Agreed, but the 30-inch screens are still among the smallest dot-pitch you get in any mainstream desktop LCD (compare to the madness of 27-inch 1920x1200, for example, or even a 19-inch 1280x1024). A quarter of 2560x1600 is after all 1280x800, and there are supposedly useful laptops sold with 14.1 or 15.4 screens that are still only WXGA.
Long story short, I'm very happy with my dual 3007 setup. I'll probably hang on to them for 2-3 more years, hoping for OLED and/or higher DPI at the next iteration.
Well, that would make Hungarian more relevant again. It doesn't matter whether the language itself can store anything in anything, you probably have a specific idea of what data should go in and out of that black box, and making the wrong assumptions can cause tricky bugs. LPCTSTR will make the people free!
What's the HD broadcaster? I certainly know that many/some digital SD feeds look like total crap compared to DVD (upscaled DVD manages to look nice, some upscaled DVB-T absolutely doesn't). I think the technology is new enough for HD feeds to vary even more.
Well, they should have done that as soon as they saw the first atoms forming, or at least when the stars started. The quantization is very much needed to give us something resembling what we have. Classical mechanics simply didn't cut it.
Detecting a well-implemented simulation is hard. It's like trying to determine for sure that you're in a VM from user mode, when you haven't even been told what a non-VM would look like. Arguments like timing assume that somehow we would still be simulated in real time. While possible, there is no reason to do it that way. If I write an explicit time-stepping scheme in a for loop, the scheme won't be affected if my code is swapped out and then loaded again. I can even hibernate the machine and start it up later, the simulated environment will be totally unaffected.
You are wrong about the chaos theory. Nothing in chaotic systems make them uncomputable in the same sense as, say, the stopping problem. They are unpredictable, in the sense that you need the state to infinite precision and detail (or: all detail and precision there really is) to predict the outcome. I can write code that demonstrates a chaotic system (like Windows Me...). What's hard is to write a chaotic system that exactly matches another chaotic system. The hypothesized guy simulating us has no (for us known) reason to imitate a specific template chaotic system, so he can very well introduce anything here.
This is not an injection bug per se, but more a string parsing bug. Parsing needs to be done as long as not all content is implicitly structured. One point in using XML for anything is to avoid doing any parsing on your own. But, think about it, would you like an e-mail address, URL or file path to be a structued list or XML snippet? And could we be sure that the structure is always the right one, so there will be no need to flatten it and reparse it and get into the same old bugs?
Finite state machines with more than a handful of states are hard (whether implemented explicitly or implicitly). They are harder for some people. We can try to make sure that the reinventing-the-wheel need is limited, but it's oh so surprising how often you want something that's only a bit different (or you find the standard interface to be so clunky that you roll your own *guilty smile*).
Yeah, but the problem is that it's an engineering project headed by scientists. They'll just take their funding and decide that they need some kind of hyperlinked document-management system, and develop a new protocol from scratch, simply because they actually believe that's the best way to do it...
Measuring precision is superior to original manufacture precision. Getting a part that's been verified and validated to be close to spec is a different thing than just being lucky. Whether that precision and lack of imperfections really contribute to the total reliability in a significant way is another issue, but even when parts come off the same factory line, the difference between them can be quite real.
There have already been some browser fixes, mainly triggering cases where characters from different scripts appear next to each other. That certainly breaks some valid cases as well, but I guess it's bearable. (So you can't just switch a single o in some domain name to a Cyrillic o and get it to show almost indistinguishably... or at least that's the idea.)
Activation of the same receptor generally doesn't mean that the complete substance is identical. (If drugs were identical to their native counterparts, the self-regulation would cope much more efficiently and degrade them. On the other hand, it's this similarity and a feedback loop that really causes the addiction...)
Strange, the main point seems to simply be that there are no breeder reactors yet, and that this is mainly due to changed plans and political reasons -- motivations that might change in favor of the technology again. Of course you pile up "waste" if no one wants to use the fuel. The fact that what was previous "only" waste is now a useful product is still important, even if the users are still missing.
