No competition? Mentioning Mozilla as an example? Hmm, let me see...
http://en.wikipedia.org/wiki/List_of_web_browsers_ for_Unix/Linux There is plenty of competition. During the last year I've switched my distribution of choice more times than I switched web browser ( yes, that includes the browser version number ). You just don't do that with closed source software. You don't just try out a Mac for the evening and use the package manager to download the latest version of Office for it. When you switch between closed source software you plan to use the alternative for a while, or otherwise you just wasted the license fee. Compare that with a free distribution. With open source you could set up 4-5 partitions and have a quadruple boot with Debian, Fedora, OpenBSD and Open Solaris and with a fast broadband connection you could have most of your favorite software running on each of them within a few hours. This means you have very strong competition between projects. That some projects appear to have more or less a monopoly is mainly because they work, and people don't feel the need for anything else. Where you do have demand for alternatives people will either write it or get existing software to support it.
Simply put, for SOME calculations you MIGHT be able to get ONE answer, the shiny part is that it is sometimes possible to make the computer give you the interesting answer and none of the ones you don't care about. I.e if you try to factorize a large integer you can set up a QC to calculate the result of dividing it by every other integer, and if you do things right you can make it output only those integers which yield a result of 0. I'm not completely sure, but I think this is possible for only certain problems with certain properties, but given that integer factorization is of central importance to mathematics, and an exponential time problem for a classical computer, that problem alone is worth the effort of making one. It would among other things allow you to brute force an RSA key in the same time it takes to generate one. This is not to say that all encryption would be broken. Only those schemes where a QC can brute force the key in short time would be affected. It would not allow you to read a one time pad as an example. It would however drastically reduce the number of secure encryption algorithms as a large number of calculations that are "slow" for a classical computer are "quick" in a QC.
The main problem with building a quantum computer is that you have to keep the quantum states separate from the environment or they collapse. So if there is a 90% chance that any one of your qubits will be successfully isolated for the duration of the calculation, the chance that all of your 333 qubits will be ok is less than 10^-15. What is worse, because each of the qubits is entangled with the other ones, any single one of them interacting with the environment will destroy not just that bit but ALL the qubits. Imagine trying to build a computer where an error in a single bit will reset all your RAM to random values. Now try to do that for a 2048 bit RSA key. To have just a 10% chance of success each of your qubits would need a probability of success greater than 99.9%. This may not seem difficult since classical computers manage it fine, but they don't have to deal with a practically 0 error tolerance. For the serial port anything between -3V and -12V is good enough to represent a 1. For a corresponding quantum system it needs to 1 , not 1.5 or 1.1 or even 1.001 It needs to be 1, no more, no less. This is why quantum computers typically involve cooling the whole thing close to absolute zero. If you don't the thermal noise of the system will be enough to nuke your memory. Remember, it is not required that you try to measure the value for the system to cause a collapse, it is sufficient that any reaction takes place which means that you could theoretically have measured the value if you were monitoring every single particle that wasn't part of your computer. If a single helium atom in your super cooled vacuum chamber accidentally strikes one of the 2048 beryllium ions that is held in your magnetic trap that could be it. If the laser you use to read/write to your system is off by a fraction of its wavelength you might be fucked too. What you need to do is ensure that the quantum state you create is in a form which has a very low probability of interacting with anything at all, yet retain a manner in which you can cause it to react with the other qubits so that you can entangle them. After that you need to find a way to set up a circuit of them which allows you to recover just those results you are interested in. I'm amazed they have managed to do this for even a few bits. This is sci-fi which makes a light-saber look like a trivial device.
Take 20 boxes and then let a bunch of hacker lose on them. Pay them $money for every box they manage to crack. Make 10 of the boxes run fully patched Windows and 10 run the stable branch of OpenBSD and stick complete computer novices behind them. In fact, make the OpenBSD boxes run the OpenBSD project's apache version, OpenSSH server, give the hackers an account on it and have every daemon listen to every port and enable X11 forwarding through SSH. The windows machines can run a fully patched Vista with all the ports under a firewall. I bet most people would still prefer trying to compromise a Windows box.
Seriously, don't come and tell me there wouldn't be fewer security problems if windows went away. Vista's security model is based on the "how do we design this so we can blame the user" while the open source distros are based on "lets be open about vulnerabilities so we can fix them asap". Heck, even if the open source ones were as vulnerable as windows I would still prefer them because at least then you can be relatively certain they will be open about it. With Microsoft you are more likely to get told of for being a user when they break something.
If you think current copyright law is not broken, considder:
Strange Drink v3.4 Copyright Suspect Foods INC 2007, All rights reserved.
Terms of consumption: This drink is provided 'as is' and comes with no warranty or fitness for any particular purpose to the extent permissable by law. Suspect Foods INC accepts no responsibility for any adverse effects that may be caused by consuming this drink, including , but not limited to, eating, drinking or smelling the drink in question. The drink remains the property of Suspect Foods INC at all times, and this license can be revoked at any time. You do not have any righst to consume, redistribute or modify this drink other than as specifically permitted by this license...
Sure, abolishing copyright law may have undesireable effects, but as long as it remains in force the GPL is the next best thing.
