Right now I have six cables under my desk, for power. If there was no USB, I'd have two more. If there was no WiFi, another two cables. If there was no Bluetooth, I'd have another two cables hanging around. And I don't have an iPod, yet.
I don't mind having a wire around the desk, it's having a dozen that I mind. Each cable that I can get rid off helps. This thing would decrease the number of cables on, around, and under my desk by more than any other wireless tech did.
And yes, the minor annoyance of having to pick up a cable from the floor, route it around the other cables and the various things on my desk (yes, I need them, and no, I don't have room for a bigger desk), find which of the seven holes in the back of my laptop fits the plug, several times every day of my life, and the same for the phone, and spending a half hour re-arranging the cables every week, is something I'd pay a decent amount of money to go without.
You're mostly right about the LCD panels -- they can't fix this without head-tracking. Head-tracking would work very well, I think, with something like the Wii's sensor on the glasses, but only for a single viewer.
However, for cinema I think it's less important, as the scenes are generally "far away" and the viewers "not very mobile", thus the brain expects very little parallax change.
Physics is only about creating a model for how the universe works: you put numbers in, you get numbers out. What happens when we aren't looking (putting numbers in but not looking at the numbers coming out) has no real relevance and is unverifiable.
Wait, I have a question about this. If what happens when we're not looking is unverifiable, then what are these guys doing in their experiment? I mean, they claim their results tell something about what happens (or doesn't) when we're not looking. So that is at least sometimes verifiable, or they're very clearly talking out of their asses. No physicist is right all the time, but I'd be surprised by such complete failure.
Actually, I think there's a language issue here. Obviously you can't tell if your prediction is true without looking at the results, I think they're doing something a bit more subtle here. (Like, for instance, looking less often than before, and noticing the results are different at the end, despite the fact that just looking shouldn't have changed the results that much.)
Yes, that's what I was curious about. I won't bet a cent on the security properties of this thing, but I'm very curious how well it might work in keeping out the wireless signals of the neighbors. I never get fewer that three foreign access points around the house, and I don't even live in a very dense apartment block. I'd need a repeater within certain frequencies for cell-phone and other out-of-the-house wireless I might want. (This would be double-nice: since cell phones communicate through the repeater/amplifier, they would need lower transmission power!)
Also, is the paint reflective or absorbent? In the former case, it might even boost my signal a bit inside the house, provided that I'm careful to use it only on external walls.
And how much does this thing cost? Since they're marketing it as a snake oil derivative, I guess it's too much to be worth it.
It's a bummer that all comments are about the bogus security instead of the nice signal-shaping possibilities.
Re:Gentoo has a good kernel hardening guide
on
SELinux by Example
·
· Score: 1
I think the definition is more like "unreasonable belief that someone is trying to get you". For a simple example, just because a terrorist intends to, say, "kill any and all Americans" doesn't mean that all paranoids in America become instantly sane (or, at least, not paranoid anymore).
Because birds can't transport tons (literally) of materials and people over thousands of kilometers in a few hours. For that matter, planes can't land on tree branches.
If we wanted an artifact that can do what birds can (e.g. land on a tree branch) we probably would start by building a mechanical bird. It's quite hard, and I doubt we can do that today. We can't build a mechanical brain (i.e. an artifact that can do what a brain does), either.
I'm sure it's _very_ nontrivial---which is why it's not done often---but lots of what a parser does could be done in parallel. For instance, you really don't _need_ to find whitespace linearly. You can split the string (for large programs) in pieces. Stitching the results is going to be a bitch, but it's possible. For a million processors you _could_ actually load one character on each processor and do everything in parallel. I never saw an algorithm that does this on text, but I've seen pattern recognition and edge detection done with a-processor-per-pixel algorithms, and if you think about it properly parsing is something analogue: you have to find the tree hidden in that mess of characters.
Again, this would be a bitch to program, perhaps only possible by automatic tools, but it could work.
I simply meant that free markets really do solve a vast set of economic problems, but not _all_ of them, and not _all_ problems we need to solve are purely economic. For instance, there are social, moral and ethic problems. These are intertwined with economic issues (with perhaps some extraordinary exceptions), but this means that the solution is not always purely an economic one (again, with some possible exceptions).
I didn't think I had to mention this in my post, because it actually dealt with such issues: the fact that people tend to think of medicine and doctors more from the point of view of ethics, morality, trust --- and often very irrationally. The reason my post was so economy-oriented was just because I wanted to show why people should think of this (also) as an economic problem.
You did it much more incisively and tersely than I was capable.
(Though I don't agree fully with you: (1) I don't think the rising cost of health care is always fully caused by increases in quality, and even if it were it might not necessarily be acceptable. There's no reason a better treatment can't be cheaper, at least sometimes. (2) I have only little formal training in economy, but as far as I understand it pure laissez-fare doesn't solve everything; unless I misunderstand the terms, it allows e.g. monopolies. I think there needs to be some equilibrium between the size and number of agents before free markets work "well". (3) People often tend to argue for "free health care", as you said. But that's not the only meme, there's also "some people simply can't pay; what do we do about it?" I don't mean I support a particular answer, that's for another discussion. But free markets give a precise answer --- we don't do anything, let them deal with it --- which may not be a "good answer".)
This is very interesting. To my (admittedly very limited) knowledge of things economical, a "very cooperative society" as you describe translates to a communist country. Which (I'm sure it's been said before) always sounds like a very good idea, but was never implemented acceptably _on a large scale_. (Meaning several million people. If you feel I'm wrong about this, please explain in reasonable detail why.)
I tried to figure out the apparent paradox: cooperation should be more efficient, but competition seems to work better in practice. I'm sure the answer is very complex, but the two things that first come to my mind are (a) "efficient" is only a small part of "better" almost every time and --I had a bit more trouble with this one-- (b) "cooperation" is only possible if the participants' intentions are very compatible.
a) efficient != better This should be quite easy to notice. There are other goals beside "bare efficiency". Consider flexibility, resiliency, evolution. This is the classic trade-off between monolithic and modular. Since this is/., a vague analogy: the most efficient way to do some task is to write a monolithic hand-optimized assembler-written (or hardware-implemented) block that does it. (Note that this is for efficiently _executing_ the task, not writing it.)