The article also touches upon the technological hurdles, but they seem minor.
It's a heavily reduced subset of.NET, so if you (have some tool that) can generate MSIL, you'r basically set. XAML, and the Silverlight subset, can also be generated quite easily. (As far as Microsoft XML formats go, it's not too bad.)
5, Insightful, and off by an order of magnitude. 99 % false positive. I think the intent of any system in this area should be a fairly high percentage of false positives. The best point, as posted by someone in another comment, is however whether this 1 % is higher than the baseline, and by how much. If the profiling would manage a 10:1 concentration of "correct" targets, it might be seen as a success, for example. The presented numbers don't seem to indicate that, though.
I haven't looked too much at the foldable ones, but I can tell you that it is light and they keys just about full-size when projected properly. The size is basically to get the proper distance between the projector and the detector. From a technology standpoint, I see it more as a proof-of-concept, where the projector and detector could be integrated into a cellphone, without really increasing the size that much. Again, the ability to switch between touchpad and keyboard functionality is also kind of neat, something you can't do (on the same surface) with an ordinary foldable keyboard. In the context of this article, it's also interesting to imagine the fixed keyboard being replaced by at least a monochrome variable picture, I think the device have uses a fixed physical stencil.
I got one as a gift from a 10^100 company. I felt quite comfortable with it, and I am a touch typer. I think that it generally just detected hits for normal keys (hence not resting). The real issue for me was that it simply wasn't able to track my normal speed of writing, the specs said it could detect 10 hits per second, and I certainly exceed that in quite long bursts. This is reasonably something that can be improved by better processing. The lack of tactile feedback will of course always be an issue, though.
In fact, I found the "virtual touchpad" mode of the keyboard more useful.
With a static system, that is. CO2 in the atmosphere is reintegrated in biomass and inorganic carbonate. Part of the largest uncertainty in the models (when considering the pure CO2 level, not the climate outcome) is what happens with these processes on a midlong scale.
But, yeah, we've gone from somewhere around 280 ppm to 360 (don't trust these numbers, very approximate out of my a** right now) and we're climbing quickly. The CO2 in the atmosphere has a very clear human impact. Any doubt is centered on the issue whether that's important or not. I think it is.
Slashdot Mirror
Long story short, I'm very happy with my dual 3007 setup. I'll probably hang on to them for 2-3 more years, hoping for OLED and/or higher DPI at the next iteration.
Well, that would make Hungarian more relevant again. It doesn't matter whether the language itself can store anything in anything, you probably have a specific idea of what data should go in and out of that black box, and making the wrong assumptions can cause tricky bugs. LPCTSTR will make the people free!
What's the HD broadcaster? I certainly know that many/some digital SD feeds look like total crap compared to DVD (upscaled DVD manages to look nice, some upscaled DVB-T absolutely doesn't). I think the technology is new enough for HD feeds to vary even more.
Heck, silly me who believed I can use my 15.4" WUXGA laptop...
Hey, I'm sure the rocket scientists at NASA manages to hack uptime, even on vxware.
Sorry, but they landed in January 2004. The GP's point was that they know start their 5th year.
Well, they should have done that as soon as they saw the first atoms forming, or at least when the stars started. The quantization is very much needed to give us something resembling what we have. Classical mechanics simply didn't cut it.
It's just that Ev...Deanna accidentally damaged the isolinear chips handling the security protocols when she tried to grab an apple.
Detecting a well-implemented simulation is hard. It's like trying to determine for sure that you're in a VM from user mode, when you haven't even been told what a non-VM would look like. Arguments like timing assume that somehow we would still be simulated in real time. While possible, there is no reason to do it that way. If I write an explicit time-stepping scheme in a for loop, the scheme won't be affected if my code is swapped out and then loaded again. I can even hibernate the machine and start it up later, the simulated environment will be totally unaffected.
