Solder is already conductive, so the eddy current losses won't be localized in the iron particles. Further, copper traces are even more conductive.
This must be based on the hysteresis losses in the iron B-H curves. That means he's probably got a very high frequency magnetic field generator that he's using to heat up the iron. Seems like a simple principle.
Could also be from friction heating as the iron particles get moved around within the solder paste.
If the soldering was done using this technique your chips would never get a good solder joint in the first place. Ever "cold soldered" (not heated a joint up enough)? Yeah, it doesn't work well - the solder doesn't adhere. Now, you can make up for this by heating the solder even hotter so it spreads its heat to the joint - but then you may as well have soldered the normal way.
Wow, talk about not getting it. The iron particles are inductively heated through an alternating magnetic field. They heat the tin/lead alloy through intimate contact. Heating is localized because (a) the iron particles are only in the solder mixture and (b) a good soldering design will have focused magnetic fields or magnetic shielding to ensure that heating is approximately limited to where it is needed. Eventually, the solder will reach the normal melting temperature but since heating is applied inductively (to which the iron particles are far more sensitive than most everything else in an assembled PCB), the heating is localized to where it is needed, rather than globally applied as is the case with solder wave bath, reflow ovens, and vapor-phase reflow, or semi-locally as is the case with hot air streams.
The interesting part of this technology will be to understand how it differs in implementation and efficiency from a decent focused beam IR method since it would seem to have the same advantages of localized heating.
"the researchers say that by varying electric current to a secured computer"...
Um, if they have physical access to the computer (in order to monkey with the power), why would it be considered secure?
The faults described by the paper are so... what's the word... specialized that it challenges believability. Not only does the attacker have to have physical access -- and likely pretty good physical access -- they have to know precisely when the encryption algorithms are being performed so that the faults can be induced then and only then otherwise the operation of the computer will be compromised. Furthermore, the faults must be induced at a reasonable, but not too great, rate, and at randomly varying times in the computation, so as to explore the full error space and have insight into the keys. And the computations have to be repeated MANY times over in order to extract enough information. So, not only do attackers have to know exactly, to the microsecond, when the system under attack is computing the RSA algorithm, they also have to be able to vary the voltage to the CPU. Their physical proof of concept, as much as it is described in the paper, is contrived. Their assertion that the technique does not require physical access is wholly unsupported. Color me skeptical. Anyone with this level of access is going to be able to do more than trigger faults.
The paper asserts that the probes can be done without leaving any trace. I don't know about the authors, but the voltages on my computers are monitored by software and excursions logged so that I can know if/when there are problems. Since the RSA-breaking technique requires substantial exploration of the response to voltage tweaks, it is likely to be detected by a decent monitoring program.
Finally, the PDF does not carry any publication information suggesting strongly that it describes work that is not peer-reviewed. It is shoddy science to bypass peer review and release to the general public.
We're also assuming that the cost given hasn't already taken that into account. This article is pure marketing, you can expect them to use every trick in the book.
And that eBay, etc., didn't get highly preferred pricing in exchange for being able to state that they are early-adopting customers saving big bucks.
(I'm just waiting for the Google press release stating that they've received NRC approval to construct a nuclear plant.)
You sound like a quick study. Go buy a copy of "Introduction to Algorithms" by Cormen, Leicerson, and Rivest (yes, *that* Rivest). It's a very, very good book. But, make no mistake, any book on algorithms that is easy is a bad one, and this one, while clear, is difficult because there are some algorithms, some very important ones, that are hard to understand. Pick up a used copy from Amazon.
Your brain works very hard to integrate vision from the two eyes. That said, we can probably guess how a single-eyed prosthesis user would feel by comparing with how sighted people who have significantly different vision in their two eyes feel. Drawing from personal experience, one of my eyes has much sharper vision than the other, and when I'm not wearing my glasses, I can tell that most of my vision comes from the better eye. But, importantly, not all of it. Depth perception, for example, uses both eyes, and, again personally, if I close my poorer eye, there is a large difference.
So, my best guess is that someone with a VGA-grade implant on one side and a normal eye on the other would have better vision than someone with only a single functioning eye, but most of the visual experience would still come from the healthy eye.
Here's the thing with any sort of augmented vision: there's no way you can justify the risks of implantation when a fully external device that shows whatever mapped, morphed, or manipulated version of vision will work as well or better.