However, tasks where resiliency is important use redundancy a lot. I've even heard of (though I can't name any, they may be hypothetical) systems that, in order to be error resilient, have several redundant implementations of the same functionality on different hardware designed by different teams running different software written by different teams, each implementation checking the others.
Flexibility is important too. Several different groups doing the same things in parallel can try different ways of doing it. Many improvements can be made this way that would be impossible with a single "efficient" implementation. (Take the problem of 3D graphics. Suppose at one point raster algorithms are more efficient. If everything is done using raster algorithms, though, no innovations in ray-tracing are possible, though perhaps there is one that would make ray-tracing much more efficient than raster. This is the common "local minima" problem. See the paragraph after the next.)
Also, the conditions of the game (i.e., the definition of efficiency) may change. This is (AFAIK) often noticed in biology: species that are very efficient at something are very specialized, so can't adapt where the conditions change.
There's also the problem that often the problem is just very hard to solve. Meaning that, while there is a cooperative solution that is very efficient, it is not known. It's often more advantageous to form several teams that compete trying to solve the problem rather than just one trying to improve itself.
b) full cooperation needs people that want the same thing This is a bit more philosophical, I don't think I know enough to give lots of arguments. But I think that often people simply want different things.
The simplest case is resource contention: say you and I want an X, but there's just one X. Suppose X can't be shared. Our wishes are mutually exclusive. This is an intrinsically competitive problem; no solution (cooperative or not) can give us both X, so there needs to be some competition for it.
Of course, we don't have to start bashing our heads with sticks to resolve our differences. Society and governments are (at least partially) a cooperative method of solving these problems. A communist government tries (ideally) to solve the difference by estimating what everyone _needs_ and allocating all resources in such a way as to maximally satisfy those needs. A capitalist government system (ideally) sets up some rules that approximate the "free market" concept and then lets people do what the _want_ (not need!), within those rules. When a resource is limited, in the latter system leads (theoretically) to the resource being distributed by "how good people are at
I disagree. There's everything wrong with becoming a doctor to make money. Without the passion for helping people, the training means nothing; like any profession, you have to give a damn about what you're working on. [...] I'm just saying that the AMOUNT of money you make is peripheral to a good doctor: It should be enough to live comfortably; anything less is unfair, anything more makes his services prohibitively expensive.
Of course, below the surface, I'm agreeing with you, it's just that I'm picking nits. A good doctor performs a valuable service, and as a result should be compensated justly for it. The nit that I'm picking is that 'doing it for the money' implies that in this theoretical doctor's mind, the money is paramount.
Since we're picking nits anyway, I'll pick one more: (And yes, I agree that we agree below the surface:))
I know you're exaggerating a bit, but the view that "doing it for the money" means "they dream green and they see people as wallet-carriers" is quite spread, and especially so regarding doctors. People don't usually see anything wrong with becoming an architect or an engineer "just for the money" -- despite the fact that many more people die because of the latters' mistakes than the doctors'. (OK, I pulled that "many more" out of my ass. Here's hoping you'll agree with the premise.) I think there's a lot of suspicion around doctors simply because they tend to be around when bad things happen to people. There's also that half-irrational "trust" thing you mentioned... (Don't get me wrong, I do that too.) Not sure why that applies that much to doctors.
After all, a car salesman could sell cars without thinking about their safety, only of the money, and most people don't worry much about that in my experience. Granted, people don't tend to think much about car salesmen, either...
I guess what I'm trying to say is that when doctors are involved people tend to be irrationally strict. And your arguments seem to confirm this: for example, there's no reason why you couldn't be a very good doctor without having "a passion for helping people". And you could have a great passion for helping people but be absolutely incompetent (or perhaps worse, unobviously so). It's true that some well-trained, competent people do a crappy job if they don't care much about it, but that's by no means a rule. Too much enthusiasm can do just as much evil as too little.
And sorry if I'm repeating myself, but the AMOUNT of money one makes should (almost) never be unimportant and even peripheral to a good professional, in any commercial system. I dislike "unfettered capitalism" as much as any reasonable person does, and I know free markets aren't the answer to any problem. But "charge whatever you need to live comfortably" and "trust only those who do it from passion" is very idealistic. "Doing X for the money" doesn't mean money is paramount: it means one does X because they want the amount of money obtainable that way, and would do Y* if X couldn't provide the desired amount of money. This says absolutely nothing about how well they do it. It says nothing about how much they like it**. It doesn't even say they want as much money as possible.
(*: Generally anyone who is capable of becoming a decent doctor is capable of any number of high-paying jobs.)
(**: There is a positive correlation, but it's far from perfect. Being a passionate but hungry artist is romantic and all, but doesn't mean you're any good. And being a sell-out doesn't mean you're not good at what you do.)
You also say that "the AMOUNT of money [a doctor makes] should be enough to live comfortably; anything less is unfair, anything more makes his services prohibitively expensive" -- which is absolutely not true, and could be applied to (probably) any job and still wouldn't be true. I won't go into details, I think it's easy to see there's a very large distance between "absolutely minimum" and "prohibitively expensive". Not to mention
1) There's nothing wrong with becoming a doctor to make money, as long as you're a decent doctor (i.e., you don't actually hurt people through incompetence or malice). That's true for all professions. You might like helping people, or have a natural inclination towards it, but this doesn't mean you shouldn't do it for the money.
2) Even if you're not in it for the money, this doesn't mean you have to sacrifice your life and not even break even. (In our case, remember the guy still has debts from school.) Why aren't you in Africa helping people?
In general (and I'm only saying this because we're on slashdot), remember that software libre works because if Linus writes a kernel, it can be used by ten million people with almost zero marginal cost (and most of it borne by them, not him). A doctor has to see each patient individually, which means that we can't make do with just a few brilliant people, we actually need lots and lots of doctors to work all the time. Which means for this kind of jobs there needs to be an incentive, and quite a strong one. (If the incentive is low, there are fewer persons interested in the job, so you have fewer persons to select from, so you'll have more mediocre and bad doctors. Because smart guys go to better payed jobs, even if they would like to help people. Compare with what happens in children education in most places. Low pay means low quality service.)
