Believe me, that is not my intent... I was speaking from the view of the patent holders. (I and still feel rather dirty from it...) I'd love to see software patents obliterated. Or at least reduced to 6 month-1 year lifetimes.
This has been brought up several times as a reason why companies don't like to release their source code. Software patents are so screwed up that likely just about every piece of code out there infringes on someones patent. The system would collapse under it's own weight if most source code was open...
They've done an amazing amount considdering their tiny budget. They've shown what can be done with a evolutionary development program, as opposed to NASA's revolutionary program (ie, throwing away Apollo hardware and starting over with the Shuttle).
Soyuz is probably a good deal safer then the shuttle...it has an escape system. Soyuz boosters have exploded before, but they've never killed their crews. The last Soyuz fatality was over 20 years ago.
The intent was always to send up a final Proton flight to de-orbit it somewhat more safely over the Pacific.
Are you sure it's supposed to be a Proton? I thought it was just a reconfigured Progress (with fuel takes in place of the normal storage bay), and thus to launch on a much cheaper Soyuz...
The question seems to be the cost of a Soyuz spacecraft. The Soyuz booster is around 15-20 million per launch. But I've heard $100 million for the spacecraft...so unless they can get them cheaper they're out of luck.
And, of couse, a Soyuz can only hold three people, one of which has to be a pilot.
Metal fatigue in airplanes only comes about because of the tremendous strain put on the airframe during takeoffs and landings.
This isn't quite right. The main cause of metal fatigue in jets is pressurization and depressurization during flight. Of course only parts of Mir suffer from this type of fatigue (mainly the airlock and docking node).
The main structural problem with Mir is corrosion. Mir has been up there a long time and during most of that time the interior has been damp with both water and various (corrosive) coolants.
Radiation isn't really a problem in LEO. The Van Allen belts which shield us here on the ground do just as good a job in low orbit. This is only a problem when you move through the belts themselves or go above them.
And, of course, the danger of impact is basically constant. Mir is no more dangerious (from this point of view) then it has ever been in the past. In fact, with Mir's prior experience with the Progress tug, it's shown that it can take a puncture and not respond with a catastrophic pressure loss. I don't know that I would expect some of the ISS modules to do that well.
Much more worrying are the Mir subsystems...the environmental system, for instance. And things like the EVA hatch, which is held together with C-clamps. There are also the fundamental design flaws, like running ducts and cables though hatchways.
The US has no launcher capable of launching the Service Module. It won't fit in the shuttle's payload bay. It might fit on a Titan IV, but those are terribly expensive and (lately) tend to blow up a lot. It's probably the least of all evils just to wait for Proton to be reapproved. I don't think the software for the US lab module is ready yet, anyway...
Might have helped if the article was either acurate or interesting, or not shuffled off somewhere beyond the front page.
While I agree that the subsections don't seem to be working very well (I don't think more then 10 people actually know they exist), I don't see anything wrong with the accuracy of this story...NASA is indeed closing down this site. And it's also rather interesting, as it's more evidence of just how bad things have gotten at nasa...
f I had a nickel everytime I hear someone bitch about how crippled the current x86 architecture is by the industry dependence on backwards compatibility with an ancient and outdated architecture, and why we can't throw it out and design a much better chip from the ground up?
We can. Just don't run MS software and switch to another architecture... PowerPC comes to mind, as do MIPS, Sparc, and ARM. All these chips are much cleaner then x86 (or IA32 or whatever they like to call it this week), not to mention the fact that they give much more bang for your buck (and your watt...).
The whole "intel problem" is caused by stupid people making dumb decisions about what software they feel they must run. When if they would just step back they would see their employees wasting huge amounts of time fiddling around with font attributes in Word, instead of actually being productive.
No, IIRC NTSC is the second FCC color standard, and do remember that there were a number of early television experiments since the 20's.
The original color standard was basically going to transmit the R, G and B channels as sequential black and white signals, and the reciver would have a synchronized spinning wheel in front of the tube with R, G and B filters.
The spinning color wheel was (iirc) CBS's system. It was just a screwy idea, akin to Baird's mechanical television systems. It was never an actual standard, though. It was only used by CBS in some experimental broadcasts.
