Let's face it: starting this late and with only a vague idea of hardware and software components required, he isn't going to win anyways. In the best of cases, the answers he gets here will make the difference between having an entry for the competition, and not having one at all.
I have yet to receive a government PDF form here in Canada. Go look on the government of Canada and the provincial (I live in BC) web pages. Pretty much any form you'll ever need to fill in is there as a PDF, and many of them (I'd say about half of the ones I've had to deal with) are instrumented for being filled in electronically. Of course, I you just open them in a reader that does not support forms then you'd never know.
The Evince developers are working on a form filling function for it. So I hope I never have the need to install Acrobat Reader on my home Linux system. I agree that is good news, so this may be an evolving possibility.
At work on XP I use Foxit reader as my Acrobat Reader installation is so fucked up. From the original blog: Foxit is vulnerable as well, although the user is required to interact with the document in order to launch the exploit.
That is one fo the reasons why the government forms are using PDF, I am sure. PDF is an open format, so you can get free software to print it and fill in the form, even if it is cumbersome. Of course you really want to keep electronic copies for your own reference in many cases, so you can leater do searches for them, etc.
the world span perfectly ok before their existence, and i'm sure it will spin just fine without it Well, duh. Before PDF forms most official forms were provided as hardcopy only, so you had to fill them in with a pen. Do you really want to go back to the day where you have to start over again when you made a mistake in one of the boxes?
then i suggest sending out a plain text file and request that it is filled in, it's something that i seriously can't see much of a requirement for. sorry. Then you obviously have never worked in a professional environment. Just the other day I used PDF to sign two legal documents (forms) with my certified electronic signature. I now have one copy on my file system that I can index for search, and the party I was dealing with has another copy with a legally binding signature that they can save electronically and/or print at their leisure. Until the free PDF readers include these critical workflow features they are not fit for any serious environment.
what corporation actually makes use of forms? Only every single one I've ever worked for. Some government offices here in Canada also provide PDF forms for situations where you have to submit a printed version of the form in the end. You could achieve something similar with web forms, except the printed version would look different depending on browser. Sometimes a consistent formatting is a real advantage. So it is either PDF forms or Word, and given a choice between the two, I definitely vote for PDF.
The proper country to compare us to if you want to understand what Ron Paul seeks is not North Korea but Switzerland. The Swiss don't keep troops in foreign countries and don't try to rule the world, but do trade successfully with numerous other nations. You may have noticed that a lot fewer people hate the Swiss than hate Americans. Note, however, that the Swiss are part of the UN. One reason the Swiss aren't hated as much as Americans is because they submit to the notion of international law, which is precisely what Ron Paul and the likes propose to eliminate.
How right you are. And the 900+ pages are just the tip of the iceberg, since they do not seem to include the documentation for the 3D part. Expect the full documentation to be sevaral thousands of pages. By the time that is all implemented and stable enough for regular use, several new generations of GPUs will have come and gone.
This is a great first step towards open source drivers, but there is a lot of catching up to do.
Not new, not innovative, but Calvin is not a high end tech school so you have to give them some slack.
Not if they put out a media release, you don't.
If it had just been a student web page of a course or honor's project, I'd have agreed. But the faculty involved should have had the judgement to not push this as "news".
Of course there is enough compute power and IO bandwidth. That is not the point. To play media without glitches, you need to feed the audio chip with data at a sustained rate, which means you need a teensy bit of CPU time in regular intervals. Likewise, you need the I/O for the audio to work slowly but STEADILY.
If you have a scheduler that treats all processes as equal, then it may NOT be willing to preempt a compute-intensive app just so the audio player may get its small time slice at just the right time. An in fact, if you want peak performance (say on a server), then that is probably the right decision. Likewise, if your I/O module just sees abstract data, then the download of a large zip file could easily stall the audio stream for a short period of time. These things happen all the time even on the most powerful platforms. That is why the Linux kernel guys argue over schedulers, and why most OSes start adding some lightweight realtime features (realtime OSes can provide guarantees on CPU and bandwidth availability although at a cost, as mentioned in my original post).