Also, to retain your nerd and/or geek credentials, you must be able to quote from two or more of Star Wars, Star Trek, Babylon 5, Stargate, Firefly, or Andromeda. You will be expected to pick one of these as a religion* and from time to time wage holy war on the rest for forsaking The One True Way. Also you must be able to recite on demand the Spam sketch, the Dead Parrot sketch, and 90% of the Princess Bride script**.
back in the day, that list used to include Monty Python and Star Wars
Back in the day people used to read a post before answering.
As far as I remember (not much above your level of exposure), the trick is like that:
You have an opaque screen with two slits of appropriate size.
(Not sure exactly what 'appropriate size' means, but I'm almost sure it's not necessary they be microscopic. Also, I'm sure you don't need _slits_ (actually, the article does this with a stencil), but it's easier to calculate the results that way, so you can test the theory.)
On one side of the slitted screen, you have a panel of photon detectors, in a plane parallel with the screen. (A matrix, like the CCD of a digital camera.) This allows you to detect photons hitting the detector panel.
On the other side, you have a "photon gun" that can send photons in the general direction of the slitted screen, from a fixed point. (That is, all photons pass approximately through the same point, or a small circular opening.)
When you shoot photons (individually or lots of them), some hit the screen and are not detected, some pass through the slits and do get detected by some of the photon detectors. Of course, because you have a matrix of detectors, you can tell where the photon hit the panel. Because the photon acts as a PARTICLE when you measure it, for each photon shot by the photon gun _exactly_ one detector will be activated. (Sort of; its "wave function" collapses, which means about as much for me as it means for anyone)
The WAVE part of the experiment is this: if you shower the screen with photons and count how many photons hit the detector panel _in_each_pixel_, you can form a grayscale image (black and white photo) that shows the "density of photon hits".
Now, if photons acted like particles in flight, the photo would show two lines of high-density, and the rest no hits. (About how the shadow of two macroscopic rectangles in a piece of paper would look like; the wave effect is visible only for small sizes, not sure exactly how small.) But because the slits are not detectors (the measurement is done by the detector panel, not the slits), the photons keep their wave-status while traveling, so the photo would actually look very different. (Like interference waves.)
In particular, you would have hits detected in spots that are not on a straight line between the photon gun and the slits. (A particle could not get in those points.) This is very important, because of the next thing:
If you shoot only single photons, they will be able to get to the same spots! That is, they will be detected in spots that a particle could not have hit, given the geometry of the slits and the panels! Even more, if you do the experiment multiple times (you shoot a photon, record where it hit, you shoot a photon again later, record when that one hits, etc.), and then you add all the detections, you'll see that the distribution of the photons (where are they more or less likely to hit) is the same as what you got when shooting lots of photons, which was an wave interference-pattern. But since you only shot photons one-by-one, which means there was always at most one photon passing through the slits, it means each photon interferenced with itself!
So, the conclusion is that while a photon flies happily through space (or transparent matter), it acts like a wave that expands through the entire space. (For example, it can interference with itself passing through two different slits at the same time.) If you try to do measurements (with the detector panel, for instance), it will instantly "make up its mind" and behave like a particle, with its position chosen randomly _but_ with a probability governed by how the wave was distributed in space at that moment.
(I'm not sure what happens when you try to also record the photons while they pass through the slits. The extra measurement should collapse the wave function at the slits, and the detector wall should loose the interference pattern. But the extra measurement also changes the way the photons move, so I'm not sure what will appear on the detector. I'm guessing it would look like a 'normal shadow" of two slits, but blurry.)
How many centimeters in a third of a meter? 33.3333333333333333333333333333333333333333333.
Or, about 33 cm. (Or 333.3 mm if you need the maximum eye-on-ruler precision.)
How about how many inches in a fifth of a yard or a foot? It's not that much harder, but when you need to do three or four operations like this (say if you need to get to volumes, or energy, or mass or density -- say to approximate how much energy you'll need to warm or cool a room) it's usually easier to abstract everything to decimal numbers.
Also, 1/3 m can be used just as well as 5 1/4 inch, in a quick calculation, and it's easy to envision if your mind works in metric (that's what I always answer when someone wonders how much a foot means in meters). It's when you need more than one or two calculations that metric helps.
(And yes, time _is_ bothersome sometimes, I always convert to seconds if I need to do more than one or two divisions.)
I'm sorry about the ambiguity, I should have been more careful. I meant "hack" and "exploit" as in "use superior knowledge of the system for outcomes different from what it was intended for or from what less experienced people might think to use it for".
This includes those who do it under the "hacker ethic" you pointed to, as well as those who do it for bad purposes. I don't like the fact that we don't have a different word for "good hacker" (as "cracker" is for the bad guys), and indeed my whole argument was about the bad guys. The term "hacker" is overloaded, whether we like it or not.
(Note that I don't include script kiddies in the hacker class, but I do include the guys who created the 'sploitz. Misguided or not, they do fit the loose definition in the first paragraph of this post. In the previous post, this would be the distinction between a crook and the lawyer he hired.)
Your assumption that government is free of con men and cheats is scary. It's also hilarious that you think private industry generates more paperwork than the ultimate bureaucracy. All large institutions generate useless paperwork. There is a particular kind of organism that thrives on ass covering and rule following, and anytime something gets big enough that it starts needing rules to function, they move in and take over.
Just eliminate such people with automation. A bureaucrat and an expert system have a lot in common- but the one big difference is that the expert system is incorruptable.
I encountered this idea a few times before, but I'm not very convinced it would work any better than a government of people.
First of all, consider that an expert system needs someone to write the rules -- the legislative branch of a government. And you can't do that once and be done with it. There's a reason why the Senate/House of Representatives/Parliament exists all the time instead of writing some laws and dissolving themselves -- actually, there are several reasons, and "just to have something to do" is not the whole of it.
Second, you need people to feed the system date, interpret it, and enforce it. This means lawyers, notaries & clerks, judges, and policemen. (No, I don't think Robocop was a good idea.) That's pretty much what the judicial branch of government does.
