Yeah, I love used book stores. I've never had much luck with libraries. The hours in my area aren't that great and more importantly, the amount of time I have to read can vary wildly, so its nice not have a deadline for when I have to finish/renew the book.
Now if only I could find a used bookstore that had a good selection of non-fiction audio-books, I would be in heaven. I used the use library for that but I've gone through all the interesting titles at my local branch. I can request stuff from nearby branches, but if I look for something specific, they usually don't have it, and the website is pretty much useless for browsing. It only shows like 8 entries per page, and it shows every copy in all the libraries in the area, so you get page after page just listing the same title over and over again.
On my system it seems to get the commercials in the middle of the show alright, but does a very poor job with false positives at the end. Pretty much anytime a show has a short extra scene at the end, MythTV will skip it, jumping back to the main menu, since the show is done. Then you have to reopen the show disable auto-skip, fast-forward to the end of the show to see if you missed anything or not. I find that to be a bigger hassle than having to press the skip button when a commercial comes on, so I just leave the auto-skip feature off.
Oh, one other caveat, is that when you quantize the blurred image (assign each pixels a discrete, say 24-bit, value), you will also loose some information.
Furthermore, I should mention that given the size of peoples faces, and the amount of blur that Google is likely to use, the entire blurred section will be near enough to the edge to loose significant information, so it is unlikely that much recovery will be possible.
So, nothing I said was really applicable to this situation:) I was just surprised myself to learn that a blurred image is not the same as a lower resolution image, and so I thought I'd share.
You can't add pixels that aren't there, and an out of focus picture is effectively a lower resolution. No, it isn't. Think about an unfocused camera - all the light is still hitting film/CCD, it is just spread out. So from an information theory point of view you haven't lost any data, you just put it into another form. If you consider what would be a single point of light, the energy in that point is spread out in a normal distribution (aka bell curve, aka gaussian). So the blurred image is just all these Gaussians functions overlayed on top of each other. Computer blurring algorithms do pretty much the same thing.
From a signal processing perspective, this is the same as convolving with a Gaussian. And if you take the Fourier transform of that blurred image, you get the transform of the image multiplied by the transform of the Gaussian (which is just another Gaussian). From there all you have to do is divide by this Gaussian, take the inverse transform, and walla, you have the desired non-blurred image. This is called a deconvolution, and I've written code to do this for an image processing class.
There are some caveats. You have to guess how blurred the image is - what focal length is and what not. Noise and compression can kill you, so you need to filter those out first (or limit your deconvolution filter to low frequency content). In addition at the edges of the image (or edge of the blur boundary) information is genuinely lost as the gaussian falls outside the boundary and is discarded.
Focus Magic is a commercial package that refocuses blurred images, and they have some interesting sample photos.
Yeah, it makes a lot more sense after reading SiliconEntity's post. They wrote their routine to be called iteratively - it takes the previous value of a buffer and a random source and combines the two to create a more random value. They didn't bother to initialize the buffer the first time through since it wouldn't hurt to have garbage data in there. The "fix" made it so that the current value of the buffer was never used when getting the next value, so the randomness of the generator was only as good as the last data point from the random source, which was significantly weaker than the original design.
This is just strange. From the bug report it sounds like the only thing they did was remove the code which read into uninitialized memory, and fed the value into the random number generator. If this really was the only seed used by the RNG that seems like a pretty poor design decision on the part of OpenSSL developers to begin with.
On the other hand it is possible that the Debian maintainer didn't understand what he was doing, and his changes did more than remove that read into uninitialized memory. In either case it I can't imagine why someone would modify a RNG or any other cryptographic code without checking with the original author first before distributing it. That is just unacceptable.
This isn't an about-face. The fact of the matter is that NASA has been required by law to contract out nearly all of its launch production, facilities and maintenance for a long time now. All of its probe launches are done with Boeing and Lockheed rockets. NASA has also gone out of its way to offer contracts to the smaller private companies from the vary beginning of the new launcher plans. If you look at the contracts they almost appear to be intentionally catered to the strengths of these specific start-ups.