If you have normal sight, or even nearly normal sight, then why have an implant that carries significant risk, will be large and potentially painful for some time to come, will require frequent recharging, will be expensive as getout, when you can put on a special pair of glasses with a heads-up-display that does more? Telescopic vision, IR, UV, macroscopic, x-ray, edge enhanced, color shifted, depth enhanced, whatever you can think of, it is easier to do it with a head-worn high-tech display that you can take off at will.
In contrast, having an implant means -- for any kind of implant that is under current consideration -- fixed resolution, and, unless you're willing to undergo significant, expensive surgeries, many of the interface parameters will be technologically fixed. Yes, there's a lot you can do with reprogramming, but it's essentially impossible to change the stimulating electrodes and their drivers.
Trust me, you do not want a visual prosthesis unless you need one. The normal visual system, enhanced with purely external devices, will always be better.
Any visual implant that is currently under discussion
First of all, 30-40,000 lines of code is not lots of code. Try, 250,000 of code.
Depends on the language. For most of the popular languages (C / C++, Lisp, Scheme, Perl, Python, PHP, Ruby, Fortran, Matlab, Ada, ALGOL, whatever), you are spot-on. For languages that have higher density (APL primary among them, but I might also include hand-written assembler), 30-40k lines is seriously daunting.
The difference between bands like Girl Talk who sample music to create new pieces, and someone copying someone else's words into a paper they're writing, is that Girl Talk doesn't claim to have made the samples. One of the aspects of why plagiarism is seen as wrong is because you're taking credit for someone else's work. When you're sampling music, you're crediting them.
Agreed. Often, sampled sounds are clearly recognizable as they are intended to bring an association from the original material: take something everyone knows, twist it, shuffle with other sounds, and make something new. The artists are not trying to hide the origin of their samples, but paying homage to them. Indeed, there is a long-standing history of one artist performing works by another, adding their own touch to the music. Furthermore, when that happens and the second artist makes money at it (sometimes even when they don't), they have to pay royalties to the first artist.
Unattributed lifting without manipulation is not the same. Didn't we just have an article earlier today about cheating in CS classes at Stanford? It's not just illegal, it's unethical.
Just make the punishment for cheating sufficiently harsh. You cheat.. you get kicked out. Simple.
I dunno about Stanford but when I went to school my CS classes (especially the earlier ones) were huge. I never met most of my classmates. I would be *extremely* pissed off to have my academic standing affected by someone else's cheating.
The summary omits the fact that raising the weight of the final is not the ONLY action that happens. The student who is caught is at least reprimanded, if not much worse. The point of the novel mechanism is that everyone else in the class (even if the single individual is expelled) is also punished by re-weighting the grade evaluation to more heavily favor the final exam. Furthermore, that at each instance of cheating, there is another adjustment. Presumably peer pressure or some fear of potential guilt is supposed to act as a deterrent, but I don't see that. It's not an effective motivation for someone who is likely to cheat.
When I was at MIT, teaching the core EECS classes (6.001 through 6.004) late 80s and early 90s, cheating was taken *very* seriously. The very best a student could hope for is to fail the class (I never heard of such a case, though). Nominally it meant leaving the department (with a forced change of majors), if not expulsion from the university. That said, since the punishment was so harsh, and since cheating is difficult to prove in general, students whom we suspected of cheating were usually given a stern warning and put on informal probation. Still, there were disciplinary committee meetings from time to time to deal with students who were provably shown to have cheated. The chair of the committee for many years was a family friend --- and, boy, he was tough. You did not want to be even accused of cheating.
I am very careful to avoid viruses and other malware (always was when i was healthy and Win32 was only a secondary OS for me then). But to be stabbed in the back would be utterly devastating to me. It could be weeks before I could get qualified help (Nerd Herd, etc. need not apply).
I work in the field of prosthetics (although on the sensory, rather than motor side), so I am vitally aware of the importance of interfaces like the one you are using.
That said, since things like bad patches, viruses, trojans, hardware failures, and the like do happen, no matter how careful you might be, perhaps you should think about having a spare, secondary system, in case your primary system fails for some reason. Also, it might be a good idea to implement some kind of automated backup system so that you can roll back to the previous known-good configuration if something goes wrong. Unfortunately, I'm not qualified to make suggestions for how that might work with Windows XP, but the ideas aren't new, so there must already be a solution.