And then you need people who can take decisions quickly to react to unforeseen events (no expert system can cover everything) -- the executive. (I think police is usually listed here, but I'm sure you get my point.)
And all that doesn't do anything to prevent hackers -- people who understand the rules better that others and can exploit any weakness for profit. That's exactly what criminals (in the general sense) do, from thieves and kidnappers to corrupt administrations. Someone who exploits the tax code (for instance) is a hacker just as much as someone who exploits a vulnerability of a computer program. It's just the machine that's different.
As to your other two statements, I don't agree with the first's premise, and I think the second is self-contradictory.
What other form of application runs without a separate step of installation?
That's your reason for accepting applications with few features and incredibly poor performance? That's like saying "I could make ten times more money, but then I'd have to pay more taxes." It doesn't make sense.
It's not as wrong as you make it sound:
True, a dedicated application running locally is most of the time superior to a web one, enough so to justify the usually minor hassle of installing it.
However, this cost is not always minor. Accessing gmail or gtalk or google maps from any computer with a web browser is much more important to me than having them run quicker, with less memory. (And in 3D, for the maps thing.) I've used these from school, other friend's computers, internet cafes, even a coin-operated internet terminal it a railway station. In most of these places I can't install programs or change POP settings. But I can log on to gmail or use google maps just fine. (OK, I have to risk keyloggers, but the "installing an app" way wouldn't help with that either.)
Which, as a matter of fact, is what the whole point of AJAX is.
Just because a massive gorilla sending you a letter doesn't mean there's evil intent behind it. That's just saying anything more powerful than you is evil, automatically, just because they can.
I'd say that's a lot more cynical than believing that if someone who can force you to do something (that's damaging to them) just asks you nicely, it might not be just for PR.
I didn't fully miss your point, I just failed to answer clearly. I was referring only to the "children born darker" sentence when I talked about missing the point -- of course, the chest thing with twins is perfectly appropriate for the story, and is a similar example to the ones I gave.
Where I think you are wrong is just the last thing, the relation between environment and gene propagation and activation. There are several possible ways in which the environment "influences genes" (that I know of).
Consider a population with the usual random distribution of genes: for example, more or less dark people. (There are many genes influencing skin color, before (at birth) and after UV-induced tan, but my argument is general, so let's ignore the details for a bit).
(a) The environment can select who reproduces. For example, very light-skinned people may have skin cancer more often, thus have fewer offspring. This changes the distribution of genes in the population (after many generations!).
(b) The environment can activate some genes after birth. For example, people will get darker sun-tans if they are subject to more UV. This makes the population exposed to UV darker-skinned, but _doesn't_ change the distribution of genes in the next generations _at_all_. (I.e., if the UV-radiation becomes suddenly weaker, the population will get lighter-skinned _after_just_one_generation_.)
(c) The environment can influence the process of reproduction (for example, some chemicals can theoretically kill sperm cells that have some variant of a gene but not others), thus changing the distribution of genes in the population _after_some_generations_.
The standard theory of natural evolution (Darwinism) deals pretty much _only_ with (a). In particular, this is (as far as I know) the only likely mechanism that can cause future humans to be more dark-skinned because of UV.
The reason I said you (may have) missed the point is that I think you stated that case (b) will lead to the results of case (a), a couple of posts ago.
Case (c) ---which I think is what you are talking about in the last post--- is a bit more complicated. In the _vast_majority_ of cases, the changes in the gene distribution (after some generation) are not correlated in effect with the cause.
In our example, rises in UV radiation can (this is all theoretical, remember) kill sperm cells (or ovules) that have a certain gene, thus making that gene less common in future generations. But there is no reason for it to kill gametes with light-skin genes. They can just as well kill sperm cells with long leg-genes or sicklemia-genes. This is because the selection (e.g. killing* the gametes) happens _before_ the genes express their traits. Because there are lots of genes, and only a few are linked to skin color, it's statistically _very_ improbable that the cause is correlated to the effect (remember, the genes didn't express yet!). Essentially, it _can_ happen, but it's just random chance.
*) It doesn't have to kill gametes, just make them less likely to fertilization, e.g. make sperm cells less mobile.
There is another subtlety here to be wary of: there may be biological mechanisms in _some_ animals where there is a way for _some_ combination of genes to be selected in the gametes as a direct result of environmental factors before birth. For example (this is very speculative!), UV radiation may cause some hormone change in the mothers that kill embryos with light-skin genes. This would be an expression of an evolved trait of the _mother's_ genes, not a random coincidence, and this is a complex variant of (a) rather than (c).
I don't know if the hypothetical situation in the last paragraph actually happens in any existing species of animals. I _think_ it's much less likely than normal (a)-case ---thus, it happens much rarer--- because it's indirect evolution. (It's indirect evolution because the expression of a gene doesn't affect the individual with the genes, but its descendants.)
When I was in the Netherlands (Holland), I stayed in a town where there were no stop lights, and no stop signs (Almelo). [...] The only accident I saw while I was there, was a 'farmer' who sped through a blind intersection and clipped the bumper of another car (he was going nearly double the speed limit).
I guess it is a bit bothersome to remember three or four sentences back when you can just throw a little barb. Feels better that way, doesn't it?
I think you miss the point. The idea is that _a single individual_ of a species (including humans) has in its genome the potential for several different phenotypes ("shapes" in short).
Which is bloody obvious: body-builders don't look the way they do because of genes, but because they pump iron. (Of course, some people _do_ find it easier to do body building because their genes favorize muscle mass, and some people really can't do body building for genetic reasons, but these are only the extreme cases.)
I can't say if this applies to all athletic disciplines. For example, I have no idea if professional basketball players are usually tall because (a) only genetically-tall people will be good enough to play professional basketball or (b) playing lots of basketball makes you tall. But things like muscle mass and bone density are clearly strongly influenced by stresses on the body (i.e. the environment).
In response to your question about the ozone, as far as the story is concerned, no: people will just tend to be darker because they get more sun-tan. They will all be born exactly as white as their (close) ancestors were, but they'll get darker in time. Which is bloody obvious too, I'm not sure what's the big deal with the article.