This isn't about public vs private - it is about NASA's desire to stop being dependant on a small number of large aerospace corporations. It is about their desire for space exploration grow in anyway possible. Everybody who works there wants to see SpaceX, t-Space, and the others succeed, as much as the folks here do.
I've been looking into getting one of these, just because I am sick of my magnetic strip getting screwed up a month after I get the card and then having to request another one.
I've heard that at least some of the touch-and-pay systems aren't just passive RFID, but use a challenge-response system which would actually more be secure than a credit card, since the merchant / snooper never sees your card number. If I can verify this then I definitely will be getting one.
So sure, maybe theoretically someone could sit next to me on the bus, and gather enough CR samples to recover the key. But considering all the places that I have used my CC number online, and who knows how many of the merchants store that info, I think that a brute-force cryptanalysis of my keyfob is the least of my concerns.
And besides, since it is a real credit card (not a debit card) the CC company will pay for any fraud anyway, so making sure the system is secure is their problem not mine.
My reaction at VIA was more of befuddlement than anything else. I mean they went to through all the effort to write these drivers, and they were nice enough to make the source available, but then there was just a complete breakdown of communication when it came to letting people do the last 10% that was necessary to make the drivers useful.
Like at first they had a binary download, but then to get the source you had to sign up and be a "serious" open source developer. I just wanted to get the source so I could recompile it with my kernel (which was not compatible with their binary), I filled out their form and then never heard back from then. They would release source saying it was under a certain license, but when the developers of the fork would look through it they would find all sorts of other claims of proprietary license in and accompanying the code, sometimes by third parties, and weren't sure which to believe. Inquires to VIA about such things often seemed to disappear into a blackhole.
I don't know what was going on inside VIA - if they hadn't decided whether they wanted to maintain the software themselves or if they wanted the community to do it, or if the development work was being done at VIA Taiwan and they hadn't given anyone at VIA America authority to handle relations with developers, or what, but it was a completely bungled arrangement.
I have no problem with companies depending on the community to maintain the drivers - that can be a very productive arrangement for everyone involved - but communication is absolutely essential for it to work. VIA is an interesting company, and I think they are in a unique position to benefit from a closer connection with the open source community - the encryption features in their processors are a good example of where they have done things right in the past. Hopefully their video/chipset drivers will see the same success in the future.
Via has "supported" linux in the past, and all it amounted to was dumping some poorly written and undocumented code, and then not doing anything to maintain the code themselves, and not accepting accepting patches, not responding to queries for documentation/clarification from those that wanted to improve the drivers themselves.
I hope they are doing the right thing this time, and will gladly praise them if they do, but I can understand why some people would be skeptical until then.
I've played with one of those - sadly it will only bring you beer if your pre-program the path to the bar and back, and have a bar-tender to put the beer in it's holder, and send it off on it's way. It also goes eats through batteries like crazy.
But it does an awesome little dance to the cantina song if you ask it to! And it can play tag.
One of the projects on my list of things to do (most of which will never happen till I retire) has been to build a R2D2 karaoke machine, complete with lyrics projector, and lightsaber^W microphone ejecting action! And a gin and tonic dispenser, of course.
I wasn't thinking of using anywhere near as good of components, though. This really is impressive.
I personally think it is because Sun insists on tying OO.o together with the JRE. You can disable the JRE: Tools > Options > Java > disable. Only a few components use it. Doing so does improve the start up time quite a bit, but I haven't seen any difference whatsoever with runtime performance, so I don't think the JRE is to blame there.
There are reasons that the Aptera has three wheels and not four, and they are entirely regulatory and not technical. Part of that is just the red tape required to prove that the car meets the requirements, but not even Aptera claims that they meet or exceed all the government requirements for passenger vehicles, just the ones they considered most important for safety.
I have little reason to disbelieve auto manufacturers when they say it is impossible to build a 100 MPH automobile, according to the legal definition of automobile. Not that it matters to me at all whether the vehicle I buy is technically classified as an automobile or not.
Well, yeah, if you use a refill kit then the amount of ink in the cartridge is entirely dependent on how much you put in, so what I said wouldn't even make sense. This was talking about buying third party cartridges. They did look at refill kits. I don't remember what the results were - I think they had quality issues with the ones they tested, but I've let my membership lapse, so I can't check.