What many people don't seem to realise (and indeed something I missed myself when I was playing around with a copputer simulation of Stimulated Raman Scattering in inertial confinement fusion as part of my undergraduate degree) is that temperture is not just the average of the kinetic energy of the particles involved. Consider what woudl happen if you fire a bullet at a very high speed. The average kinetic energy of the particles in the bullet is quite high, but seen from the rest frame of the bullet they only move due to thermal vibrations. It is the velocity of hydrogen nuclei relative to one another that matters. In a plasma these velocities are due to the high temperature, and it is therefore very hot. However, there isn't any particular reason why you couldn't carefully fire the nuclei against one another in a particle accelerator. In such a case, you could accelerate all the nuclei to pretty much the same velocity, and it wouldn't be "hot" in the thermal sense. If you do it with a fusion reaction that doesn't produce neutrons you can even extract the kinetic energy of the resulting particles directly, circumventing the carnot limit of efficiency because you don't let the energy get converted into heat. In practice there are all kinds of reasons why such a scheme is not practical, and a hot plasma confined in a magnetic field is the most promising way to generate power, but it is far from the only plausible way ( at least from a theoretical point of view). As for the polywell scheme it is nonsense. If you are to confine a plasma using a magnetic field you will need a very strong field to avoid heat losses, period. You can confine a plasma fairly easily ( a flourescent light tube manages just fine ) but to be useful as a power source it has to do so at a sufficiently high density, and more importantly, with low energy losses to the surroundings. If you try to use a weak magnetic field your plasma will end up with a lowe density and large surface area, meaning it will rapidly lose energy to the surroundings. Furthermore, even if you manage to confine it ( as modern reactors do ) that is only the start of your problems. Once the reaction actually gets going you will see your vessel exposed to neutron radiation even stronger than what you have in the core of a traditional nuclear powerplant. Now in a traditional nuclear reactor this is not a [show stopping] problem because the coolant, moderator, nuclear fuel, and controll rods all soak up a whole lot of neutrons. In a fusion reactor you have a fairly empety vacume chamber, and your reactor vessel, magnets, hydrogen injection system, divertor, temperature sensors etc are all directly exposed to the hottest plasma in the solar system ( yes that includes the core of the sun ). The polywell scheme hasn't even managed to achieve fusion rates similar to those used in traditional tabletop devices ( yes they exists ) that the Oil industry use for prospecting. Even if he could get his scheme to confine the plasma, he has not proposed any way to get the fusion products out of the plasma, how to deal with the neutron radiation, nor how to make it produce enough excess power to be useful as a powerplant. In sumary, it goes. "I think this might work, but I can't give details as to why, I think I saw 3 neutrons (there are 10^23 nuclei in 1 gram of hydrogen ) but I can't show you how I did it because my device is broken. Please give me some money." Tabletop fusion is easy ( easy enough that teenagres do it in their basements ) making a practical power source out of it is a completely different thing.
All you people who wonder how one could have missed something so obvious, well, it isn't obvious. Firstly, what actually happens inside a solid-state laser is an absolute pain to work out rigorously. The quantum mechanics involved is sufficiently complicated that the preferred method of finding a good lasing material is pretty much trial and error. You are talking about systems where adding a few fractions of a percent of impurities changes the energy levels, and consequentially the physical properties of your semi-conductor. Changing the temperature at which you grow your silicon crystal, or at what rate you cool it, or how you add the doping material, or weather the moon is in the seventh phase, can have implications for how these things work. You are dealing with processes that occur at nanometer length scales, where you can't just fit things together using a bit of duct tape and a can of WD40 ( tho don't quote me on that, because it is just far too likely that someone crazy enough will find a way to use WD40 in a laser ).
What I'm trying to say is that sticking a diode to your home made radio circuit is one thing, finding a way to do it to a laser based on an ultra-pure silicon crystal, which changes its behavior if you do anything more severe than to think about it is hard (and keep in mind, this is QM we are talking about, chances are thinking about it does in fact kill the cat... ). My experience with solid state physics is rather limited, but I know one thing. It sure as hell isn't obvious... (If you doubt it, go look up the explanation for why gallium arsenide makes better photo voltaic cells than silicon).
There is one significant advantage of computers over pen-and paper. They accept a wide number of possible interfaces, which make them useful for people who have various disabilities. I had a friend with quite severe dyslexia who simply wasn't able to proof read his essays on his own. When it started to become acceptable for us to use spell checkers for our essays he suddenly found his assignments no more difficult than the rest of us. Another of my friends have problems writing for long periods of time because of an injury. Speak to text software is obviously very useful for him. There is a lot of ways in which computers can help with learning, but it is important tor realize that it is a complement to, not a substitute for, competent teachers and a well prepared curriculum.
Say what you want about Theo and the obscure license policies of OpenBSD, but OpenSSH is in my opinion one piece of software which is simply better than sliced bread ( server as well as client). Since I was introduced to it I use it on a daily basis and it has been rock solid since the first login. It's the kind of software that inspires you to write software yourself. Two thumbs up.
He is promoting the right thing, for the wrong reasons, meaning I don't trust him to actually do it. In analogy, removing suppressive dictators with the use of force is in my opinion the right thing to do, but if you do it only because you have a big Oil lobby backing you, then I don't trust you to do it right, or to do it when it happens somewhere that doesn't have Oil. What reason do we have to suspect that Obama will not suddenly change his mind the moment a republican opens a my-space site quoting the democratic campaign? He may not even back on the copyright front but instead push the libel front in the wrong direction. There are still many reasons to prefer him from his opponents, but I really don't trust him on this one.
Eliminating or fixing windows ( lets face it, that is what this is all about ) would not make a system perfectly secure, and thus the security industry would still have a place in providing security as a service. It would change the security business model quite a bit, and the companies would probably get the main share of their revenue from developers rather than end users, but the need to audit code for flaws and vulnerabilities would still be there. The only way the security industry would not be needed would be if there were no attackers, and that is not about to happen any time soon.
The reason bar codes are not sufficient is that once they are read, they can be easily copied. The same goes for any static message transmitted by an RFID tag. Also, the database can obviously be corrupted by an evil government or disgruntled worker. If you really want to have a forge-proof solution you will need to implement something like OpenPGP in every passport. I can't wait until the day where politicians and media will have to be careful with their creditability or risk having a significant number of people revoke their certificate... Want people to trust you about the foreign policy? Well lets just have a look at that signature of yours...