There is a genetic part too, but that isn't something you'll notice: people who, because of random variations in genes, are born a bit darker will tend to get skin cancer a bit less often, and thus will tend to reproduce a bit more often, than those who are born a bit lighter.
Note the subtle fact that the increase in UV radiation _does_not_ select the embrio between the possible embrios of the reproducing pair, it in fact selects between the possible reproducing pairs. You won't see people being born darker, you will see darker people giving birth.
(This is all theoretical, I'm not saying all men will be dark in a million years; the UV intensity changed before, too, and the effect is minute.)
He was already driving at _double the speed limit_. This means he was already breaking the law (by quite a big margin). If someone is (or was) _already_ breaking the law, what makes you think that _adding_ a new law is going to fix things?
Do you not have a manual, non-electric engine starter? How do you start in cold weather without forcing a bit more gas through the carburetor and adjusting the advance of the ignition just right? How do you force a choked engine to start without turning the hand crank a few times? Hint: it can be done automatically, too.
I drive a manual gearbox, but there are (now, at least some) automatic gearbox cars that can do everything you said. I've been in one that did automatically notice when you released the gas pedal a bit quick and did engine braking automatically (all controlled through the gas- and brake-pedal feedback, it was quite intuitive). It did drop a gear automatically if I pressed the gas pedal suddenly, to give quick acceleration.
Incidentally this was in the winter, in the mountains (I was at sky in the Alps), snowy/icy roads (it was above zero during the day, the snow melted, then at sundown it all froze quickly, and then it started snowing quite hard). Traction control was better than I was.
And yes, the engine was started by pressing a button.
Slashdot Mirror
Right now I have six cables under my desk, for power. If there was no USB, I'd have two more. If there was no WiFi, another two cables. If there was no Bluetooth, I'd have another two cables hanging around. And I don't have an iPod, yet.
I don't mind having a wire around the desk, it's having a dozen that I mind. Each cable that I can get rid off helps. This thing would decrease the number of cables on, around, and under my desk by more than any other wireless tech did.
And yes, the minor annoyance of having to pick up a cable from the floor, route it around the other cables and the various things on my desk (yes, I need them, and no, I don't have room for a bigger desk), find which of the seven holes in the back of my laptop fits the plug, several times every day of my life, and the same for the phone, and spending a half hour re-arranging the cables every week, is something I'd pay a decent amount of money to go without.
I'm too lazy to check exactly, but I did read most of TFA, and Larry, Eric and Serghei together hold about 70% of the votes, IIRC.
You're mostly right about the LCD panels -- they can't fix this without head-tracking. Head-tracking would work very well, I think, with something like the Wii's sensor on the glasses, but only for a single viewer.
However, for cinema I think it's less important, as the scenes are generally "far away" and the viewers "not very mobile", thus the brain expects very little parallax change.
Yes, that's what I was curious about. I won't bet a cent on the security properties of this thing, but I'm very curious how well it might work in keeping out the wireless signals of the neighbors. I never get fewer that three foreign access points around the house, and I don't even live in a very dense apartment block. I'd need a repeater within certain frequencies for cell-phone and other out-of-the-house wireless I might want. (This would be double-nice: since cell phones communicate through the repeater/amplifier, they would need lower transmission power!)
Also, is the paint reflective or absorbent? In the former case, it might even boost my signal a bit inside the house, provided that I'm careful to use it only on external walls.
And how much does this thing cost? Since they're marketing it as a snake oil derivative, I guess it's too much to be worth it.
It's a bummer that all comments are about the bogus security instead of the nice signal-shaping possibilities.
I think the definition is more like "unreasonable belief that someone is trying to get you". For a simple example, just because a terrorist intends to, say, "kill any and all Americans" doesn't mean that all paranoids in America become instantly sane (or, at least, not paranoid anymore).
Because birds can't transport tons (literally) of materials and people over thousands of kilometers in a few hours. For that matter, planes can't land on tree branches.
If we wanted an artifact that can do what birds can (e.g. land on a tree branch) we probably would start by building a mechanical bird. It's quite hard, and I doubt we can do that today. We can't build a mechanical brain (i.e. an artifact that can do what a brain does), either.
Are those free and open source games? I'd guess not, but if I'm wrong, just port them to OpenGL :)
I'm sure it's _very_ nontrivial---which is why it's not done often---but lots of what a parser does could be done in parallel. For instance, you really don't _need_ to find whitespace linearly. You can split the string (for large programs) in pieces. Stitching the results is going to be a bitch, but it's possible. For a million processors you _could_ actually load one character on each processor and do everything in parallel. I never saw an algorithm that does this on text, but I've seen pattern recognition and edge detection done with a-processor-per-pixel algorithms, and if you think about it properly parsing is something analogue: you have to find the tree hidden in that mess of characters.
Again, this would be a bitch to program, perhaps only possible by automatic tools, but it could work.
I simply meant that free markets really do solve a vast set of economic problems, but not _all_ of them, and not _all_ problems we need to solve are purely economic. For instance, there are social, moral and ethic problems. These are intertwined with economic issues (with perhaps some extraordinary exceptions), but this means that the solution is not always purely an economic one (again, with some possible exceptions).
I didn't think I had to mention this in my post, because it actually dealt with such issues: the fact that people tend to think of medicine and doctors more from the point of view of ethics, morality, trust --- and often very irrationally. The reason my post was so economy-oriented was just because I wanted to show why people should think of this (also) as an economic problem.
You did it much more incisively and tersely than I was capable.
(Though I don't agree fully with you: (1) I don't think the rising cost of health care is always fully caused by increases in quality, and even if it were it might not necessarily be acceptable. There's no reason a better treatment can't be cheaper, at least sometimes. (2) I have only little formal training in economy, but as far as I understand it pure laissez-fare doesn't solve everything; unless I misunderstand the terms, it allows e.g. monopolies. I think there needs to be some equilibrium between the size and number of agents before free markets work "well". (3) People often tend to argue for "free health care", as you said. But that's not the only meme, there's also "some people simply can't pay; what do we do about it?" I don't mean I support a particular answer, that's for another discussion. But free markets give a precise answer --- we don't do anything, let them deal with it --- which may not be a "good answer".)
