us.)
Totally agree with your reasoning, and add: square is nearly the only shape that, given a lip so it can't drop into the hole, can be turned such that it will. If "can't drop into the hole" were the only reason, you could go with triangles or anything with more than about 5 sides, and the major diameter - minor diameter width would be exceeded by the lip so they wouldn't either. In other words, not only does history roll its eyes at this question, so does basic geometry. I've been asked this question in an interview and got a surprisingly positive reception when I gave my "this is a stupid question and here's why" response, so keep it in your bag of interview tricks.
Since you seem to know something about it, maybe you could answer this: why is this suddenly such a big deal? I thought the original article was a big deal because he'd made synthetic aperture radar out of stuff he picked up on ebay -- it was a really amazing DIY hack, in other words. But militaries have had SAR for decades, right? So how is what he's doing different than what's been available for many years? Obviously it is, because this is making news all over the place, but I feel like I'm missing some important part of the research/development.
Small warm-blooded animals have a tough time keeping warm, particularly in rain. Some people who study hibernation have theorized that wings and feathers both came from the need to have something like a rainjacket, that could deflect rain, but could also be opened up to vent excess heat during exercise, based on the huge primary feathers of many waterfowl, that cover their whole backs and sides.
FWIW we've printed our own LEGO bricks on a Stratasys and they've worked pretty well. We needed a rapid mechanical prototyping system and LEGO bricks are perfect, so we built some custom baseplates that are basically LEGO-covered c-clamps and they've done an awesome job.
When I was a kid I lived in a town that had the first LEGO manufacturing facility outside Denmark (brief franchise with Samsonite) and I knew some of the people doing the plastic injection molding there. They said the LEGO manufacturing tolerances were higher than American companies making medical equipment: it took forever to meet LEGO specs even though they were using the LEGO molds.
Plug, wire, etc. maintenance is still easy (except in designs that crowd the engine bay). It's a design that's known to work fairly reliably, and when it doesn't, it can be fixed. Try replacing the engine in something like a Subaru sometime... or even getting at the plugs.
On the other hand, I'd love to see more inline engines. They have a lot of the same benefits.
Interesting you say that. The last time I did an engine swap on my Subaru Legacy, it did take *quite* a while, but that's because it's a 4WD, and the only reason it was more difficult than when I pulled the engine out of my Jeep is because the Subaru engine takes up a lot more of the engine compartment than that inline six in the Jeep. The engine was also much easier to deal with because it was so much lighter than the Jeep engine/transmission/transfer case. It was also much easier than pulling the engine in the 1967 Oldsmobile, which has about the same amount of engine compartment clearance as the Subaru, simply because it was so much lighter.
And I'm really surprised by the claim about sparkplugs. It takes me like 30 minutes to change the plugs on the Soob -- a little longer than on the inline six in the Jeep, because the Jeep plugs are so easy to access, but *far* easier than changing the 8 plugs on the Olds, because I have to loosen and tilt the power brake booster to get to the furtherest-back sparkplug on the Olds.
I'd far rather work on Subarus with their nice little flat boxer engines. They fit so nicely in the hood space. Everything's easily accessible.
I wonder if that's true. I know that existing pharmaceutical companies that operate in the existing patent environment spend billions of dollars looking for new drugs and so, to sustain themselves, they need long-term patent protection. ... I wonder, though, what sorts of drugs would be invented if the patent system were overhauled to shorten the monopoly? Maybe the next multi-billion-dollar penis pill for 70-year-old men won't be invented and, instead, we'll cure something. It'd be nice if it worked out that way, huh?
Ian
The problem for drug companies is that like 1% of the cost of getting a drug to market is actually doing the chemistry to produce it -- and that's the part that someone else can copy, avoiding the other 99% of the cost. The reason for that is because something like 20% of the get-drug-to-market cost is advertising (I'm not claiming that's a positive thing) and the lion's share of it is paying off A: all the stuff that didn't work and B: all the testing and clinical trials involving thousands of patients and hundreds of doctors and statisticians to prove to the FDA that it is 1: effective and 2: not horribly dangerous. Pharmaceutical companies produce thousands of potential drugs, of which dozens get as far as actual trials in animals, of which a very small handful get to clinical trials in humans, of which maybe one actually goes to the market, and that's a *huge* amount of money spent, all of which has to be recovered with the one that got to market.
Now before everyone goes awwww poor pharmaceutical companies, this isn't a bad thing from their viewpoint, because having an extremely expensive certification process means small players can't get in the game. This is a gatekeeping function that the pharmaceutical companies have had a hand in putting in place, because only a multibillion dollar company can afford a billion dollar certification process. However, one side-effect is that they need patent protection to pay for their gate. I'm not saying that's a good idea, but that's the way it is, and if we're going to keep getting new & improved drugs, slashing patent protection alone is not going to work (and trying to fix two separate major regulatory problems is roughly four times as hard as just trying to fix one, I suspect.)
Both time figures were somewhat unrealistic, but the point is copying something is usually a lot cheaper and quicker than developing something from scratch. Certainly I think something could be ripped off after release long before an inventor would see his profit.
That, and some "simple" things come out of long periods of trying to solve a problem. The end solution might be a simple widget, but coming up with that widget as a solution to the problem may have involved significant resources. If someone can then just start producting that widget with no money going to the people who came up with it, you'll see people a lot less willing to spend money inventing.
I'm not sure this answers your question: I work in an area -- integrated circuit design for consumer markets -- where we can patent things until we're blue in the face, and someone else can still come up with another implementation that does the same thing we do, and get it to market. (Despite what so many people claim, the US patent office *does* reject things for being too obvious, like, oh, say, building an LED driver that can work with wall dimmers, so if you want to build those you have to build a specific implementation of a dimmer driver, and patent that, and then someone else will just find another clever way of doing it.) So we do the research, build a product, and everyone else copies it. But we still make money, and the way we do that is by choosing markets where we know it'll take long enough for the copies to hit the market that we'll have already repaid our investment and made a profit. I'm not saying that's the best way of doing things, but one nice side-benefit is that instead of just one solution to the problem, there are a half-dozen, and that's a huge amount of information being dumped into the public pool for future inventors to use. We almost don't need patents at all, because they don't actually help us -- but they do help the world, by requiring people to publish their implementations to problems. We keep investing on innovation because we make money, and the world gets a lot of value from that, as a very nice side-effect.
Now where that doesn't work is, say, drugs. It costs us like I dunno let's say a million dollars to bring a new IC to market. We can recoup that in six months. It takes more like a billion dollars to get a new drug on the market, so you either need to find something that is going to sell a billion dollars in six months or you need long-term patent protection. Maybe that means we should have variable patent lengths, maybe patents should expire in 5 years unless you reapply for continued protection. I don't know. But patents have some obvious and huge benefits over trade secrets. We have trade secrets we've kept for 30 years, and some of them appear to me to be exactly the sort of stuff that'd really help a lot of other designers. I don't think that's a great idea either.
I'll ask around and see what I can find. Work has a lab to specifically measure human eye sensitivity across the spectrum, but unfortunately it's in Illinois. However, I'm wondering if you couldn't build a DIY spectrophotometer using a DVD as a diffraction grating, and see what your sensitivity/wavelength/intensity graph would be.
One o/s...monitoring...and controlling....every sensor in the city?
One o/s to rule them all
One o/s to find them
One o/s to bring them all
And in the darkness bind them...