I've had a look at OpenBSD and while it is indeed very encouraging to have an OS that is secure without tweaking, as opposed to one where you have to know about, and disable, everything that could be a risk, there are still a few things that trouble me.
a)If you want to follow the stable branch you need to compile from source. The OpenBSD developers correctly discourage you from compiling from source as it is more likely to break things, yet it is required for teh stable branch, which is annoying.
b)The install is a bit complicated, especially if you are going to compile from source ( which you have to if you want to follow the stable branch ). There is quite a lot of work required here to get a system that is "secure by default".
c)The price of the CD images. Yes, I know this is in order to fund the project, but it only really affects new users since those who have been using OpenBSD for some time are probably fine with the net-install or will donate money anyway. This policy hits new users, and if you don't really want to spend a lot of money on a system just to try it, this leaves you with an even more complicated install procedure. Maybe it could be an idea to have a "minimal" install CD for the very basics of the system, and then charge for the rest of the stuff.
Most of these problems seem to be down to limited resources rather than the capabilities of the OpenBSD team. I don't really care much about the license policy, because quite frankly its their project and they are free to license it as they like. Personally I prefer the GPL for things I would write myself, but as a user having less restrictions is never a bad thing. All in all it's a nice project and I will probably try it out once I get some more experience with *NIX systems. For now I will stick with my Debian install however.
Just have a look at this thread. Virtually nobody complains. If not even a bunch of slash dot trolls bother to bring prophesies of doom about the choice of Ubuntu, then it just shows that all the worries about what distribution Dell would pick have just been squashed, big time. Just look at this thread. There are a couple of worries about the hardware, a few reliefs that they didn't go with Novell, but there isn't really any sharp criticism of the choice of Ubuntu. Not that strange of course. As everyone said before. If the hardware is right, stick whatever you want on it, we can fix the software ourselves...
The main cost of Nuclear power is not the labour, but the capital investment necessary to build the plant. This has reduced drastically over the last few decades ( more so than solar panels ) and is set to sink even more. Furthermore, there is an easy way to reduce the costs of labor.. You increase reactor power. You don't need much more personell for a 1000MW reactor than you need for a 140MW one. Now, contrary to solar, there is no fundamental limit to reactor output other than safety limits deliberately incorporated by design ( and these limits rise as technology improve ). The only physical limit of a reactor's output is the energy content of uranium, which is HUGE. Just think about it for a second. A few kilograms of uranium contains enough energy to power a nuclear weapon capable of demolishing a city. When used in a more peaceful manner it allows you to produce almost unlimited amounts of energy for any forseable future ( no, really, look up "breeder reactors" if you doubt me ). A cubic meter of uranium contains enough energy to provide a constant energy output of thousands of megawatts for more than a year. Thus I would argue that nuclear has a much greater potential for reducing its overall costs, since the only thing limiting is technology, whereas solar cell's face a physical limit of production, set by the amount of energy they receive from the sun.
Contrary to the misconception people keep throwing arround, it wasn't 20 years of 20 years ago. The confusion arises because one was talking about different things. One estimate was when we would reach break-even. That eastimate was for year 2000, and at the time ( 1970) it was 30 years into the future. As it happens, the JET reactor has managed to heat a plasma to the temperatures needed for break-even, but that doesn't mean it is practical as a powerplant. I have a 30 year old book about electricity generation, which estimates the first powerplant for 2050. Furthermore, last time I heard "it was always X years ago", X was 30. Before that X was and had "always" been 50 years ( Tho my Swedish book still says 2050 and was written in the 70ies ). I bet in 2040 we will hear people saying how widescale worldwide deployment of fusion powerplants was "always" 10 more years. When in fact, the estimate of today is that the technology needed to build a practical powerplant ( not necessarily an economically competative one ) is 2027. These "that is what they said back then" quotes usually have no substance in reality. It is just like saying "well they said chernobyl was safe", which of course nobody ever claimed ( in contrast the department of energy stated that no water cooled graphite moderated reactor would be licensed in the US ). However, the claim sounds so damning that people want to believe it. It is the same thing with fusion. The scientists never claimed we would be using fusion plants today. They claimed that IF funding was continued, and IF projects were not cancelled, then we would be able to have a controlled fusion reaction by the year 2000. As it happens we have done better than that. We have managed to initiate fusion reactions that produce more energy than is needed to sustain them. This is however not the same thing as an economically competative powerplant, and it is not the same as ignition ( a fusion plasma that needs no external energy input once it is burning). If you keep changing the goal to be something more difficult, then yes, the goal will always be in the future, that doesn't mean the original estimate was wrong tho. It just mean you were talking about something else.
So just why would you recover CO2 from the air when it would be much easier to do so before it leaves the chimney? Until every single fossil fuel plant uses CCS, this is a waste of time, and if every fossil fuel plant used CCS we wouldn't really have much of a problem anyway. The easiest way to recover CO2 is to not emit it in the first place.
Is it plagiarism? Sure is. Copyright infringement, you bet it is. Immoral, oh yea. The guy deserves a decent slap? Yup.
Is it theft? Hell no. Theft refers to depriving someone of their property. To copy a work without permission is copyright infringement. To claim someone else's work is your own is plagiarism. The only ones who benefit from blurring these terms are those who seek to rewrite the law without having to bother about the "inconvenient" proper procedure involved in doing so.
Here we go again. Let me sum arise it.
-The cheapest, most efficient AND easiest way to collect solar energy is as heat.
-If this was cheap enough, people would use solar heating all over the place.
-Solar heating remains of limited popularity
-If solar heating is not competitive with other energy sources, despite a dramatically lower price than photovoltaics, and despite better efficiencies than are even theoretically possible with photovoltaics, then photovoltaics, which will inevitably be less efficient, more expensive, and less durable, is not going to be competitive. EVER.