I spent the majority of my childhood until I was 18 picking rock and bailing hay on a farm. You think you're in a tedious, repetitive and boring job? Actually, I've worked mindless tedious repetitive manual labor jobs in the past, and the truth is I much prefer them to IT work that is every bit as uninteresting and tedious, but which requires more thought. I don't mind hard work, and the act of having to stay mentally focused on something that bores me for 10 hours at a stretch is more draining than physical labor. Fortunately, only about half my time is spent on that sort of work, and the other half (not to mention pay) make up for it.
Hi, this might be an old issue, but stale data has a way of sticking around on the internet, and I've had trouble digging up a definitive answer when looking into this recently. Perhaps there are folks here that are actively using XFS and/or follow the kernel mailing lists more closely than I, so I'll ask - Have all the 4k stack issues with XFS been resolved?
Well, because it wasn't a rational or conscious choice. It was just that as time progressed, I had fewer and fewer reasons to lead me to think that god existed, and what faith I had left pretty much evaporated as I realized more and more that the things that I valued in religion had nothing to do with god. I can't decide to believe in god now any more than I could be reasoned into disbelieving him when I was a christian.
I do agree with you, and it is a big reason why I am not an "evangelical" atheist (the other being that I am as about as non-confrontational as you can get). I think faith can be a very beneficial thing, and as long as you aren't hurting anyone and aren't trying to pass your opinions off as science, I have no problems with it.
Another issue that I didn't see after skimming this article, as well as part one of the series, is that many of the third party ink cartridges don't contain as much ink as the name brand. Consumer Reports tested a bunch of inks, and found that most of the cheapest inks were actually more expensive per page than the brandname ink. Which inks fared well varied from printer to printer.
Unfortunately, it looks like the full article is only available to subscribers, and there are just a few short blubs summarizing the results available to everyone.
I've been trying to figure out why those darn kids keep attacking my gazebo. They've obviously been playing too much D&D! When will they learn to separate games from reality?
That is actually a very interesting conclusion. It really is true that just "thinking happy thoughts" does make you more likely to achieve them, but for naturally cynical people this is easier said than done. If I try to sit there and talk myself up, it usually backfires because the cynic in my just can't help poking holes in everything I'm saying and I end up talking myself down instead. It was easier when I believed in god, because it acted as a way of suspending disbelief.
What they are saying here is that just the act of imagining yourself as being better, even in the context of an artificial world that has no bearing on reality, has some of the same effects of imagining yourself being a better person in the real world. That seems like it would be a very useful technique. It is probably also part of the reason that MUDs can help asocial people be more social. I had always assumed that it was just because it sidestepped one's fear of external expectations/judgment, but the fact that it also bypasses internal judgment as well is something I hadn't thought of. Oh, and playing MUDs will help me be a better person IRL, so there naysayers:)
The energy in question is coming from the sun, and was going to enter the biosphere anyway. Some of it would have, but some of it would also be reflected. On average, the earth has an albedo (fraction of light reflected) of about 37%.
To a certain extent, the effect will be the exact opposite of what you are thinking, as the sunlight would have most assuredly heated the land, sea and air, but beamed down to the electrical grid, it will be stored in other forms In the long run it will all be converted to heat. Furthermore, there are very few uses of electricity that result in storage as potential energy of some form. Looking at California data, the Residential, Commercial, TCU and Streetlights will all be AC/lights/electronics which will be converted to heat immediately. The mining sector and industrial sectors will result in some potential (lifted mass, increased chemical potential of stable compounds, etc) But the machines they use to do this are not very efficient. Even if we are very generous and say that half of their energy is used for these purposes, and those machines are 50% efficient, that gives 5% of total energy use being converted to potential form.
So if the energy efficiency of the panel/beam is greater than about 100%-37%-5% = 58%, then this system will result in more heat than would normally occur from the sunlight.
Of course, even if it does significantly increase the amount of heat generated for the fraction of sunlight that it captures, that is still a tiny fraction of the sky that is covered, and the net result will be completely negligible compared to just about anything else.
Hmm, it look like you are right. Last time I looked (which was a while ago) the drivers were split into 70 and 75 series, and I was told that the 70 series only worked with 2.4 kernels.