What became NTSC was RCA's totally electronic color system. As you said it had the huge advantage of being compatible with B&W recievers (which used the original RCA B&W system), plus it didn't need a big wheel spinning away all the time. I don't think the CBS system was ever seriously considdered as a viable color standard.
So they wedged the color data into a "colorburst" (high speed high-density data pulse with color info in the blanking interval).
Umm, not quite. The color burst is just a marker indicating that the signal is in color. The real color information is contained in a high frequency subcarrier hanging on the original B&W luminance signal. It takes up space that was originally the high frequency part of the B&W signal, so you trade a little B&W resolution for color information. The color is also transmitted using some modulation technique (which I cannot remember the name for) which uses phase changes to record information... It's a form of analog compression, really. Phase shifts caused by multipath are why NTSC color sucks to badly... PAL systems solve this last problem by trading a bit of color resolution for a form of analog error correction. NTSC systems solve it with a HUE control....................
DTV is good for everybody, the consumer and the broadcasters, as it lets the customer choose the channels they want to get. No more Food network if you don't want it.
"DTV" is good for only one thing: making cable companies more money. They get to free up part of their bandwidth (which they can sell for more money to someone else) and provide crappy over-compressed video to their customers. Plus it makes it seriously harder for them to upgrade to HDTV service in the future, since the previously available bandwidth is now used up by data services.
Half of their problem is already solved: people are just now upgrading to a medium (DVD) which doesn't have the capacity to handle high definition video...
AFAIK, you can get decent quality in 2Mbps, and IIRC you can send 6Mbps through a classic TV "channel",
Uh, no. You can get crappy quality though 2 Mbps. This is about what TCI's horrible "digital cable" system uses. It is truly a step backwards. Really amazing that people will put up with it.
A 6MHz channel will give you about 19MBps using the HDTV modulation (of which I forget the name) technique. This is why broadcasters like digital, they can use that bandwidth in whatever way they see fit. So they can stick a crappy 2-3 MBps video stream in there and sell the rest as a highspeed data service.
Cable companies don't care for this, though. They have never really given out full 6MHz channels in the first place. They don't gain any extra bandwidth by going to HDTV, they loose it, especially if they can go to a TCI-like digital cable system instead.
Sure. At the time the FCC had a backbone. They not only demanded that the new standard be compatible with the old B&W sets (the broadcast format of which was unnamed, as far as I know) but also that the new standard (NTSC) keep within the original 6 MHz bandwidth. The resulting system worked pretty well, with somewhat decent color on color sets, and only a slightly fuzzier image on B&W sets.
Of course today the FCC is totally spineless. They couldn't even handle the decision of what resolutions and frame rates to support.
The cost issue isn't nearly as bad as it was with color, though. Color sets were horribly expensive for many many years after NTSC came around. With our modern computer industry environment and Moore's law we should see prices of hdtv-capable sets come down very rapidly. All it will take is for someone to start pumping out integrated Mpeg2 framebuffer chipsets...
Also, unlike color/bw, the new digital tv sets are not automatically compatible with the old standard. You have to buy an expensive additional device for that.
Uh, I think you've got that backwards. All the hdtv's I've seen can handle NTSC just fine. The problem (if one can call it that) is that older ntsc sets can't handle hdtv signals.
Personally, I feel that using NTSC in what is supposed to be the most technologically advanced country in the world is simply pathetic. It was the first color television standard for gods sake. It's time to move on. If the FCC had gotten it's act together in the mid-late 80s we would be done with the conversion by now, but they didn't so we get to go through it now (now that there are three or four TVs per home...hmm...).
So, what's special about my 19" that makes it capable of telling the OS what it can do?
Your new monitor supports DDC (I think thats the name). So it can tell the OS what timings it supports. XFree86 4.x can use this as well.
The main point I was trying to make above is that the only reason people see windows as being easier is that the monitor manufacturers themselves have done all the work. If the manufacturers cared about unix users at all they'd just print a few mode lines in the manual and call it good.
Why is it that my monitors and video cards have always "just worked" with MacOS, BeOS, and Windows, but when i try to use linux, it becomes a pain to configure?