The ONLY way solar energy can become competitive is for the price of other forms of energy generation to sky-rocket. Whereas this may be true for fossil fuels, nuclear, wind, hydro, geothermal and biomass are not set to become more expensive any time soon. On the contrary, developments in reactor technology is set to make nuclear costs comparable to gas ( and that includes waste disposal and decommissioning ). Wind power is seeing improvements as we speak. Geothermal is not set to get any more expensive, and biomass is already competitive with fossil fuels. Solar simply doesn't stand a chance. Even if solar cell's were as cost efficient as solar heat collectors, they would still lose out compared to the alternatives. Solar cells are good for remote applications where you can't stick a power plant, like in an orbiting satellite. For pretty much everything else they are rubbish.
First of all table-top fusion is really simple to do, and devices exist that achieve it using a 9V battery. Such devices are routinely used for various kinds of scanners, as neutron-sources for nuclear experiments, various kinds of material testing... etc Getting D-D fusion or D-T fusion is sufficiently easy for hobbyists to do it in their basement. What is tricky, however, is to generate a controllable plasma that can produce enough energy for it to be practical as a power source, and this is orders of magnitude more difficult. Every month I hear some new plan about how to achieve fusion, the truth is, getting fusion to work is not hard. What would be interesting would be if this device could demonstrate a high triple-product. I.e if it can achieve a high plasma density, high temperature, AND high confinement time simultaneously. In practice THAT is really difficult to do, mainly because for any feasible pressure the temperature required will be in the range of hundreds of millions of degrees, meaning it will radiate A LOT of energy in the form of electromagnetic radiation, leading to a low confinement time. ( the sun gets away with "only" ten million centigrades because of the intense pressure in the core ). The only way this could possibly work would be if he has actually reduced bremsstrahlung losses A LOT. If I understand it correctly he claims to have done that by separating nuclei and electrons, which quite frankly is bullshit. 1 gram of hydrogen contains [roughly] 10^23 nuclei, giving 10000 coulomb's of charge if not kept neutral by electrons. Now, for those of you who know your electrostatics, try sticking 10000 coulomb into coulomb's law of electrostatic repulsion for a device that separates the charges by a distance of 1 meter or so, and then tell me this scheme will work. There is a reason you need a strong containment field for a fusion reaction...
Ultimate solution to this is obviously to switch to fiber-optics. Laminate the chips in a small layer of a transparent conductor ( yes it is very possible ) then do inter-chip comunications optically. Faraday - 1 , Hackers - 0. For the slightly more affordable aproach you can use a conducting cage aka-tinfoil hat. As long as you ensure that the only signal going in and out of the cage is optic good old Faraday takes care of the rest. Power suply is the main issue, but a high-quality low-pass filter can fix that. As a bonus, a computer built this way would also be hardened against EMP's and power spikes. If it is hard for signals to leak out, it is also hard for power spikes to get in. Man, physics is cool some times...
So the idea is that you use a piece of legacy technology (the barcode) to tag another piece of legacy tech (optical disks) in order to encode information in a manner that, unlike a website, can't be changed... Explain just how this is better than a URI and a camera with text-recognition ? Furthermore, why would it replace the original barcode? You already have a perfectly good product-ID right there in the good old fashioned one. You have a phone connected to the web...
Hey, I have an idea. Why don't we punch a string of holes at the edge of the packaging to encode a demo of the software sold? Yea, that would be great...
This is why your browser ideally shouldn't be able to read your entire home directory. People talk about running as admin or not, but your most sensitive data is your personal files that you have read access to as your limited user. Running as admin or root is bad mainly because it can open security holes which can cause further mischief, but if your most personal information, and your most important files, are right there for your browser to read, it won't matter if the exploit hits the kernel or simply your browser. The way I have it set up my browser runs as a separate user which connects as an un-trusted X-client. Files that I don't care about are in a directory with the group set so that the browser can read them, while personal documents, e-mail... etc is readable by my user only.Now, in practice I am not very secure. I still trust google with my e-mail, I allow sites to set cookies etc... I set this up mainly as a proof of principle thing. There isn't any good reason why your browser, which is arguably the most exposed part of your system, should be able to fuck up your entire home directory and send your most private data somewhere it doesn't belong.
Could someone explain to me why you are at such an increased risk if your contact details are online? Most abuse is from family or close relatives, or people known to the family. School staff, neighbors, local sports club etc... I mean I'm sure I have given away sufficient information about myself to track me down at some point, but why would anyone bother? The way I see it the real risk is people online persuading others to "meet up" in places where they are more at risk, or easier to exploit. Thus the real danger is not that people will find your contact details passively. Heck, you can use the phone book or just stalk someone from the local school to do that. The real danger is if people are not careful when they meet new people ( in real life or online ). If you do meet up with someone ensure its in a public place, bring a friend ( preferably adult ). Do it during hours when there will be many people around etc... If you happen to have children, explain the risks to them, and tell them not to make contact with people they find online. Really, online profiles are little more of a risk than the phone book is. The problem arises because whereas few people would just go out and hook up with some random stranger who call them on the phone or send them a letter ( tho I'm sure some do ) quite a few people have problems doing so with someone they found on an online web-forum.
The way I see it, and what official agencies keep telling us, is that there are three things that keep reoccurring among school shootings.
a)A history of bullying or rejection b)Mental health problems c)Guns (duh)
What is not on that list is video games, the media, the "culture of fear".. etc. The whole problem is basically down to improving availability and standards of care for people with mental health problems, reduce bullying in schools, and actually implement some gun restrictions. "Guns don't kill people, people kill people" yea sure, but it is a whole lot easier for people to kill people with a gun than it is with a knife. It should be rather obvious... Gun control isn't THE solution, but it would help bring down the body count...
As for the positive correlation between violent media and violent behavior ( yes, it is real ) I'd like to see some peer-reviewed papers on distinguishing cause and effect. It doesn't surprise me that there is a correlation between violence and violent media, but it doesn't directly follow that media is the cause and violent personalities the effect. After all, if people didn't appreciate violence such media would not be very successful, would it?