Note to self: Must remember that this new form does not work in konqueror - 3rd post I've had to retype today.
Also, if you want 100% rock solid drivers that are supported out of the box and cream the competition in terms of stability (speaking about Linux here), you buy an Intel GPU. I wouldn't go that far. I've had stability issues with my intel graphics. Some OpenGL screensavers and some games running under Wine will crash or lockup X, regardless of what settings I use in my xorg.conf (XAA vs EXE, Composite on/off). Furthermore, several extentions (like composite) that are fairly stable with NVidia drivers are still buggy as hell with the intel drivers.
I never had any stability issues whatsoever with the last NVidia card I bought. Then again, that card is now useless to me since NVidia stopped releasing accelerated drivers for new kernels, which is why I went with intel last time around.
So yeah, if you discount the market leader in terms of driver stability and volume of sales, and care only about speed then yes, Intel isn't competitive. Which is exactly what dreamchaser said. Intel keeps claiming that the performance of their new chipsets will be competitive and every time it isn't even close. So why should we take what they say seriously? I sure as hell don't.
I don't regret my decision to go with intel, and would do it again if I was buying a computer today. However, the way things are looking, I have more faith in the ATI driver situation improving than in intel ever producing a GPU with good performance.
Slashdot Mirror
Yeah, I love used book stores. I've never had much luck with libraries. The hours in my area aren't that great and more importantly, the amount of time I have to read can vary wildly, so its nice not have a deadline for when I have to finish/renew the book.
Now if only I could find a used bookstore that had a good selection of non-fiction audio-books, I would be in heaven. I used the use library for that but I've gone through all the interesting titles at my local branch. I can request stuff from nearby branches, but if I look for something specific, they usually don't have it, and the website is pretty much useless for browsing. It only shows like 8 entries per page, and it shows every copy in all the libraries in the area, so you get page after page just listing the same title over and over again.
On my system it seems to get the commercials in the middle of the show alright, but does a very poor job with false positives at the end. Pretty much anytime a show has a short extra scene at the end, MythTV will skip it, jumping back to the main menu, since the show is done. Then you have to reopen the show disable auto-skip, fast-forward to the end of the show to see if you missed anything or not. I find that to be a bigger hassle than having to press the skip button when a commercial comes on, so I just leave the auto-skip feature off.
Oh, one other caveat, is that when you quantize the blurred image (assign each pixels a discrete, say 24-bit, value), you will also loose some information.
:) I was just surprised myself to learn that a blurred image is not the same as a lower resolution image, and so I thought I'd share.
Furthermore, I should mention that given the size of peoples faces, and the amount of blur that Google is likely to use, the entire blurred section will be near enough to the edge to loose significant information, so it is unlikely that much recovery will be possible.
So, nothing I said was really applicable to this situation
From a signal processing perspective, this is the same as convolving with a Gaussian. And if you take the Fourier transform of that blurred image, you get the transform of the image multiplied by the transform of the Gaussian (which is just another Gaussian). From there all you have to do is divide by this Gaussian, take the inverse transform, and walla, you have the desired non-blurred image. This is called a deconvolution, and I've written code to do this for an image processing class.
There are some caveats. You have to guess how blurred the image is - what focal length is and what not. Noise and compression can kill you, so you need to filter those out first (or limit your deconvolution filter to low frequency content). In addition at the edges of the image (or edge of the blur boundary) information is genuinely lost as the gaussian falls outside the boundary and is discarded.
Focus Magic is a commercial package that refocuses blurred images, and they have some interesting sample photos.
Yeah, it makes a lot more sense after reading SiliconEntity's post. They wrote their routine to be called iteratively - it takes the previous value of a buffer and a random source and combines the two to create a more random value. They didn't bother to initialize the buffer the first time through since it wouldn't hurt to have garbage data in there. The "fix" made it so that the current value of the buffer was never used when getting the next value, so the randomness of the generator was only as good as the last data point from the random source, which was significantly weaker than the original design.
This is just strange. From the bug report it sounds like the only thing they did was remove the code which read into uninitialized memory, and fed the value into the random number generator. If this really was the only seed used by the RNG that seems like a pretty poor design decision on the part of OpenSSL developers to begin with.