2. Go to the store and fill a shopping bag full of fatty snax, Doritos, Pringles, Kit-Kat bars, Coke, Red Bull, etc.
3. Bring your computer and the bag to the university IT department and beg for help. Let them know that you don't care about the computer (because compared to N years of effort, one computer is nothing), just the contents of the hard drive.
4. While the IT department is working on your computer, go to your departmental administrative office and talk to the secretary in charge of accepting doctoral dissertations. Beg for a one-day extension. Involve your advisor's secretary.
5. Return to the IT department to retrieve your dissertation. Do whatever you were going to do to get it printed.
6. Once everything works out, reward the people who saved your ass with more goodies and/or flowers.
I had looked at the moon forever with the naked eye, and a few times through a couple of lenses, including a backyard telescope. Then, because there was going to be an occultation, I had a chance to look at Jupiter through a pair of binocs. I was blown away that something that close had that much structure -- you could see just scads of moons!
To instil a sense of Science, give the kids tools that they can use well beyond your classes. To this day, when I see Orion in the night sky, he seems like an old friend. When I see Mars waxing and waning, it helps me keep track of time through its two year cycle. Have them look at impressive things, but make sure you give them the tools to find the same objects with their naked eyes. Unless these kids become astronomers, they'll spend at most a few hours behind glass, but will have the rest of their lives to look up at the night sky with their eyes.
One of the coolest things I've learned as a closet amateur astronomer is that the dark side of the moon is illuminated by backwash from the earth, or earthshine. OK, that makes sense if you think about it. But this fact was known almost 100 years ago, and we have a nearly complete record of the brightness of the dark side of the moon since then. That record is important because earthshine is directly related to the amount of terrestrial cloud cover, and that is related to climate. I wish I had known that as a kid, because I'd often wondered why you could see the dark side of the moon better (brighter) on some nights than on others.
For a server running, say, a big web site, or a database, or something else where time is money, and there are a lot of zeros involved, uptime is crucial. When a stock broker's trading floor system goes down, the loss is measured in millions of dollars per second (disclaimer, my brother used to work for a Wall Street firm, his wife used to work for another, and I have two close friends who still work at a third; my estimate is based on things they have told me). Downtime is just not acceptable under some circumstances.
Sure, if my GoDaddy-hosted web site goes off the air for a minute or two while the virtual server gets rekicked, I can't really complain. I end up rebooting my laptop once or twice per week. My desktop gets rebooted maybe twice per year for some hardware update. Users of single-user machines are generally far more tolerant of reboots since, nominally, they are the ones making the decision to reboot. When there are many users, though, rebooting needs to be coordinated, at the very least, so as not to interrupt work in progress. And, as alluded to above, when there's real money involved, sometimes reboots are not ever acceptable.
For you, rebooting might not be evil, but some people do actually depend on high availability of their computers, and some of them are running Linux.
What sort of kickback did the Ubuntu folk get from Google for this? On the face of it, it's a bad decision from the user experience standpoint, as many others have pointed out, because of the potential lack of a network connection. Therefore, it's reasonable to assume that there must be a greater good, at least from the Ubtuntu viewpoint, to screwing their users.
The problem? Most of them (except for PLoS and PLoS ONE) have a very low impact factor because although negative results are important, they aren't sexy in the least. If they were sexy, they would have been published in more mainstream journals. Because publishing a paper requires significant effort, a scientist is unlikely to spend his most precious resource -- time -- publishing a negative result if he can publish a positive one. Positive results get referenced, negative ones, by-and-large, do not. References in important journals lead to advancement as a scientist through grants, promotion, etc. So, unless the result is going to have significant impact -- like contradicting a previous result, or disproving dogma -- there's little motivation for a scientist to expend the effort to write up and publish a negative result, rather than do more research.
It also might not take millions of dollars to do. This could potentially be solved by someone in their garage.
No, that's just not the case. It will take millions of dollars and lots of equipment and infrastructure. We're not talking about technology, we're talking about biology. There are already hundreds if not thousands of people working on the problem (I'm among them). The limiting factors are not the power of our computers, or the whizziness of our mechanical machines, but the understanding of (a) how we can make permanent high-fidelity implants in the brain that do not pose an undue risk to the health of the patient, (b) what, exactly, the language for communicating through these implants is. While the last 100 years has seen tremendous, fantastic progress in understanding the brain, we are still pretty much in the dark as to the fine details, and it's the fine details that matter for a machine-brain interface. Fortunately, recent technological advances (two photon microscopy coupled with ultra-high resolution 3d tissue reconstruction) are going to give us a huge push toward understanding the details in the next few years.