Slashdot Mirror
No competition? Mentioning Mozilla as an example? Hmm, let me see... http://en.wikipedia.org/wiki/List_of_web_browsers_ for_Unix/Linux There is plenty of competition. During the last year I've switched my distribution of choice more times than I switched web browser ( yes, that includes the browser version number ). You just don't do that with closed source software. You don't just try out a Mac for the evening and use the package manager to download the latest version of Office for it. When you switch between closed source software you plan to use the alternative for a while, or otherwise you just wasted the license fee. Compare that with a free distribution. With open source you could set up 4-5 partitions and have a quadruple boot with Debian, Fedora, OpenBSD and Open Solaris and with a fast broadband connection you could have most of your favorite software running on each of them within a few hours. This means you have very strong competition between projects. That some projects appear to have more or less a monopoly is mainly because they work, and people don't feel the need for anything else. Where you do have demand for alternatives people will either write it or get existing software to support it.
Simply put, for SOME calculations you MIGHT be able to get ONE answer, the shiny part is that it is sometimes possible to make the computer give you the interesting answer and none of the ones you don't care about. I.e if you try to factorize a large integer you can set up a QC to calculate the result of dividing it by every other integer, and if you do things right you can make it output only those integers which yield a result of 0. I'm not completely sure, but I think this is possible for only certain problems with certain properties, but given that integer factorization is of central importance to mathematics, and an exponential time problem for a classical computer, that problem alone is worth the effort of making one. It would among other things allow you to brute force an RSA key in the same time it takes to generate one. This is not to say that all encryption would be broken. Only those schemes where a QC can brute force the key in short time would be affected. It would not allow you to read a one time pad as an example. It would however drastically reduce the number of secure encryption algorithms as a large number of calculations that are "slow" for a classical computer are "quick" in a QC. The main problem with building a quantum computer is that you have to keep the quantum states separate from the environment or they collapse. So if there is a 90% chance that any one of your qubits will be successfully isolated for the duration of the calculation, the chance that all of your 333 qubits will be ok is less than 10^-15. What is worse, because each of the qubits is entangled with the other ones, any single one of them interacting with the environment will destroy not just that bit but ALL the qubits. Imagine trying to build a computer where an error in a single bit will reset all your RAM to random values. Now try to do that for a 2048 bit RSA key. To have just a 10% chance of success each of your qubits would need a probability of success greater than 99.9%. This may not seem difficult since classical computers manage it fine, but they don't have to deal with a practically 0 error tolerance. For the serial port anything between -3V and -12V is good enough to represent a 1. For a corresponding quantum system it needs to 1 , not 1.5 or 1.1 or even 1.001 It needs to be 1, no more, no less. This is why quantum computers typically involve cooling the whole thing close to absolute zero. If you don't the thermal noise of the system will be enough to nuke your memory. Remember, it is not required that you try to measure the value for the system to cause a collapse, it is sufficient that any reaction takes place which means that you could theoretically have measured the value if you were monitoring every single particle that wasn't part of your computer. If a single helium atom in your super cooled vacuum chamber accidentally strikes one of the 2048 beryllium ions that is held in your magnetic trap that could be it. If the laser you use to read/write to your system is off by a fraction of its wavelength you might be fucked too. What you need to do is ensure that the quantum state you create is in a form which has a very low probability of interacting with anything at all, yet retain a manner in which you can cause it to react with the other qubits so that you can entangle them. After that you need to find a way to set up a circuit of them which allows you to recover just those results you are interested in. I'm amazed they have managed to do this for even a few bits. This is sci-fi which makes a light-saber look like a trivial device.
Take 20 boxes and then let a bunch of hacker lose on them. Pay them $money for every box they manage to crack. Make 10 of the boxes run fully patched Windows and 10 run the stable branch of OpenBSD and stick complete computer novices behind them. In fact, make the OpenBSD boxes run the OpenBSD project's apache version, OpenSSH server, give the hackers an account on it and have every daemon listen to every port and enable X11 forwarding through SSH. The windows machines can run a fully patched Vista with all the ports under a firewall. I bet most people would still prefer trying to compromise a Windows box. Seriously, don't come and tell me there wouldn't be fewer security problems if windows went away. Vista's security model is based on the "how do we design this so we can blame the user" while the open source distros are based on "lets be open about vulnerabilities so we can fix them asap". Heck, even if the open source ones were as vulnerable as windows I would still prefer them because at least then you can be relatively certain they will be open about it. With Microsoft you are more likely to get told of for being a user when they break something.
If you think current copyright law is not broken, considder:
Strange Drink v3.4
Copyright Suspect Foods INC 2007, All rights reserved.
Terms of consumption:
This drink is provided 'as is' and comes with no warranty or fitness for any particular purpose to the extent permissable by law. Suspect Foods INC accepts no responsibility for any adverse effects that may be caused by consuming this drink, including , but not limited to, eating, drinking or smelling the drink in question. The drink remains the property of Suspect Foods INC at all times, and this license can be revoked at any time. You do not have any righst to consume, redistribute or modify this drink other than as specifically permitted by this license...
Sure, abolishing copyright law may have undesireable effects, but as long as it remains in force the GPL is the next best thing.