On the other hand it is possible that the Debian maintainer didn't understand what he was doing, and his changes did more than remove that read into uninitialized memory. In either case it I can't imagine why someone would modify a RNG or any other cryptographic code without checking with the original author first before distributing it. That is just unacceptable.
This isn't an about-face. The fact of the matter is that NASA has been required by law to contract out nearly all of its launch production, facilities and maintenance for a long time now. All of its probe launches are done with Boeing and Lockheed rockets. NASA has also gone out of its way to offer contracts to the smaller private companies from the vary beginning of the new launcher plans. If you look at the contracts they almost appear to be intentionally catered to the strengths of these specific start-ups.
This isn't about public vs private - it is about NASA's desire to stop being dependant on a small number of large aerospace corporations. It is about their desire for space exploration grow in anyway possible. Everybody who works there wants to see SpaceX, t-Space, and the others succeed, as much as the folks here do.
I've been looking into getting one of these, just because I am sick of my magnetic strip getting screwed up a month after I get the card and then having to request another one.
I've heard that at least some of the touch-and-pay systems aren't just passive RFID, but use a challenge-response system which would actually more be secure than a credit card, since the merchant / snooper never sees your card number. If I can verify this then I definitely will be getting one.
So sure, maybe theoretically someone could sit next to me on the bus, and gather enough CR samples to recover the key. But considering all the places that I have used my CC number online, and who knows how many of the merchants store that info, I think that a brute-force cryptanalysis of my keyfob is the least of my concerns.
And besides, since it is a real credit card (not a debit card) the CC company will pay for any fraud anyway, so making sure the system is secure is their problem not mine.
My reaction at VIA was more of befuddlement than anything else. I mean they went to through all the effort to write these drivers, and they were nice enough to make the source available, but then there was just a complete breakdown of communication when it came to letting people do the last 10% that was necessary to make the drivers useful.
Like at first they had a binary download, but then to get the source you had to sign up and be a "serious" open source developer. I just wanted to get the source so I could recompile it with my kernel (which was not compatible with their binary), I filled out their form and then never heard back from then. They would release source saying it was under a certain license, but when the developers of the fork would look through it they would find all sorts of other claims of proprietary license in and accompanying the code, sometimes by third parties, and weren't sure which to believe. Inquires to VIA about such things often seemed to disappear into a blackhole.
I don't know what was going on inside VIA - if they hadn't decided whether they wanted to maintain the software themselves or if they wanted the community to do it, or if the development work was being done at VIA Taiwan and they hadn't given anyone at VIA America authority to handle relations with developers, or what, but it was a completely bungled arrangement.
I have no problem with companies depending on the community to maintain the drivers - that can be a very productive arrangement for everyone involved - but communication is absolutely essential for it to work. VIA is an interesting company, and I think they are in a unique position to benefit from a closer connection with the open source community - the encryption features in their processors are a good example of where they have done things right in the past. Hopefully their video/chipset drivers will see the same success in the future.
Via has "supported" linux in the past, and all it amounted to was dumping some poorly written and undocumented code, and then not doing anything to maintain the code themselves, and not accepting accepting patches, not responding to queries for documentation/clarification from those that wanted to improve the drivers themselves.
I hope they are doing the right thing this time, and will gladly praise them if they do, but I can understand why some people would be skeptical until then.
I've played with one of those - sadly it will only bring you beer if your pre-program the path to the bar and back, and have a bar-tender to put the beer in it's holder, and send it off on it's way. It also goes eats through batteries like crazy.
But it does an awesome little dance to the cantina song if you ask it to! And it can play tag.
One of the projects on my list of things to do (most of which will never happen till I retire) has been to build a R2D2 karaoke machine, complete with lyrics projector, and lightsaber^W microphone ejecting action! And a gin and tonic dispenser, of course.
I wasn't thinking of using anywhere near as good of components, though. This really is impressive.
There are reasons that the Aptera has three wheels and not four, and they are entirely regulatory and not technical. Part of that is just the red tape required to prove that the car meets the requirements, but not even Aptera claims that they meet or exceed all the government requirements for passenger vehicles, just the ones they considered most important for safety.