Like I said, I work in the field. To do a very small -- SMALL -- experiment with only half a dozen volunteers who will have a temporary brain implant for two weeks, the non-recoverable costs are about $500,000. That's just for the hospital stays, the costs of the operating room, and paying support staff and the like, and assumes that the surgeon's time is donated, along with all of the important hardware. Remember, this is actually brain surgery. And yes, I have that cost baked into my budget.
My understanding -- as a complete outsider to the field -- is that a lot of the elements are already there.
My understanding -- as an insider in the field -- is that you are correct. I work in the field of visual prosthetics. There are Phase II clinical trials underway for visual prostheses based on retinal stimulation, and a handful of researchers, like myself, who are looking at alternate approaches that include a more direct brain interface. To create a crude machine-brain visual interface, you need: (1) a digital imaging device, like a web cam, (2) a means to translate the image into the neural signal, like a wearable computer, (3) a computer-controlled multi-channel stimulator, like are used for cochlear implants, (4) a brain electrode, like are used to treat Parkinson's disease through Deep Brain Stimulation, or are used on the cortical surface to treat epilepsy. The parts are all there; it's really just a matter of integration, optimization, and getting FDA approval to try it in blind volunteers.
Here's the actual retraction, rather than reporting on reporting on the retraction:
The Lancet, Early Online Publication, 2 February 2010 doi:10.1016/S0140-6736(10)60175-7
Retraction—Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children
The Editors of The Lancet
Following the judgment of the UK General Medical Council's Fitness to Practise Panel on Jan 28, 2010, it has become clear that several elements of the 1998 paper by Wakefield et al(1) are incorrect, contrary to the findings of an earlier investigation.(2) In particular, the claims in the original paper that children were "consecutively referred" and that investigations were "approved" by the local ethics committee have been proven to be false. Therefore we fully retract this paper from the published record.
References
1 Wakefield AJ, Murch SH, Anthony A, et al. Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. Lancet 1998; 351: 637-641 2 Hodgson H. A statement by The Royal Free and University College Medical School and The Royal Free Hampstead NHS Trust. Lancet 2004; 363: 824.
Cutting boards made of glass are quite common. To be fair, though, I believe they are of somewhat different composition.
And they're thick. Scratch the surface, and there is still glass beneath. With the waterproof coating, once the silica layer has been breached, all of the nice properties are lost and, as the great grandparent posting suggests, the coating will likely start to flake off. While the waterproofing characteristics are very interesting, I'm betting there is a lot of hard work left to make the coating durable enough for an industrial food processing setting.
If it's that thing of a layer, wont it be prone to breaking off and becoming airborne? Sounds like silicosis-fun-times to me.
Yeah. Silica-based glass is not very hard. Although this coating is reportedly flexible, I'm betting that it will be readily breached with a sharp edge, so that the example application, on food processing surfaces, at least ones that come in contact with knives, tools and containers, won't be that useful. Stainless steel works by more-or-less the same idea (a thin, hard oxide forms at the surface), except that it has the advantage that when -- not if -- the oxide layer is damaged, a new one automatically forms.
I wonder if this could help make graffiti removal easier. Spray this on a clean road sign, and then just wash it with water if it gets tagged. Sure could help new drivers in Los Angeles.
The New York Subway system has been reportedly using teflon coating for exactly this reason for a very long time. Urban pinheads -- ah, I mean artists -- determined to make their mark despite the paint-shedding properties of teflon discovered that you can scratch the surface instead, thus creating what's called scratchiti, a horrible, far more defacing version of graffiti.
Slashdot Mirror
Solder is already conductive, so the eddy current losses won't be localized in the iron particles. Further, copper traces are even more conductive.
This must be based on the hysteresis losses in the iron B-H curves. That means he's probably got a very high frequency magnetic field generator that he's using to heat up the iron. Seems like a simple principle.
Could also be from friction heating as the iron particles get moved around within the solder paste.