What many people don't seem to realise (and indeed something I missed myself when I was playing around with a copputer simulation of Stimulated Raman Scattering in inertial confinement fusion as part of my undergraduate degree) is that temperture is not just the average of the kinetic energy of the particles involved. Consider what woudl happen if you fire a bullet at a very high speed. The average kinetic energy of the particles in the bullet is quite high, but seen from the rest frame of the bullet they only move due to thermal vibrations. It is the velocity of hydrogen nuclei relative to one another that matters. In a plasma these velocities are due to the high temperature, and it is therefore very hot. However, there isn't any particular reason why you couldn't carefully fire the nuclei against one another in a particle accelerator. In such a case, you could accelerate all the nuclei to pretty much the same velocity, and it wouldn't be "hot" in the thermal sense. If you do it with a fusion reaction that doesn't produce neutrons you can even extract the kinetic energy of the resulting particles directly, circumventing the carnot limit of efficiency because you don't let the energy get converted into heat. In practice there are all kinds of reasons why such a scheme is not practical, and a hot plasma confined in a magnetic field is the most promising way to generate power, but it is far from the only plausible way ( at least from a theoretical point of view). As for the polywell scheme it is nonsense. If you are to confine a plasma using a magnetic field you will need a very strong field to avoid heat losses, period. You can confine a plasma fairly easily ( a flourescent light tube manages just fine ) but to be useful as a power source it has to do so at a sufficiently high density, and more importantly, with low energy losses to the surroundings. If you try to use a weak magnetic field your plasma will end up with a lowe density and large surface area, meaning it will rapidly lose energy to the surroundings. Furthermore, even if you manage to confine it ( as modern reactors do ) that is only the start of your problems. Once the reaction actually gets going you will see your vessel exposed to neutron radiation even stronger than what you have in the core of a traditional nuclear powerplant. Now in a traditional nuclear reactor this is not a [show stopping] problem because the coolant, moderator, nuclear fuel, and controll rods all soak up a whole lot of neutrons. In a fusion reactor you have a fairly empety vacume chamber, and your reactor vessel, magnets, hydrogen injection system, divertor, temperature sensors etc are all directly exposed to the hottest plasma in the solar system ( yes that includes the core of the sun ). The polywell scheme hasn't even managed to achieve fusion rates similar to those used in traditional tabletop devices ( yes they exists ) that the Oil industry use for prospecting. Even if he could get his scheme to confine the plasma, he has not proposed any way to get the fusion products out of the plasma, how to deal with the neutron radiation, nor how to make it produce enough excess power to be useful as a powerplant. In sumary, it goes. "I think this might work, but I can't give details as to why, I think I saw 3 neutrons (there are 10^23 nuclei in 1 gram of hydrogen ) but I can't show you how I did it because my device is broken. Please give me some money." Tabletop fusion is easy ( easy enough that teenagres do it in their basements ) making a practical power source out of it is a completely different thing.
All you people who wonder how one could have missed something so obvious, well, it isn't obvious. Firstly, what actually happens inside a solid-state laser is an absolute pain to work out rigorously. The quantum mechanics involved is sufficiently complicated that the preferred method of finding a good lasing material is pretty much trial and error. You are talking about systems where adding a few fractions of a percent of impurities changes the energy levels, and consequentially the physical properties of your semi-conductor. Changing the temperature at which you grow your silicon crystal, or at what rate you cool it, or how you add the doping material, or weather the moon is in the seventh phase, can have implications for how these things work. You are dealing with processes that occur at nanometer length scales, where you can't just fit things together using a bit of duct tape and a can of WD40 ( tho don't quote me on that, because it is just far too likely that someone crazy enough will find a way to use WD40 in a laser ).
What I'm trying to say is that sticking a diode to your home made radio circuit is one thing, finding a way to do it to a laser based on an ultra-pure silicon crystal, which changes its behavior if you do anything more severe than to think about it is hard (and keep in mind, this is QM we are talking about, chances are thinking about it does in fact kill the cat... ). My experience with solid state physics is rather limited, but I know one thing. It sure as hell isn't obvious... (If you doubt it, go look up the explanation for why gallium arsenide makes better photo voltaic cells than silicon).
There is one significant advantage of computers over pen-and paper. They accept a wide number of possible interfaces, which make them useful for people who have various disabilities. I had a friend with quite severe dyslexia who simply wasn't able to proof read his essays on his own. When it started to become acceptable for us to use spell checkers for our essays he suddenly found his assignments no more difficult than the rest of us. Another of my friends have problems writing for long periods of time because of an injury. Speak to text software is obviously very useful for him. There is a lot of ways in which computers can help with learning, but it is important tor realize that it is a complement to, not a substitute for, competent teachers and a well prepared curriculum.
Say what you want about Theo and the obscure license policies of OpenBSD, but OpenSSH is in my opinion one piece of software which is simply better than sliced bread ( server as well as client). Since I was introduced to it I use it on a daily basis and it has been rock solid since the first login. It's the kind of software that inspires you to write software yourself. Two thumbs up.
He is promoting the right thing, for the wrong reasons, meaning I don't trust him to actually do it. In analogy, removing suppressive dictators with the use of force is in my opinion the right thing to do, but if you do it only because you have a big Oil lobby backing you, then I don't trust you to do it right, or to do it when it happens somewhere that doesn't have Oil. What reason do we have to suspect that Obama will not suddenly change his mind the moment a republican opens a my-space site quoting the democratic campaign? He may not even back on the copyright front but instead push the libel front in the wrong direction. There are still many reasons to prefer him from his opponents, but I really don't trust him on this one.
Come on people, you know how to tag this one ; )
Eliminating or fixing windows ( lets face it, that is what this is all about ) would not make a system perfectly secure, and thus the security industry would still have a place in providing security as a service. It would change the security business model quite a bit, and the companies would probably get the main share of their revenue from developers rather than end users, but the need to audit code for flaws and vulnerabilities would still be there. The only way the security industry would not be needed would be if there were no attackers, and that is not about to happen any time soon.
The reason bar codes are not sufficient is that once they are read, they can be easily copied. The same goes for any static message transmitted by an RFID tag. Also, the database can obviously be corrupted by an evil government or disgruntled worker. If you really want to have a forge-proof solution you will need to implement something like OpenPGP in every passport. I can't wait until the day where politicians and media will have to be careful with their creditability or risk having a significant number of people revoke their certificate... Want people to trust you about the foreign policy? Well lets just have a look at that signature of yours...