Of course, if you just start the audio player on your PC and walk away, you'll never see these kinds of problems. They only occur when you mix media with other demanding apps.
Most likely, the music will skip under some circumstances on XP, like when you start a new (big) job, or when you download from a (really) fast server. Most likely you either don't even notice, or you don't find it disruptive, since you are at that time focused on whatever other job you asked the computer to do.
Vista, however, tries to be "media-friendly", i.e. it wants to be installed on living room PCs, and in that environment any skipping is much less acceptable. So they added some characteristics of real-time operating systems, in that they tried to allocate audio (and presumably video) threads a certain guaranteed amount of CPU time. Unfortunately, it is REAL HARD to add realtime OS features without unduely sacrificing the performance of non-realtime apps (read: nobody really knows how to do it).
So essentially, they decided to sacrificed desktop performance for media performance. Most likely they thought they could get away with it because (they thought) media usage in "serious" environments is rare (hah!).
The real lesson to be learned here is that you maybe don't wnat to use the exact same OS configuration for all possible application scenarios. This is something that the Linux community should have a close look at as well, especially in the light of the recent scheduler debate!
Actually, you can create any color with a mixture of 2 (not 3) fully saturated (i.e. pure spectral) colors. For any given color, find its location in the CIE horseshoe diagram, and then draw any line through that point. The intersections of this line with the boundaries of the horseshoe diagram represent the spectral colors that you need to blend, and the distance of your target color to these point sgives you the weighting ratio.
The reason why current displays have to use 3 primaries is because the primaries are fixed, so you cannot create just ANY spectral wavelength.
Hex constants in Perl, like C/C++ have to start with "0", so the correct syntax for what you describe would be 0xCCCCC. Without the leading 0, the expression gets interpreted as a variable name.
Yeah, I saw it at Siggraph last year. I wasn't too impressed: this stuff makes a hell of a lot of noise, the number of dots/s seems quite limited, and the safety concerns are serious. At Siggraph, they had a whole area arounf the thing cordoned off.
This is a nice curiosity, but I can't see it going anywhere for real-world applications. Other display technologis at Siggraph (both last year and this year) are much more promising.
Why not just convert it to square feet, the standard floor space measure in the US and Canada? 1 m^2 = 10.764 sqft, so the whole thing is over 1900 sqft, or about the size of a reasonably sized house. Many one bedroom condos are around 600 sqft, so this really is quite large.
What the heck is "o-zone"? Ozone is a molecule, not some kind of atmospheric zone; that would be the ozone layer, i.e. the atmospheric layer with a high natural concentration of... wait for it... ozone.
Call me cynical, but I fully expected companies to edit wiki entries that affect their public image.
IMHO, the scary part is how pathetically stupid this particular company goes about it. One would hope that a company like Diebold knows a bit more about IT security. Just send an employee with a laptop to your local wifi coffee shop already. Jeez.
I wouldn't worry about that. Unlike some other countries, Germany is fairly strict about jurisdiction. Unless you carry any "offending" tools with you on your visit to Germany, or host the tools on a web page that is specifically aimed at german readers (say, in german language or on a.de server), they are not going to go after you for developing these tools in other countries where it is legal.
Going from theory to practice is the science... Actually, that would be the engineering. As in "there is no rocket science, there is physics, and then there is rocket engineering". Meanwhile, publishing the initial idea for other people (like, uh, an actual engineer) to built on is very valuable.
Oh, and your premise is wrong: building a MEMS chip of a non-trivial size pretty quickly runs in the hundreds of thousands of $, even with educational discounts. So pretty much you have to get the design ready, then ask for funding to build the thing, which is what they are presumably doing.