If the soldering was done using this technique your chips would never get a good solder joint in the first place. Ever "cold soldered" (not heated a joint up enough)? Yeah, it doesn't work well - the solder doesn't adhere. Now, you can make up for this by heating the solder even hotter so it spreads its heat to the joint - but then you may as well have soldered the normal way.
Wow, talk about not getting it. The iron particles are inductively heated through an alternating magnetic field. They heat the tin/lead alloy through intimate contact. Heating is localized because (a) the iron particles are only in the solder mixture and (b) a good soldering design will have focused magnetic fields or magnetic shielding to ensure that heating is approximately limited to where it is needed. Eventually, the solder will reach the normal melting temperature but since heating is applied inductively (to which the iron particles are far more sensitive than most everything else in an assembled PCB), the heating is localized to where it is needed, rather than globally applied as is the case with solder wave bath, reflow ovens, and vapor-phase reflow, or semi-locally as is the case with hot air streams.
The interesting part of this technology will be to understand how it differs in implementation and efficiency from a decent focused beam IR method since it would seem to have the same advantages of localized heating.
"the researchers say that by varying electric current to a secured computer"...
Um, if they have physical access to the computer (in order to monkey with the power), why would it be considered secure?
The faults described by the paper are so ... what's the word ... specialized that it challenges believability. Not only does the attacker have to have physical access -- and likely pretty good physical access -- they have to know precisely when the encryption algorithms are being performed so that the faults can be induced then and only then otherwise the operation of the computer will be compromised. Furthermore, the faults must be induced at a reasonable, but not too great, rate, and at randomly varying times in the computation, so as to explore the full error space and have insight into the keys. And the computations have to be repeated MANY times over in order to extract enough information. So, not only do attackers have to know exactly, to the microsecond, when the system under attack is computing the RSA algorithm, they also have to be able to vary the voltage to the CPU. Their physical proof of concept, as much as it is described in the paper, is contrived. Their assertion that the technique does not require physical access is wholly unsupported. Color me skeptical. Anyone with this level of access is going to be able to do more than trigger faults.
The paper asserts that the probes can be done without leaving any trace. I don't know about the authors, but the voltages on my computers are monitored by software and excursions logged so that I can know if/when there are problems. Since the RSA-breaking technique requires substantial exploration of the response to voltage tweaks, it is likely to be detected by a decent monitoring program.
Finally, the PDF does not carry any publication information suggesting strongly that it describes work that is not peer-reviewed. It is shoddy science to bypass peer review and release to the general public.
is anyone shocked this is coming from the not quite all there department of kdawson?
Note to self: stop reading Slashdot when kdawson is editing because the signal-to-silliness ratio goes to hell.
We're also assuming that the cost given hasn't already taken that into account. This article is pure marketing, you can expect them to use every trick in the book.
And that eBay, etc., didn't get highly preferred pricing in exchange for being able to state that they are early-adopting customers saving big bucks.
(I'm just waiting for the Google press release stating that they've received NRC approval to construct a nuclear plant.)
You sound like a quick study. Go buy a copy of "Introduction to Algorithms" by Cormen, Leicerson, and Rivest (yes, *that* Rivest). It's a very, very good book. But, make no mistake, any book on algorithms that is easy is a bad one, and this one, while clear, is difficult because there are some algorithms, some very important ones, that are hard to understand. Pick up a used copy from Amazon.
Your brain works very hard to integrate vision from the two eyes. That said, we can probably guess how a single-eyed prosthesis user would feel by comparing with how sighted people who have significantly different vision in their two eyes feel. Drawing from personal experience, one of my eyes has much sharper vision than the other, and when I'm not wearing my glasses, I can tell that most of my vision comes from the better eye. But, importantly, not all of it. Depth perception, for example, uses both eyes, and, again personally, if I close my poorer eye, there is a large difference.
So, my best guess is that someone with a VGA-grade implant on one side and a normal eye on the other would have better vision than someone with only a single functioning eye, but most of the visual experience would still come from the healthy eye.
I work in Visual Prosthetics.
Here's the thing with any sort of augmented vision: there's no way you can justify the risks of implantation when a fully external device that shows whatever mapped, morphed, or manipulated version of vision will work as well or better.
If you have normal sight, or even nearly normal sight, then why have an implant that carries significant risk, will be large and potentially painful for some time to come, will require frequent recharging, will be expensive as getout, when you can put on a special pair of glasses with a heads-up-display that does more? Telescopic vision, IR, UV, macroscopic, x-ray, edge enhanced, color shifted, depth enhanced, whatever you can think of, it is easier to do it with a head-worn high-tech display that you can take off at will.