This is very interesting. To my (admittedly very limited) knowledge of things economical, a "very cooperative society" as you describe translates to a communist country. Which (I'm sure it's been said before) always sounds like a very good idea, but was never implemented acceptably _on a large scale_. (Meaning several million people. If you feel I'm wrong about this, please explain in reasonable detail why.)
/., a vague analogy: the most efficient way to do some task is to write a monolithic hand-optimized assembler-written (or hardware-implemented) block that does it. (Note that this is for efficiently _executing_ the task, not writing it.)
I tried to figure out the apparent paradox: cooperation should be more efficient, but competition seems to work better in practice. I'm sure the answer is very complex, but the two things that first come to my mind are (a) "efficient" is only a small part of "better" almost every time and --I had a bit more trouble with this one-- (b) "cooperation" is only possible if the participants' intentions are very compatible.
a) efficient != better
This should be quite easy to notice. There are other goals beside "bare efficiency". Consider flexibility, resiliency, evolution. This is the classic trade-off between monolithic and modular. Since this is
However, tasks where resiliency is important use redundancy a lot. I've even heard of (though I can't name any, they may be hypothetical) systems that, in order to be error resilient, have several redundant implementations of the same functionality on different hardware designed by different teams running different software written by different teams, each implementation checking the others.
Flexibility is important too. Several different groups doing the same things in parallel can try different ways of doing it. Many improvements can be made this way that would be impossible with a single "efficient" implementation. (Take the problem of 3D graphics. Suppose at one point raster algorithms are more efficient. If everything is done using raster algorithms, though, no innovations in ray-tracing are possible, though perhaps there is one that would make ray-tracing much more efficient than raster. This is the common "local minima" problem. See the paragraph after the next.)
Also, the conditions of the game (i.e., the definition of efficiency) may change. This is (AFAIK) often noticed in biology: species that are very efficient at something are very specialized, so can't adapt where the conditions change.
There's also the problem that often the problem is just very hard to solve. Meaning that, while there is a cooperative solution that is very efficient, it is not known. It's often more advantageous to form several teams that compete trying to solve the problem rather than just one trying to improve itself.
b) full cooperation needs people that want the same thing
This is a bit more philosophical, I don't think I know enough to give lots of arguments. But I think that often people simply want different things.
The simplest case is resource contention: say you and I want an X, but there's just one X. Suppose X can't be shared. Our wishes are mutually exclusive. This is an intrinsically competitive problem; no solution (cooperative or not) can give us both X, so there needs to be some competition for it.
Of course, we don't have to start bashing our heads with sticks to resolve our differences. Society and governments are (at least partially) a cooperative method of solving these problems. A communist government tries (ideally) to solve the difference by estimating what everyone _needs_ and allocating all resources in such a way as to maximally satisfy those needs. A capitalist government system (ideally) sets up some rules that approximate the "free market" concept and then lets people do what the _want_ (not need!), within those rules. When a resource is limited, in the latter system leads (theoretically) to the resource being distributed by "how good people are at
Since we're picking nits anyway, I'll pick one more: (And yes, I agree that we agree below the surface :))
I know you're exaggerating a bit, but the view that "doing it for the money" means "they dream green and they see people as wallet-carriers" is quite spread, and especially so regarding doctors. People don't usually see anything wrong with becoming an architect or an engineer "just for the money" -- despite the fact that many more people die because of the latters' mistakes than the doctors'. (OK, I pulled that "many more" out of my ass. Here's hoping you'll agree with the premise.) I think there's a lot of suspicion around doctors simply because they tend to be around when bad things happen to people. There's also that half-irrational "trust" thing you mentioned... (Don't get me wrong, I do that too.) Not sure why that applies that much to doctors.
After all, a car salesman could sell cars without thinking about their safety, only of the money, and most people don't worry much about that in my experience. Granted, people don't tend to think much about car salesmen, either...
I guess what I'm trying to say is that when doctors are involved people tend to be irrationally strict. And your arguments seem to confirm this: for example, there's no reason why you couldn't be a very good doctor without having "a passion for helping people". And you could have a great passion for helping people but be absolutely incompetent (or perhaps worse, unobviously so). It's true that some well-trained, competent people do a crappy job if they don't care much about it, but that's by no means a rule. Too much enthusiasm can do just as much evil as too little.
And sorry if I'm repeating myself, but the AMOUNT of money one makes should (almost) never be unimportant and even peripheral to a good professional, in any commercial system. I dislike "unfettered capitalism" as much as any reasonable person does, and I know free markets aren't the answer to any problem. But "charge whatever you need to live comfortably" and "trust only those who do it from passion" is very idealistic. "Doing X for the money" doesn't mean money is paramount: it means one does X because they want the amount of money obtainable that way, and would do Y* if X couldn't provide the desired amount of money. This says absolutely nothing about how well they do it. It says nothing about how much they like it**. It doesn't even say they want as much money as possible.
(*: Generally anyone who is capable of becoming a decent doctor is capable of any number of high-paying jobs.)
(**: There is a positive correlation, but it's far from perfect. Being a passionate but hungry artist is romantic and all, but doesn't mean you're any good. And being a sell-out doesn't mean you're not good at what you do.)
You also say that "the AMOUNT of money [a doctor makes] should be enough to live comfortably; anything less is unfair, anything more makes his services prohibitively expensive" -- which is absolutely not true, and could be applied to (probably) any job and still wouldn't be true. I won't go into details, I think it's easy to see there's a very large distance between "absolutely minimum" and "prohibitively expensive". Not to mention
I think I'm feeding a troll here, but:
1) There's nothing wrong with becoming a doctor to make money, as long as you're a decent doctor (i.e., you don't actually hurt people through incompetence or malice). That's true for all professions. You might like helping people, or have a natural inclination towards it, but this doesn't mean you shouldn't do it for the money.
2) Even if you're not in it for the money, this doesn't mean you have to sacrifice your life and not even break even. (In our case, remember the guy still has debts from school.) Why aren't you in Africa helping people?