If it's all been modern hardware then it either came with "driver" disks (a monitor doesn't need a driver...but the disk contains hardware specs), or microsoft included the information on the windows CD. Go back and try to get an old no-brand SVGA monitor with no surviving documentation to work with windows. You're very lucky if you can get anything more then VESA modes to work.
But then walk around inside an office building and look at the windows displays on most people's desks...the vast majority are still running in 640x480x8bits...even when the hardware can likely do 1280x1024x24bit. The underlying difference is clear: windows users just don't care. They're the same people who watch movies on VHS on a 13" screen 20 feet away.
The current crop of unix / X11 users are different. They're old laserdisc owners. They like to see if they can squeeze 50 extra pixels onto their desktops.
There is motion between the fields, so you are really seeing 60 new images every second.
While there is motion between the fields (which is why a sports broadcast looks better on video then it does on 24FPS film), by the time it's displayed on a CRT the two fields are on the screen at the same time... So wether you end up with any actual extra information or just a smear is somewhat academic.
NTSC (and every other analog televison standard) uses interlacing to reduce flicker. 24, 25, or 30 fps video if displayed in real time flickers horribly. Film avoids the problem by displaying each frame multiple times...televesion with interlacing...
When was the last time you had your monitor in interlaced mode, if ever?
Uh, I'm sitting in front of a 1024x768x81Hz Interlaced display at this very moment.
Perceptually, there is a huge difference.
Perception is VERY different when you're talking about full motion video viewed from across the room. The american B&W television system choose interlaceing as a trade off between frame rate, flicker, and bandwidth. 60Hz interlaced only takes as much bandwidth as 30 Hz progressive, but provides a far better picture. Interlacing is not automatically bad, despite what most of our video industry would have you believe.
Slashdot Mirror
D'oh!
Believe me, that is not my intent... I was speaking from the view of the patent holders. (I and still feel rather dirty from it...) I'd love to see software patents obliterated. Or at least reduced to 6 month-1 year lifetimes.
This has been brought up several times as a reason why companies don't like to release their source code. Software patents are so screwed up that likely just about every piece of code out there infringes on someones patent. The system would collapse under it's own weight if most source code was open...
They've done an amazing amount considdering their tiny budget. They've shown what can be done with a evolutionary development program, as opposed to NASA's revolutionary program (ie, throwing away Apollo hardware and starting over with the Shuttle).
Soyuz is probably a good deal safer then the shuttle...it has an escape system. Soyuz boosters have exploded before, but they've never killed their crews. The last Soyuz fatality was over 20 years ago.
The intent was always to send up a final Proton flight to de-orbit it somewhat
more safely over the Pacific.
Are you sure it's supposed to be a Proton? I thought it was just a reconfigured Progress (with fuel takes in place of the normal storage bay), and thus to launch on a much cheaper Soyuz...
The question seems to be the cost of a Soyuz spacecraft. The Soyuz booster is around 15-20 million per launch. But I've heard $100 million for the spacecraft...so unless they can get them cheaper they're out of luck.
And, of couse, a Soyuz can only hold three people, one of which has to be a pilot.
Metal fatigue in airplanes only comes about because of the tremendous strain put on the airframe during takeoffs and landings.
This isn't quite right. The main cause of metal fatigue in jets is pressurization and depressurization during flight. Of course only parts of Mir suffer from this type of fatigue (mainly the airlock and docking node).
The main structural problem with Mir is corrosion. Mir has been up there a long time and during most of that time the interior has been damp with both water and various (corrosive) coolants.
Radiation isn't really a problem in LEO. The Van Allen belts which shield us here on the ground do just as good a job in low orbit. This is only a problem when you move through the belts themselves or go above them.
And, of course, the danger of impact is basically constant. Mir is no more dangerious (from this point of view) then it has ever been in the past. In fact, with Mir's prior experience with the Progress tug, it's shown that it can take a puncture and not respond with a catastrophic pressure loss. I don't know that I would expect some of the ISS modules to do that well.
Much more worrying are the Mir subsystems...the environmental system, for instance. And things like the EVA hatch, which is held together with C-clamps. There are also the fundamental design flaws, like running ducts and cables though hatchways.