In contrast, having an implant means -- for any kind of implant that is under current consideration -- fixed resolution, and, unless you're willing to undergo significant, expensive surgeries, many of the interface parameters will be technologically fixed. Yes, there's a lot you can do with reprogramming, but it's essentially impossible to change the stimulating electrodes and their drivers.
Trust me, you do not want a visual prosthesis unless you need one. The normal visual system, enhanced with purely external devices, will always be better.
Any visual implant that is currently under discussion
First of all, 30-40,000 lines of code is not lots of code. Try, 250,000 of code.
Depends on the language. For most of the popular languages (C / C++, Lisp, Scheme, Perl, Python, PHP, Ruby, Fortran, Matlab, Ada, ALGOL, whatever), you are spot-on. For languages that have higher density (APL primary among them, but I might also include hand-written assembler), 30-40k lines is seriously daunting.
Hello, kdawson.
Note to self: stop reading Slashdot when kdawson is editing because the signal-to-silliness goes to hell.
The difference between bands like Girl Talk who sample music to create new pieces, and someone copying someone else's words into a paper they're writing, is that Girl Talk doesn't claim to have made the samples. One of the aspects of why plagiarism is seen as wrong is because you're taking credit for someone else's work. When you're sampling music, you're crediting them.
Agreed. Often, sampled sounds are clearly recognizable as they are intended to bring an association from the original material: take something everyone knows, twist it, shuffle with other sounds, and make something new. The artists are not trying to hide the origin of their samples, but paying homage to them. Indeed, there is a long-standing history of one artist performing works by another, adding their own touch to the music. Furthermore, when that happens and the second artist makes money at it (sometimes even when they don't), they have to pay royalties to the first artist.
Unattributed lifting without manipulation is not the same. Didn't we just have an article earlier today about cheating in CS classes at Stanford? It's not just illegal, it's unethical.
Just make the punishment for cheating sufficiently harsh. You cheat.. you get kicked out. Simple.
I dunno about Stanford but when I went to school my CS classes (especially the earlier ones) were huge. I never met most of my classmates. I would be *extremely* pissed off to have my academic standing affected by someone else's cheating.
The summary omits the fact that raising the weight of the final is not the ONLY action that happens. The student who is caught is at least reprimanded, if not much worse. The point of the novel mechanism is that everyone else in the class (even if the single individual is expelled) is also punished by re-weighting the grade evaluation to more heavily favor the final exam. Furthermore, that at each instance of cheating, there is another adjustment. Presumably peer pressure or some fear of potential guilt is supposed to act as a deterrent, but I don't see that. It's not an effective motivation for someone who is likely to cheat.
When I was at MIT, teaching the core EECS classes (6.001 through 6.004) late 80s and early 90s, cheating was taken *very* seriously. The very best a student could hope for is to fail the class (I never heard of such a case, though). Nominally it meant leaving the department (with a forced change of majors), if not expulsion from the university. That said, since the punishment was so harsh, and since cheating is difficult to prove in general, students whom we suspected of cheating were usually given a stern warning and put on informal probation. Still, there were disciplinary committee meetings from time to time to deal with students who were provably shown to have cheated. The chair of the committee for many years was a family friend --- and, boy, he was tough. You did not want to be even accused of cheating.
I am very careful to avoid viruses and other malware (always was when i was healthy and Win32 was only a secondary OS for me then). But to be stabbed in the back would be utterly devastating to me. It could be weeks before I could get qualified help (Nerd Herd, etc. need not apply).
I work in the field of prosthetics (although on the sensory, rather than motor side), so I am vitally aware of the importance of interfaces like the one you are using.
That said, since things like bad patches, viruses, trojans, hardware failures, and the like do happen, no matter how careful you might be, perhaps you should think about having a spare, secondary system, in case your primary system fails for some reason. Also, it might be a good idea to implement some kind of automated backup system so that you can roll back to the previous known-good configuration if something goes wrong. Unfortunately, I'm not qualified to make suggestions for how that might work with Windows XP, but the ideas aren't new, so there must already be a solution.