In general (and I'm only saying this because we're on slashdot), remember that software libre works because if Linus writes a kernel, it can be used by ten million people with almost zero marginal cost (and most of it borne by them, not him). A doctor has to see each patient individually, which means that we can't make do with just a few brilliant people, we actually need lots and lots of doctors to work all the time. Which means for this kind of jobs there needs to be an incentive, and quite a strong one. (If the incentive is low, there are fewer persons interested in the job, so you have fewer persons to select from, so you'll have more mediocre and bad doctors. Because smart guys go to better payed jobs, even if they would like to help people. Compare with what happens in children education in most places. Low pay means low quality service.)
Oh, wait, this is /.
As far as I remember (not much above your level of exposure), the trick is like that:
You have an opaque screen with two slits of appropriate size.
(Not sure exactly what 'appropriate size' means, but I'm almost sure it's not necessary they be microscopic. Also, I'm sure you don't need _slits_ (actually, the article does this with a stencil), but it's easier to calculate the results that way, so you can test the theory.)
On one side of the slitted screen, you have a panel of photon detectors, in a plane parallel with the screen. (A matrix, like the CCD of a digital camera.) This allows you to detect photons hitting the detector panel.
On the other side, you have a "photon gun" that can send photons in the general direction of the slitted screen, from a fixed point. (That is, all photons pass approximately through the same point, or a small circular opening.)
When you shoot photons (individually or lots of them), some hit the screen and are not detected, some pass through the slits and do get detected by some of the photon detectors. Of course, because you have a matrix of detectors, you can tell where the photon hit the panel. Because the photon acts as a PARTICLE when you measure it, for each photon shot by the photon gun _exactly_ one detector will be activated. (Sort of; its "wave function" collapses, which means about as much for me as it means for anyone)
The WAVE part of the experiment is this: if you shower the screen with photons and count how many photons hit the detector panel _in_each_pixel_, you can form a grayscale image (black and white photo) that shows the "density of photon hits".
Now, if photons acted like particles in flight, the photo would show two lines of high-density, and the rest no hits. (About how the shadow of two macroscopic rectangles in a piece of paper would look like; the wave effect is visible only for small sizes, not sure exactly how small.) But because the slits are not detectors (the measurement is done by the detector panel, not the slits), the photons keep their wave-status while traveling, so the photo would actually look very different. (Like interference waves.)
In particular, you would have hits detected in spots that are not on a straight line between the photon gun and the slits. (A particle could not get in those points.) This is very important, because of the next thing:
If you shoot only single photons, they will be able to get to the same spots! That is, they will be detected in spots that a particle could not have hit, given the geometry of the slits and the panels! Even more, if you do the experiment multiple times (you shoot a photon, record where it hit, you shoot a photon again later, record when that one hits, etc.), and then you add all the detections, you'll see that the distribution of the photons (where are they more or less likely to hit) is the same as what you got when shooting lots of photons, which was an wave interference-pattern. But since you only shot photons one-by-one, which means there was always at most one photon passing through the slits, it means each photon interferenced with itself!
So, the conclusion is that while a photon flies happily through space (or transparent matter), it acts like a wave that expands through the entire space. (For example, it can interference with itself passing through two different slits at the same time.) If you try to do measurements (with the detector panel, for instance), it will instantly "make up its mind" and behave like a particle, with its position chosen randomly _but_ with a probability governed by how the wave was distributed in space at that moment.
(I'm not sure what happens when you try to also record the photons while they pass through the slits. The extra measurement should collapse the wave function at the slits, and the detector wall should loose the interference pattern. But the extra measurement also changes the way the photons move, so I'm not sure what will appear on the detector. I'm guessing it would look like a 'normal shadow" of two slits, but blurry.)
How about how many inches in a fifth of a yard or a foot? It's not that much harder, but when you need to do three or four operations like this (say if you need to get to volumes, or energy, or mass or density -- say to approximate how much energy you'll need to warm or cool a room) it's usually easier to abstract everything to decimal numbers.
Also, 1/3 m can be used just as well as 5 1/4 inch, in a quick calculation, and it's easy to envision if your mind works in metric (that's what I always answer when someone wonders how much a foot means in meters). It's when you need more than one or two calculations that metric helps.
(And yes, time _is_ bothersome sometimes, I always convert to seconds if I need to do more than one or two divisions.)
This includes those who do it under the "hacker ethic" you pointed to, as well as those who do it for bad purposes. I don't like the fact that we don't have a different word for "good hacker" (as "cracker" is for the bad guys), and indeed my whole argument was about the bad guys. The term "hacker" is overloaded, whether we like it or not.
(Note that I don't include script kiddies in the hacker class, but I do include the guys who created the 'sploitz. Misguided or not, they do fit the loose definition in the first paragraph of this post. In the previous post, this would be the distinction between a crook and the lawyer he hired.)
First of all, consider that an expert system needs someone to write the rules -- the legislative branch of a government. And you can't do that once and be done with it. There's a reason why the Senate/House of Representatives/Parliament exists all the time instead of writing some laws and dissolving themselves -- actually, there are several reasons, and "just to have something to do" is not the whole of it.
Second, you need people to feed the system date, interpret it, and enforce it. This means lawyers, notaries & clerks, judges, and policemen. (No, I don't think Robocop was a good idea.) That's pretty much what the judicial branch of government does.
And then you need people who can take decisions quickly to react to unforeseen events (no expert system can cover everything) -- the executive. (I think police is usually listed here, but I'm sure you get my point.)
And all that doesn't do anything to prevent hackers -- people who understand the rules better that others and can exploit any weakness for profit. That's exactly what criminals (in the general sense) do, from thieves and kidnappers to corrupt administrations. Someone who exploits the tax code (for instance) is a hacker just as much as someone who exploits a vulnerability of a computer program. It's just the machine that's different.
As to your other two statements, I don't agree with the first's premise, and I think the second is self-contradictory.
True, a dedicated application running locally is most of the time superior to a web one, enough so to justify the usually minor hassle of installing it.
However, this cost is not always minor. Accessing gmail or gtalk or google maps from any computer with a web browser is much more important to me than having them run quicker, with less memory. (And in 3D, for the maps thing.) I've used these from school, other friend's computers, internet cafes, even a coin-operated internet terminal it a railway station. In most of these places I can't install programs or change POP settings. But I can log on to gmail or use google maps just fine. (OK, I have to risk keyloggers, but the "installing an app" way wouldn't help with that either.)