I have little reason to disbelieve auto manufacturers when they say it is impossible to build a 100 MPH automobile, according to the legal definition of automobile. Not that it matters to me at all whether the vehicle I buy is technically classified as an automobile or not.
Well, yeah, if you use a refill kit then the amount of ink in the cartridge is entirely dependent on how much you put in, so what I said wouldn't even make sense. This was talking about buying third party cartridges. They did look at refill kits. I don't remember what the results were - I think they had quality issues with the ones they tested, but I've let my membership lapse, so I can't check.
Hi, this might be an old issue, but stale data has a way of sticking around on the internet, and I've had trouble digging up a definitive answer when looking into this recently. Perhaps there are folks here that are actively using XFS and/or follow the kernel mailing lists more closely than I, so I'll ask - Have all the 4k stack issues with XFS been resolved?
Well, because it wasn't a rational or conscious choice. It was just that as time progressed, I had fewer and fewer reasons to lead me to think that god existed, and what faith I had left pretty much evaporated as I realized more and more that the things that I valued in religion had nothing to do with god. I can't decide to believe in god now any more than I could be reasoned into disbelieving him when I was a christian.
I do agree with you, and it is a big reason why I am not an "evangelical" atheist (the other being that I am as about as non-confrontational as you can get). I think faith can be a very beneficial thing, and as long as you aren't hurting anyone and aren't trying to pass your opinions off as science, I have no problems with it.
Another issue that I didn't see after skimming this article, as well as part one of the series, is that many of the third party ink cartridges don't contain as much ink as the name brand. Consumer Reports tested a bunch of inks, and found that most of the cheapest inks were actually more expensive per page than the brandname ink. Which inks fared well varied from printer to printer.
Unfortunately, it looks like the full article is only available to subscribers, and there are just a few short blubs summarizing the results available to everyone.
I've been trying to figure out why those darn kids keep attacking my gazebo. They've obviously been playing too much D&D! When will they learn to separate games from reality?
That is actually a very interesting conclusion. It really is true that just "thinking happy thoughts" does make you more likely to achieve them, but for naturally cynical people this is easier said than done. If I try to sit there and talk myself up, it usually backfires because the cynic in my just can't help poking holes in everything I'm saying and I end up talking myself down instead. It was easier when I believed in god, because it acted as a way of suspending disbelief.
What they are saying here is that just the act of imagining yourself as being better, even in the context of an artificial world that has no bearing on reality, has some of the same effects of imagining yourself being a better person in the real world. That seems like it would be a very useful technique. It is probably also part of the reason that MUDs can help asocial people be more social. I had always assumed that it was just because it sidestepped one's fear of external expectations/judgment, but the fact that it also bypasses internal judgment as well is something I hadn't thought of. Oh, and playing MUDs will help me be a better person IRL, so there naysayers:)
So if the energy efficiency of the panel/beam is greater than about 100%-37%-5% = 58%, then this system will result in more heat than would normally occur from the sunlight.
Of course, even if it does significantly increase the amount of heat generated for the fraction of sunlight that it captures, that is still a tiny fraction of the sky that is covered, and the net result will be completely negligible compared to just about anything else.
Hmm, it look like you are right. Last time I looked (which was a while ago) the drivers were split into 70 and 75 series, and I was told that the 70 series only worked with 2.4 kernels.
Note to self: Must remember that this new form does not work in konqueror - 3rd post I've had to retype today.
I never had any stability issues whatsoever with the last NVidia card I bought. Then again, that card is now useless to me since NVidia stopped releasing accelerated drivers for new kernels, which is why I went with intel last time around. So yeah, if you discount the market leader in terms of driver stability and volume of sales, and care only about speed then yes, Intel isn't competitive. Which is exactly what dreamchaser said. Intel keeps claiming that the performance of their new chipsets will be competitive and every time it isn't even close. So why should we take what they say seriously? I sure as hell don't.
I don't regret my decision to go with intel, and would do it again if I was buying a computer today. However, the way things are looking, I have more faith in the ATI driver situation improving than in intel ever producing a GPU with good performance.