The US has no launcher capable of launching the Service Module. It won't fit in the shuttle's payload bay. It might fit on a Titan IV, but those are terribly expensive and (lately) tend to blow up a lot. It's probably the least of all evils just to wait for Proton to be reapproved. I don't think the software for the US lab module is ready yet, anyway...
Might have helped if the article was either acurate or interesting, or not shuffled off somewhere beyond the front page.
While I agree that the subsections don't seem to be working very well (I don't think more then 10 people actually know they exist), I don't see anything wrong with the accuracy of this story...NASA is indeed closing down this site. And it's also rather interesting, as it's more evidence of just how bad things have gotten at nasa...
f I had a nickel everytime I hear someone bitch about how crippled the current x86 architecture is by the industry dependence on backwards compatibility with an ancient and outdated architecture, and why we can't throw it out and design a much better chip from the ground up?
We can. Just don't run MS software and switch to another architecture... PowerPC comes to mind, as do MIPS, Sparc, and ARM. All these chips are much cleaner then x86 (or IA32 or whatever they like to call it this week), not to mention the fact that they give much more bang for your buck (and your watt...).
The whole "intel problem" is caused by stupid people making dumb decisions about what software they feel they must run. When if they would just step back they would see their employees wasting huge amounts of time fiddling around with font attributes in Word, instead of actually being productive.
No, IIRC NTSC is the second FCC color standard, and do remember that there were a number of early television experiments since the 20's.
The original color standard was basically going to transmit the R, G and B channels as sequential black and white signals, and the reciver would have a synchronized spinning wheel in front of the tube with R, G and B filters.
The spinning color wheel was (iirc) CBS's system. It was just a screwy idea, akin to Baird's mechanical television systems. It was never an actual standard, though. It was only used by CBS in some experimental broadcasts.
What became NTSC was RCA's totally electronic color system. As you said it had the huge advantage of being compatible with B&W recievers (which used the original RCA B&W system), plus it didn't need a big wheel spinning away all the time. I don't think the CBS system was ever seriously considdered as a viable color standard.
It's not the fault of the decoder, though. It's your cable company trying to get away with using as little bandwidth as possible.
So they wedged the color data into a "colorburst" (high speed high-density data pulse with color info in the blanking interval).
Umm, not quite. The color burst is just a marker indicating that the signal is in color. The real color information is contained in a high frequency subcarrier hanging on the original B&W luminance signal. It takes up space that was originally the high frequency part of the B&W signal, so you trade a little B&W resolution for color information. The color is also transmitted using some modulation technique (which I cannot remember the name for) which uses phase changes to record information... It's a form of analog compression, really. Phase shifts caused by multipath are why NTSC color sucks to badly... PAL systems solve this last problem by trading a bit of color resolution for a form of analog error correction. NTSC systems solve it with a HUE control....................
DTV is good for everybody, the consumer and the broadcasters, as it lets the customer choose the channels they want to get. No more Food network if you don't want it.
"DTV" is good for only one thing: making cable companies more money. They get to free up part of their bandwidth (which they can sell for more money to someone else) and provide crappy over-compressed video to their customers. Plus it makes it seriously harder for them to upgrade to HDTV service in the future, since the previously available bandwidth is now used up by data services.
Half of their problem is already solved: people are just now upgrading to a medium (DVD) which doesn't have the capacity to handle high definition video...
AFAIK, you can get decent quality in 2Mbps, and IIRC you can send 6Mbps through a classic TV "channel",
Uh, no. You can get crappy quality though 2 Mbps. This is about what TCI's horrible "digital cable" system uses. It is truly a step backwards. Really amazing that people will put up with it.
A 6MHz channel will give you about 19MBps using the HDTV modulation (of which I forget the name) technique. This is why broadcasters like digital, they can use that bandwidth in whatever way they see fit. So they can stick a crappy 2-3 MBps video stream in there and sell the rest as a highspeed data service.
Cable companies don't care for this, though. They have never really given out full 6MHz channels in the first place. They don't gain any extra bandwidth by going to HDTV, they loose it, especially if they can go to a TCI-like digital cable system instead.