Which, as a matter of fact, is what the whole point of AJAX is.
Just because a massive gorilla sending you a letter doesn't mean there's evil intent behind it. That's just saying anything more powerful than you is evil, automatically, just because they can.
I'd say that's a lot more cynical than believing that if someone who can force you to do something (that's damaging to them) just asks you nicely, it might not be just for PR.
I didn't fully miss your point, I just failed to answer clearly. I was referring only to the "children born darker" sentence when I talked about missing the point -- of course, the chest thing with twins is perfectly appropriate for the story, and is a similar example to the ones I gave.
Where I think you are wrong is just the last thing, the relation between environment and gene propagation and activation. There are several possible ways in which the environment "influences genes" (that I know of).
Consider a population with the usual random distribution of genes: for example, more or less dark people. (There are many genes influencing skin color, before (at birth) and after UV-induced tan, but my argument is general, so let's ignore the details for a bit).
(a) The environment can select who reproduces. For example, very light-skinned people may have skin cancer more often, thus have fewer offspring. This changes the distribution of genes in the population (after many generations!).
(b) The environment can activate some genes after birth. For example, people will get darker sun-tans if they are subject to more UV. This makes the population exposed to UV darker-skinned, but _doesn't_ change the distribution of genes in the next generations _at_all_. (I.e., if the UV-radiation becomes suddenly weaker, the population will get lighter-skinned _after_just_one_generation_.)
(c) The environment can influence the process of reproduction (for example, some chemicals can theoretically kill sperm cells that have some variant of a gene but not others), thus changing the distribution of genes in the population _after_some_generations_.
The standard theory of natural evolution (Darwinism) deals pretty much _only_ with (a). In particular, this is (as far as I know) the only likely mechanism that can cause future humans to be more dark-skinned because of UV.
The reason I said you (may have) missed the point is that I think you stated that case (b) will lead to the results of case (a), a couple of posts ago.
Case (c) ---which I think is what you are talking about in the last post--- is a bit more complicated. In the _vast_majority_ of cases, the changes in the gene distribution (after some generation) are not correlated in effect with the cause.
In our example, rises in UV radiation can (this is all theoretical, remember) kill sperm cells (or ovules) that have a certain gene, thus making that gene less common in future generations. But there is no reason for it to kill gametes with light-skin genes. They can just as well kill sperm cells with long leg-genes or sicklemia-genes. This is because the selection (e.g. killing* the gametes) happens _before_ the genes express their traits. Because there are lots of genes, and only a few are linked to skin color, it's statistically _very_ improbable that the cause is correlated to the effect (remember, the genes didn't express yet!). Essentially, it _can_ happen, but it's just random chance.
*) It doesn't have to kill gametes, just make them less likely to fertilization, e.g. make sperm cells less mobile.
There is another subtlety here to be wary of: there may be biological mechanisms in _some_ animals where there is a way for _some_ combination of genes to be selected in the gametes as a direct result of environmental factors before birth. For example (this is very speculative!), UV radiation may cause some hormone change in the mothers that kill embryos with light-skin genes. This would be an expression of an evolved trait of the _mother's_ genes, not a random coincidence, and this is a complex variant of (a) rather than (c).
I don't know if the hypothetical situation in the last paragraph actually happens in any existing species of animals. I _think_ it's much less likely than normal (a)-case ---thus, it happens much rarer--- because it's indirect evolution. (It's indirect evolution because the expression of a gene doesn't affect the individual with the genes, but its descendants.)
Now that I think of i
I think you miss the point. The idea is that _a single individual_ of a species (including humans) has in its genome the potential for several different phenotypes ("shapes" in short). Which is bloody obvious: body-builders don't look the way they do because of genes, but because they pump iron. (Of course, some people _do_ find it easier to do body building because their genes favorize muscle mass, and some people really can't do body building for genetic reasons, but these are only the extreme cases.) I can't say if this applies to all athletic disciplines. For example, I have no idea if professional basketball players are usually tall because (a) only genetically-tall people will be good enough to play professional basketball or (b) playing lots of basketball makes you tall. But things like muscle mass and bone density are clearly strongly influenced by stresses on the body (i.e. the environment). In response to your question about the ozone, as far as the story is concerned, no: people will just tend to be darker because they get more sun-tan. They will all be born exactly as white as their (close) ancestors were, but they'll get darker in time. Which is bloody obvious too, I'm not sure what's the big deal with the article. There is a genetic part too, but that isn't something you'll notice: people who, because of random variations in genes, are born a bit darker will tend to get skin cancer a bit less often, and thus will tend to reproduce a bit more often, than those who are born a bit lighter. Note the subtle fact that the increase in UV radiation _does_not_ select the embrio between the possible embrios of the reproducing pair, it in fact selects between the possible reproducing pairs. You won't see people being born darker, you will see darker people giving birth. (This is all theoretical, I'm not saying all men will be dark in a million years; the UV intensity changed before, too, and the effect is minute.)
He was already driving at _double the speed limit_. This means he was already breaking the law (by quite a big margin). If someone is (or was) _already_ breaking the law, what makes you think that _adding_ a new law is going to fix things?
Do you not have a manual, non-electric engine starter? How do you start in cold weather without forcing a bit more gas through the carburetor and adjusting the advance of the ignition just right? How do you force a choked engine to start without turning the hand crank a few times? Hint: it can be done automatically, too.
I drive a manual gearbox, but there are (now, at least some) automatic gearbox cars that can do everything you said. I've been in one that did automatically notice when you released the gas pedal a bit quick and did engine braking automatically (all controlled through the gas- and brake-pedal feedback, it was quite intuitive). It did drop a gear automatically if I pressed the gas pedal suddenly, to give quick acceleration.
Incidentally this was in the winter, in the mountains (I was at sky in the Alps), snowy/icy roads (it was above zero during the day, the snow melted, then at sundown it all froze quickly, and then it started snowing quite hard). Traction control was better than I was.
And yes, the engine was started by pressing a button.