Assuming this isn't a troll --
1. Sit. Down. Breathe.
2. Go to the store and fill a shopping bag full of fatty snax, Doritos, Pringles, Kit-Kat bars, Coke, Red Bull, etc.
3. Bring your computer and the bag to the university IT department and beg for help. Let them know that you don't care about the computer (because compared to N years of effort, one computer is nothing), just the contents of the hard drive.
4. While the IT department is working on your computer, go to your departmental administrative office and talk to the secretary in charge of accepting doctoral dissertations. Beg for a one-day extension. Involve your advisor's secretary.
5. Return to the IT department to retrieve your dissertation. Do whatever you were going to do to get it printed.
6. Once everything works out, reward the people who saved your ass with more goodies and/or flowers.
I had looked at the moon forever with the naked eye, and a few times through a couple of lenses, including a backyard telescope. Then, because there was going to be an occultation, I had a chance to look at Jupiter through a pair of binocs. I was blown away that something that close had that much structure -- you could see just scads of moons!
To instil a sense of Science, give the kids tools that they can use well beyond your classes. To this day, when I see Orion in the night sky, he seems like an old friend. When I see Mars waxing and waning, it helps me keep track of time through its two year cycle. Have them look at impressive things, but make sure you give them the tools to find the same objects with their naked eyes. Unless these kids become astronomers, they'll spend at most a few hours behind glass, but will have the rest of their lives to look up at the night sky with their eyes.
One of the coolest things I've learned as a closet amateur astronomer is that the dark side of the moon is illuminated by backwash from the earth, or earthshine. OK, that makes sense if you think about it. But this fact was known almost 100 years ago, and we have a nearly complete record of the brightness of the dark side of the moon since then. That record is important because earthshine is directly related to the amount of terrestrial cloud cover, and that is related to climate. I wish I had known that as a kid, because I'd often wondered why you could see the dark side of the moon better (brighter) on some nights than on others.
For a server running, say, a big web site, or a database, or something else where time is money, and there are a lot of zeros involved, uptime is crucial. When a stock broker's trading floor system goes down, the loss is measured in millions of dollars per second (disclaimer, my brother used to work for a Wall Street firm, his wife used to work for another, and I have two close friends who still work at a third; my estimate is based on things they have told me). Downtime is just not acceptable under some circumstances.
Sure, if my GoDaddy-hosted web site goes off the air for a minute or two while the virtual server gets rekicked, I can't really complain. I end up rebooting my laptop once or twice per week. My desktop gets rebooted maybe twice per year for some hardware update. Users of single-user machines are generally far more tolerant of reboots since, nominally, they are the ones making the decision to reboot. When there are many users, though, rebooting needs to be coordinated, at the very least, so as not to interrupt work in progress. And, as alluded to above, when there's real money involved, sometimes reboots are not ever acceptable.
For you, rebooting might not be evil, but some people do actually depend on high availability of their computers, and some of them are running Linux.
What sort of kickback did the Ubuntu folk get from Google for this? On the face of it, it's a bad decision from the user experience standpoint, as many others have pointed out, because of the potential lack of a network connection. Therefore, it's reasonable to assume that there must be a greater good, at least from the Ubtuntu viewpoint, to screwing their users.
This idea was already executed a while ago by the Journal of Negative Results in Ecology and Evolutionary Biology, the Journal of Negative Results in BioMedicine, the Journal of Negative Results in Speech and Audio Sciences and probably a few others that Google will help you find, just as it helped me find. But, as I recall, even PLoS had publishing negative results in its charter and specifically PLoS ONE encouraged them, being all-inclusive.
The problem? Most of them (except for PLoS and PLoS ONE) have a very low impact factor because although negative results are important, they aren't sexy in the least. If they were sexy, they would have been published in more mainstream journals. Because publishing a paper requires significant effort, a scientist is unlikely to spend his most precious resource -- time -- publishing a negative result if he can publish a positive one. Positive results get referenced, negative ones, by-and-large, do not. References in important journals lead to advancement as a scientist through grants, promotion, etc. So, unless the result is going to have significant impact -- like contradicting a previous result, or disproving dogma -- there's little motivation for a scientist to expend the effort to write up and publish a negative result, rather than do more research.
It also might not take millions of dollars to do. This could potentially be solved by someone in their garage.