Try this. Take your digital camera, put it into the highest ISO setting, open the aperture all the way, and take a really short exposure of a CRT screen. What you'll see is a block of scanlines (about 20 or so for a 1/1000s exposure) that is bright, and the rest of the screen, which is very dim by comparison. The transitions between the bright and the dark regions will be very sharp, which will show you that the phosphor decay is quite rapid, and you only see multiple illuminated lines because, even at 1/1000s, the exposure time is still too long to see a point.
Phosphors have an exponential decay, which means that they fall off to a fraction of their peak intensity fairly rapidly, but it takes forever for them to dim completely. That is why you see radiation in a dark room, but it is at a level that is MUCH smaller than the level you get even showing a black screen with an active electron gun.
Probably too early to tell. Note, however, that LCDs aren't exactly the pinnacle of energy efficiency either. Most current color LCDs transmit only around 6-8% of light for full white. The theoretical limit is around 16% (1/6 of the light; most of the energy is lost due to polarization and in the color filters).
The proposed technology avoids polarization, so that is a potential gain of a factor 2 in light efficiency right there. Of course we will only know for sure once significanlty more effort has gone into engineering.
Slashdot Mirror
Let's face it: starting this late and with only a vague idea of hardware and software components required, he isn't going to win anyways. In the best of cases, the answers he gets here will make the difference between having an entry for the competition, and not having one at all.
For a realtime system? Surely you are kidding.
Could you write a little more about the differences? Not doubting what you say, just curious.
I don't know, does it support public key signatures, yet? And comments, as somebody else has pointed out.
There is also the 3D support, but I doubt that is a big issue for most users.
That is one fo the reasons why the government forms are using PDF, I am sure. PDF is an open format, so you can get free software to print it and fill in the form, even if it is cumbersome. Of course you really want to keep electronic copies for your own reference in many cases, so you can leater do searches for them, etc.
How right you are. And the 900+ pages are just the tip of the iceberg, since they do not seem to include the documentation for the 3D part. Expect the full documentation to be sevaral thousands of pages. By the time that is all implemented and stable enough for regular use, several new generations of GPUs will have come and gone.
This is a great first step towards open source drivers, but there is a lot of catching up to do.
Not if they put out a media release, you don't.
If it had just been a student web page of a course or honor's project, I'd have agreed. But the faculty involved should have had the judgement to not push this as "news".
Of course there is enough compute power and IO bandwidth. That is not the point. To play media without glitches, you need to feed the audio chip with data at a sustained rate, which means you need a teensy bit of CPU time in regular intervals. Likewise, you need the I/O for the audio to work slowly but STEADILY.
If you have a scheduler that treats all processes as equal, then it may NOT be willing to preempt a compute-intensive app just so the audio player may get its small time slice at just the right time. An in fact, if you want peak performance (say on a server), then that is probably the right decision. Likewise, if your I/O module just sees abstract data, then the download of a large zip file could easily stall the audio stream for a short period of time. These things happen all the time even on the most powerful platforms. That is why the Linux kernel guys argue over schedulers, and why most OSes start adding some lightweight realtime features (realtime OSes can provide guarantees on CPU and bandwidth availability although at a cost, as mentioned in my original post).
Of course, if you just start the audio player on your PC and walk away, you'll never see these kinds of problems. They only occur when you mix media with other demanding apps.
Most likely, the music will skip under some circumstances on XP, like when you start a new (big) job, or when you download from a (really) fast server. Most likely you either don't even notice, or you don't find it disruptive, since you are at that time focused on whatever other job you asked the computer to do.
Vista, however, tries to be "media-friendly", i.e. it wants to be installed on living room PCs, and in that environment any skipping is much less acceptable. So they added some characteristics of real-time operating systems, in that they tried to allocate audio (and presumably video) threads a certain guaranteed amount of CPU time. Unfortunately, it is REAL HARD to add realtime OS features without unduely sacrificing the performance of non-realtime apps (read: nobody really knows how to do it).