I've had a look at OpenBSD and while it is indeed very encouraging to have an OS that is secure without tweaking, as opposed to one where you have to know about, and disable, everything that could be a risk, there are still a few things that trouble me. a)If you want to follow the stable branch you need to compile from source. The OpenBSD developers correctly discourage you from compiling from source as it is more likely to break things, yet it is required for teh stable branch, which is annoying. b)The install is a bit complicated, especially if you are going to compile from source ( which you have to if you want to follow the stable branch ). There is quite a lot of work required here to get a system that is "secure by default". c)The price of the CD images. Yes, I know this is in order to fund the project, but it only really affects new users since those who have been using OpenBSD for some time are probably fine with the net-install or will donate money anyway. This policy hits new users, and if you don't really want to spend a lot of money on a system just to try it, this leaves you with an even more complicated install procedure. Maybe it could be an idea to have a "minimal" install CD for the very basics of the system, and then charge for the rest of the stuff. Most of these problems seem to be down to limited resources rather than the capabilities of the OpenBSD team. I don't really care much about the license policy, because quite frankly its their project and they are free to license it as they like. Personally I prefer the GPL for things I would write myself, but as a user having less restrictions is never a bad thing. All in all it's a nice project and I will probably try it out once I get some more experience with *NIX systems. For now I will stick with my Debian install however.
Just have a look at this thread. Virtually nobody complains. If not even a bunch of slash dot trolls bother to bring prophesies of doom about the choice of Ubuntu, then it just shows that all the worries about what distribution Dell would pick have just been squashed, big time. Just look at this thread. There are a couple of worries about the hardware, a few reliefs that they didn't go with Novell, but there isn't really any sharp criticism of the choice of Ubuntu. Not that strange of course. As everyone said before. If the hardware is right, stick whatever you want on it, we can fix the software ourselves...
The main cost of Nuclear power is not the labour, but the capital investment necessary to build the plant. This has reduced drastically over the last few decades ( more so than solar panels ) and is set to sink even more. Furthermore, there is an easy way to reduce the costs of labor.. You increase reactor power. You don't need much more personell for a 1000MW reactor than you need for a 140MW one. Now, contrary to solar, there is no fundamental limit to reactor output other than safety limits deliberately incorporated by design ( and these limits rise as technology improve ). The only physical limit of a reactor's output is the energy content of uranium, which is HUGE. Just think about it for a second. A few kilograms of uranium contains enough energy to power a nuclear weapon capable of demolishing a city. When used in a more peaceful manner it allows you to produce almost unlimited amounts of energy for any forseable future ( no, really, look up "breeder reactors" if you doubt me ). A cubic meter of uranium contains enough energy to provide a constant energy output of thousands of megawatts for more than a year. Thus I would argue that nuclear has a much greater potential for reducing its overall costs, since the only thing limiting is technology, whereas solar cell's face a physical limit of production, set by the amount of energy they receive from the sun.
Contrary to the misconception people keep throwing arround, it wasn't 20 years of 20 years ago. The confusion arises because one was talking about different things. One estimate was when we would reach break-even. That eastimate was for year 2000, and at the time ( 1970) it was 30 years into the future. As it happens, the JET reactor has managed to heat a plasma to the temperatures needed for break-even, but that doesn't mean it is practical as a powerplant. I have a 30 year old book about electricity generation, which estimates the first powerplant for 2050. Furthermore, last time I heard "it was always X years ago", X was 30. Before that X was and had "always" been 50 years ( Tho my Swedish book still says 2050 and was written in the 70ies ). I bet in 2040 we will hear people saying how widescale worldwide deployment of fusion powerplants was "always" 10 more years. When in fact, the estimate of today is that the technology needed to build a practical powerplant ( not necessarily an economically competative one ) is 2027. These "that is what they said back then" quotes usually have no substance in reality. It is just like saying "well they said chernobyl was safe", which of course nobody ever claimed ( in contrast the department of energy stated that no water cooled graphite moderated reactor would be licensed in the US ). However, the claim sounds so damning that people want to believe it. It is the same thing with fusion. The scientists never claimed we would be using fusion plants today. They claimed that IF funding was continued, and IF projects were not cancelled, then we would be able to have a controlled fusion reaction by the year 2000. As it happens we have done better than that. We have managed to initiate fusion reactions that produce more energy than is needed to sustain them. This is however not the same thing as an economically competative powerplant, and it is not the same as ignition ( a fusion plasma that needs no external energy input once it is burning). If you keep changing the goal to be something more difficult, then yes, the goal will always be in the future, that doesn't mean the original estimate was wrong tho. It just mean you were talking about something else.
So just why would you recover CO2 from the air when it would be much easier to do so before it leaves the chimney? Until every single fossil fuel plant uses CCS, this is a waste of time, and if every fossil fuel plant used CCS we wouldn't really have much of a problem anyway. The easiest way to recover CO2 is to not emit it in the first place.
Is it plagiarism? Sure is.
Copyright infringement, you bet it is.
Immoral, oh yea.
The guy deserves a decent slap? Yup.
Is it theft? Hell no. Theft refers to depriving someone of their property. To copy a work without permission is copyright infringement. To claim someone else's work is your own is plagiarism. The only ones who benefit from blurring these terms are those who seek to rewrite the law without having to bother about the "inconvenient" proper procedure involved in doing so.
Here we go again. Let me sum arise it. -The cheapest, most efficient AND easiest way to collect solar energy is as heat. -If this was cheap enough, people would use solar heating all over the place. -Solar heating remains of limited popularity -If solar heating is not competitive with other energy sources, despite a dramatically lower price than photovoltaics, and despite better efficiencies than are even theoretically possible with photovoltaics, then photovoltaics, which will inevitably be less efficient, more expensive, and less durable, is not going to be competitive. EVER. The ONLY way solar energy can become competitive is for the price of other forms of energy generation to sky-rocket. Whereas this may be true for fossil fuels, nuclear, wind, hydro, geothermal and biomass are not set to become more expensive any time soon. On the contrary, developments in reactor technology is set to make nuclear costs comparable to gas ( and that includes waste disposal and decommissioning ). Wind power is seeing improvements as we speak. Geothermal is not set to get any more expensive, and biomass is already competitive with fossil fuels. Solar simply doesn't stand a chance. Even if solar cell's were as cost efficient as solar heat collectors, they would still lose out compared to the alternatives. Solar cells are good for remote applications where you can't stick a power plant, like in an orbiting satellite. For pretty much everything else they are rubbish.