No, that's just not the case. It will take millions of dollars and lots of equipment and infrastructure. We're not talking about technology, we're talking about biology. There are already hundreds if not thousands of people working on the problem (I'm among them). The limiting factors are not the power of our computers, or the whizziness of our mechanical machines, but the understanding of (a) how we can make permanent high-fidelity implants in the brain that do not pose an undue risk to the health of the patient, (b) what, exactly, the language for communicating through these implants is. While the last 100 years has seen tremendous, fantastic progress in understanding the brain, we are still pretty much in the dark as to the fine details, and it's the fine details that matter for a machine-brain interface. Fortunately, recent technological advances (two photon microscopy coupled with ultra-high resolution 3d tissue reconstruction) are going to give us a huge push toward understanding the details in the next few years.
Like I said, I work in the field. To do a very small -- SMALL -- experiment with only half a dozen volunteers who will have a temporary brain implant for two weeks, the non-recoverable costs are about $500,000. That's just for the hospital stays, the costs of the operating room, and paying support staff and the like, and assumes that the surgeon's time is donated, along with all of the important hardware. Remember, this is actually brain surgery. And yes, I have that cost baked into my budget.
My understanding -- as a complete outsider to the field -- is that a lot of the elements are already there.
My understanding -- as an insider in the field -- is that you are correct. I work in the field of visual prosthetics. There are Phase II clinical trials underway for visual prostheses based on retinal stimulation, and a handful of researchers, like myself, who are looking at alternate approaches that include a more direct brain interface. To create a crude machine-brain visual interface, you need: (1) a digital imaging device, like a web cam, (2) a means to translate the image into the neural signal, like a wearable computer, (3) a computer-controlled multi-channel stimulator, like are used for cochlear implants, (4) a brain electrode, like are used to treat Parkinson's disease through Deep Brain Stimulation, or are used on the cortical surface to treat epilepsy. The parts are all there; it's really just a matter of integration, optimization, and getting FDA approval to try it in blind volunteers.
Here's the actual retraction, rather than reporting on reporting on the retraction:
The Lancet, Early Online Publication, 2 February 2010
doi:10.1016/S0140-6736(10)60175-7
Retraction—Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children
The Editors of The Lancet
Following the judgment of the UK General Medical Council's Fitness to Practise Panel on Jan 28, 2010, it has become clear that several elements of the 1998 paper by Wakefield et al(1) are incorrect, contrary to the findings of an earlier investigation.(2) In particular, the claims in the original paper that children were "consecutively referred" and that investigations were "approved" by the local ethics committee have been proven to be false. Therefore we fully retract this paper from the published record.
References
1 Wakefield AJ, Murch SH, Anthony A, et al. Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. Lancet 1998; 351: 637-641
2 Hodgson H. A statement by The Royal Free and University College Medical School and The Royal Free Hampstead NHS Trust. Lancet 2004; 363: 824.
Cutting boards made of glass are quite common. To be fair, though, I believe they are of somewhat different composition.
And they're thick. Scratch the surface, and there is still glass beneath. With the waterproof coating, once the silica layer has been breached, all of the nice properties are lost and, as the great grandparent posting suggests, the coating will likely start to flake off. While the waterproofing characteristics are very interesting, I'm betting there is a lot of hard work left to make the coating durable enough for an industrial food processing setting.
If it's that thing of a layer, wont it be prone to breaking off and becoming airborne? Sounds like silicosis-fun-times to me.
Yeah. Silica-based glass is not very hard. Although this coating is reportedly flexible, I'm betting that it will be readily breached with a sharp edge, so that the example application, on food processing surfaces, at least ones that come in contact with knives, tools and containers, won't be that useful. Stainless steel works by more-or-less the same idea (a thin, hard oxide forms at the surface), except that it has the advantage that when -- not if -- the oxide layer is damaged, a new one automatically forms.
I wonder if this could help make graffiti removal easier. Spray this on a clean road sign, and then just wash it with water if it gets tagged. Sure could help new drivers in Los Angeles.
The New York Subway system has been reportedly using teflon coating for exactly this reason for a very long time. Urban pinheads -- ah, I mean artists -- determined to make their mark despite the paint-shedding properties of teflon discovered that you can scratch the surface instead, thus creating what's called scratchiti, a horrible, far more defacing version of graffiti.
No, they didn't. There weren't any locked-down boxes when Apple built itself. Apple got into the locked-down box thing with the original Macintosh.
They were called Mainframes.