So essentially, they decided to sacrificed desktop performance for media performance. Most likely they thought they could get away with it because (they thought) media usage in "serious" environments is rare (hah!).
The real lesson to be learned here is that you maybe don't wnat to use the exact same OS configuration for all possible application scenarios. This is something that the Linux community should have a close look at as well, especially in the light of the recent scheduler debate!
Actually, you can create any color with a mixture of 2 (not 3) fully saturated (i.e. pure spectral) colors. For any given color, find its location in the CIE horseshoe diagram, and then draw any line through that point. The intersections of this line with the boundaries of the horseshoe diagram represent the spectral colors that you need to blend, and the distance of your target color to these point sgives you the weighting ratio. The reason why current displays have to use 3 primaries is because the primaries are fixed, so you cannot create just ANY spectral wavelength.
Hex constants in Perl, like C/C++ have to start with "0", so the correct syntax for what you describe would be 0xCCCCC. Without the leading 0, the expression gets interpreted as a variable name.
Yeah, I saw it at Siggraph last year. I wasn't too impressed: this stuff makes a hell of a lot of noise, the number of dots/s seems quite limited, and the safety concerns are serious. At Siggraph, they had a whole area arounf the thing cordoned off.
This is a nice curiosity, but I can't see it going anywhere for real-world applications. Other display technologis at Siggraph (both last year and this year) are much more promising.
Why not just convert it to square feet, the standard floor space measure in the US and Canada? 1 m^2 = 10.764 sqft, so the whole thing is over 1900 sqft, or about the size of a reasonably sized house. Many one bedroom condos are around 600 sqft, so this really is quite large.
I never said it wasn't a pollutant. I just pointed out that it is a molecule rather than a spatial region.
What the heck is "o-zone"? Ozone is a molecule, not some kind of atmospheric zone; that would be the ozone layer, i.e. the atmospheric layer with a high natural concentration of ... wait for it ... ozone.
Call me cynical, but I fully expected companies to edit wiki entries that affect their public image.
IMHO, the scary part is how pathetically stupid this particular company goes about it. One would hope that a company like Diebold knows a bit more about IT security. Just send an employee with a laptop to your local wifi coffee shop already. Jeez.
I wouldn't worry about that. Unlike some other countries, Germany is fairly strict about jurisdiction. Unless you carry any "offending" tools with you on your visit to Germany, or host the tools on a web page that is specifically aimed at german readers (say, in german language or on a .de server), they are not going to go after you for developing these tools in other countries where it is legal.
It is still an idiotic law, of course.
Oh, and your premise is wrong: building a MEMS chip of a non-trivial size pretty quickly runs in the hundreds of thousands of $, even with educational discounts. So pretty much you have to get the design ready, then ask for funding to build the thing, which is what they are presumably doing.
Try this. Take your digital camera, put it into the highest ISO setting, open the aperture all the way, and take a really short exposure of a CRT screen. What you'll see is a block of scanlines (about 20 or so for a 1/1000s exposure) that is bright, and the rest of the screen, which is very dim by comparison. The transitions between the bright and the dark regions will be very sharp, which will show you that the phosphor decay is quite rapid, and you only see multiple illuminated lines because, even at 1/1000s, the exposure time is still too long to see a point.
Phosphors have an exponential decay, which means that they fall off to a fraction of their peak intensity fairly rapidly, but it takes forever for them to dim completely. That is why you see radiation in a dark room, but it is at a level that is MUCH smaller than the level you get even showing a black screen with an active electron gun.
Probably too early to tell. Note, however, that LCDs aren't exactly the pinnacle of energy efficiency either. Most current color LCDs transmit only around 6-8% of light for full white. The theoretical limit is around 16% (1/6 of the light; most of the energy is lost due to polarization and in the color filters).
The proposed technology avoids polarization, so that is a potential gain of a factor 2 in light efficiency right there. Of course we will only know for sure once significanlty more effort has gone into engineering.