Sure. At the time the FCC had a backbone. They not only demanded that the new standard be compatible with the old B&W sets (the broadcast format of which was unnamed, as far as I know) but also that the new standard (NTSC) keep within the original 6 MHz bandwidth. The resulting system worked pretty well, with somewhat decent color on color sets, and only a slightly fuzzier image on B&W sets.
Of course today the FCC is totally spineless. They couldn't even handle the decision of what resolutions and frame rates to support.
The cost issue isn't nearly as bad as it was with color, though. Color sets were horribly expensive for many many years after NTSC came around. With our modern computer industry environment and Moore's law we should see prices of hdtv-capable sets come down very rapidly. All it will take is for someone to start pumping out integrated Mpeg2 framebuffer chipsets...
Also, unlike color/bw, the new digital tv sets are not automatically compatible with the old standard. You have to buy an expensive additional device for that.
Uh, I think you've got that backwards. All the hdtv's I've seen can handle NTSC just fine. The problem (if one can call it that) is that older ntsc sets can't handle hdtv signals.
Personally, I feel that using NTSC in what is supposed to be the most technologically advanced country in the world is simply pathetic. It was the first color television standard for gods sake. It's time to move on. If the FCC had gotten it's act together in the mid-late 80s we would be done with the conversion by now, but they didn't so we get to go through it now (now that there are three or four TVs per home...hmm...).
Just don't try to edit anything over 64K...
If you're stuck in windows, PFE is probably the way to go. Unless you've installed Emacs, of course.
Hmm... Maybe they should make use of one of the many available HTML VALIDATORS. Crappy HTML is no excuse for anything.
A browser should not silently render a broken page.
So, what's special about my 19" that makes it capable of telling the OS what it can do?
Your new monitor supports DDC (I think thats the name). So it can tell the OS what timings it supports. XFree86 4.x can use this as well.
The main point I was trying to make above is that the only reason people see windows as being easier is that the monitor manufacturers themselves have done all the work. If the manufacturers cared about unix users at all they'd just print a few mode lines in the manual and call it good.
right here: http://gullfoss.fcc.gov:8080/cgi-bin/ws.exe/prod/o et/forms/misc/Grant_For m.hts?application_id=93726
Why is it that my monitors and video cards have always "just worked" with MacOS, BeOS, and Windows, but when i try to use linux, it becomes a pain to configure?
If it's all been modern hardware then it either came with "driver" disks (a monitor doesn't need a driver...but the disk contains hardware specs), or microsoft included the information on the windows CD. Go back and try to get an old no-brand SVGA monitor with no surviving documentation to work with windows. You're very lucky if you can get anything more then VESA modes to work.
But then walk around inside an office building and look at the windows displays on most people's desks...the vast majority are still running in 640x480x8bits...even when the hardware can likely do 1280x1024x24bit. The underlying difference is clear: windows users just don't care. They're the same people who watch movies on VHS on a 13" screen 20 feet away.
The current crop of unix / X11 users are different. They're old laserdisc owners. They like to see if they can squeeze 50 extra pixels onto their desktops.
There is motion between the fields, so you are really seeing 60 new images every second.
While there is motion between the fields (which is why a sports broadcast looks better on video then it does on 24FPS film), by the time it's displayed on a CRT the two fields are on the screen at the same time... So wether you end up with any actual extra information or just a smear is somewhat academic.
NTSC (and every other analog televison standard) uses interlacing to reduce flicker. 24, 25, or 30 fps video if displayed in real time flickers horribly. Film avoids the problem by displaying each frame multiple times...televesion with interlacing...
When was the last time you had your monitor in interlaced mode, if ever?
Uh, I'm sitting in front of a 1024x768x81Hz Interlaced display at this very moment.
Perceptually, there is a huge difference.
Perception is VERY different when you're talking about full motion video viewed from across the room. The american B&W television system choose interlaceing as a trade off between frame rate, flicker, and bandwidth. 60Hz interlaced only takes as much bandwidth as 30 Hz progressive, but provides a far better picture. Interlacing is not automatically bad, despite what most of our video industry would have you believe.