First of all table-top fusion is really simple to do, and devices exist that achieve it using a 9V battery. Such devices are routinely used for various kinds of scanners, as neutron-sources for nuclear experiments, various kinds of material testing... etc Getting D-D fusion or D-T fusion is sufficiently easy for hobbyists to do it in their basement. What is tricky, however, is to generate a controllable plasma that can produce enough energy for it to be practical as a power source, and this is orders of magnitude more difficult. Every month I hear some new plan about how to achieve fusion, the truth is, getting fusion to work is not hard. What would be interesting would be if this device could demonstrate a high triple-product. I.e if it can achieve a high plasma density, high temperature, AND high confinement time simultaneously. In practice THAT is really difficult to do, mainly because for any feasible pressure the temperature required will be in the range of hundreds of millions of degrees, meaning it will radiate A LOT of energy in the form of electromagnetic radiation, leading to a low confinement time. ( the sun gets away with "only" ten million centigrades because of the intense pressure in the core ). The only way this could possibly work would be if he has actually reduced bremsstrahlung losses A LOT. If I understand it correctly he claims to have done that by separating nuclei and electrons, which quite frankly is bullshit. 1 gram of hydrogen contains [roughly] 10^23 nuclei, giving 10000 coulomb's of charge if not kept neutral by electrons. Now, for those of you who know your electrostatics, try sticking 10000 coulomb into coulomb's law of electrostatic repulsion for a device that separates the charges by a distance of 1 meter or so, and then tell me this scheme will work. There is a reason you need a strong containment field for a fusion reaction...
Ultimate solution to this is obviously to switch to fiber-optics. Laminate the chips in a small layer of a transparent conductor ( yes it is very possible ) then do inter-chip comunications optically. Faraday - 1 , Hackers - 0. For the slightly more affordable aproach you can use a conducting cage aka-tinfoil hat. As long as you ensure that the only signal going in and out of the cage is optic good old Faraday takes care of the rest. Power suply is the main issue, but a high-quality low-pass filter can fix that. As a bonus, a computer built this way would also be hardened against EMP's and power spikes. If it is hard for signals to leak out, it is also hard for power spikes to get in. Man, physics is cool some times...
So the idea is that you use a piece of legacy technology (the barcode) to tag another piece of legacy tech (optical disks) in order to encode information in a manner that, unlike a website, can't be changed... Explain just how this is better than a URI and a camera with text-recognition ? Furthermore, why would it replace the original barcode? You already have a perfectly good product-ID right there in the good old fashioned one. You have a phone connected to the web...
Hey, I have an idea. Why don't we punch a string of holes at the edge of the packaging to encode a demo of the software sold? Yea, that would be great...
This is why your browser ideally shouldn't be able to read your entire home directory. People talk about running as admin or not, but your most sensitive data is your personal files that you have read access to as your limited user. Running as admin or root is bad mainly because it can open security holes which can cause further mischief, but if your most personal information, and your most important files, are right there for your browser to read, it won't matter if the exploit hits the kernel or simply your browser. The way I have it set up my browser runs as a separate user which connects as an un-trusted X-client. Files that I don't care about are in a directory with the group set so that the browser can read them, while personal documents, e-mail... etc is readable by my user only.Now, in practice I am not very secure. I still trust google with my e-mail, I allow sites to set cookies etc... I set this up mainly as a proof of principle thing. There isn't any good reason why your browser, which is arguably the most exposed part of your system, should be able to fuck up your entire home directory and send your most private data somewhere it doesn't belong.
Could someone explain to me why you are at such an increased risk if your contact details are online? Most abuse is from family or close relatives, or people known to the family. School staff, neighbors, local sports club etc... I mean I'm sure I have given away sufficient information about myself to track me down at some point, but why would anyone bother? The way I see it the real risk is people online persuading others to "meet up" in places where they are more at risk, or easier to exploit. Thus the real danger is not that people will find your contact details passively. Heck, you can use the phone book or just stalk someone from the local school to do that. The real danger is if people are not careful when they meet new people ( in real life or online ). If you do meet up with someone ensure its in a public place, bring a friend ( preferably adult ). Do it during hours when there will be many people around etc... If you happen to have children, explain the risks to them, and tell them not to make contact with people they find online. Really, online profiles are little more of a risk than the phone book is. The problem arises because whereas few people would just go out and hook up with some random stranger who call them on the phone or send them a letter ( tho I'm sure some do ) quite a few people have problems doing so with someone they found on an online web-forum.
The way I see it, and what official agencies keep telling us, is that there are three things that keep reoccurring among school shootings.
a)A history of bullying or rejection
b)Mental health problems
c)Guns (duh)
What is not on that list is video games, the media, the "culture of fear".. etc. The whole problem is basically down to improving availability and standards of care for people with mental health problems, reduce bullying in schools, and actually implement some gun restrictions. "Guns don't kill people, people kill people" yea sure, but it is a whole lot easier for people to kill people with a gun than it is with a knife. It should be rather obvious... Gun control isn't THE solution, but it would help bring down the body count...
As for the positive correlation between violent media and violent behavior ( yes, it is real ) I'd like to see some peer-reviewed papers on distinguishing cause and effect. It doesn't surprise me that there is a correlation between violence and violent media, but it doesn't directly follow that media is the cause and violent personalities the effect. After all, if people didn't appreciate violence such media would not be very successful, would it?