You are wrong about the chaos theory. Nothing in chaotic systems make them uncomputable in the same sense as, say, the stopping problem. They are unpredictable, in the sense that you need the state to infinite precision and detail (or: all detail and precision there really is) to predict the outcome. I can write code that demonstrates a chaotic system (like Windows Me...). What's hard is to write a chaotic system that exactly matches another chaotic system. The hypothesized guy simulating us has no (for us known) reason to imitate a specific template chaotic system, so he can very well introduce anything here.
Finite state machines with more than a handful of states are hard (whether implemented explicitly or implicitly). They are harder for some people. We can try to make sure that the reinventing-the-wheel need is limited, but it's oh so surprising how often you want something that's only a bit different (or you find the standard interface to be so clunky that you roll your own *guilty smile*).
Yeah, but the problem is that it's an engineering project headed by scientists. They'll just take their funding and decide that they need some kind of hyperlinked document-management system, and develop a new protocol from scratch, simply because they actually believe that's the best way to do it...
Measuring precision is superior to original manufacture precision. Getting a part that's been verified and validated to be close to spec is a different thing than just being lucky. Whether that precision and lack of imperfections really contribute to the total reliability in a significant way is another issue, but even when parts come off the same factory line, the difference between them can be quite real.
There have already been some browser fixes, mainly triggering cases where characters from different scripts appear next to each other. That certainly breaks some valid cases as well, but I guess it's bearable. (So you can't just switch a single o in some domain name to a Cyrillic o and get it to show almost indistinguishably... or at least that's the idea.)
Activation of the same receptor generally doesn't mean that the complete substance is identical. (If drugs were identical to their native counterparts, the self-regulation would cope much more efficiently and degrade them. On the other hand, it's this similarity and a feedback loop that really causes the addiction...)
The article also touches upon the technological hurdles, but they seem minor.
It's a heavily reduced subset of .NET, so if you (have some tool that) can generate MSIL, you'r basically set. XAML, and the Silverlight subset, can also be generated quite easily. (As far as Microsoft XML formats go, it's not too bad.)
5, Insightful, and off by an order of magnitude. 99 % false positive. I think the intent of any system in this area should be a fairly high percentage of false positives. The best point, as posted by someone in another comment, is however whether this 1 % is higher than the baseline, and by how much. If the profiling would manage a 10:1 concentration of "correct" targets, it might be seen as a success, for example. The presented numbers don't seem to indicate that, though.
What height are you imagining for a keyboard to support back-projection?
I haven't looked too much at the foldable ones, but I can tell you that it is light and they keys just about full-size when projected properly. The size is basically to get the proper distance between the projector and the detector. From a technology standpoint, I see it more as a proof-of-concept, where the projector and detector could be integrated into a cellphone, without really increasing the size that much. Again, the ability to switch between touchpad and keyboard functionality is also kind of neat, something you can't do (on the same surface) with an ordinary foldable keyboard. In the context of this article, it's also interesting to imagine the fixed keyboard being replaced by at least a monochrome variable picture, I think the device have uses a fixed physical stencil.
In fact, I found the "virtual touchpad" mode of the keyboard more useful.
The only editor you need is emacs.
With a static system, that is. CO2 in the atmosphere is reintegrated in biomass and inorganic carbonate. Part of the largest uncertainty in the models (when considering the pure CO2 level, not the climate outcome) is what happens with these processes on a midlong scale. But, yeah, we've gone from somewhere around 280 ppm to 360 (don't trust these numbers, very approximate out of my a** right now) and we're climbing quickly. The CO2 in the atmosphere has a very clear human impact. Any doubt is centered on the issue whether that's important or not. I think it is.
Finally someone summing it up succinctly.
The buzz are not about good hits, it's about funny ones, like all evil and Microsoft... (Hey, that's how I first heard of Google.)