Now we need an open source, peer to peer replacement for Tivo. When you see a commercial, you push the "commercial" button to skip it. This uploads the info that the current content is a commercial. In a few seconds, there's a consensus that there's a commercial on,
that info is both sent out and stored, and the commercial is skipped on everyone's system, now and forever.
"Thumbs up" and "Thumbs down" info can also be aggregated, to provide ratings others can use. This drives a blog system, so you can go on and discuss what's good and what sucked. That's also useful as a way to make consensus corrections to the TV schedule, since the free sources of that info can be a little off.
The fastest and most accurate commercial-button pushers get listed on a web site as high scorers.
Actually, no. Stanford's VW Touareg was provided by VW with the ability to be driveable via a serial port. This is apparently a mod they use internally for testing.
The map is given to each of the teams several (3?) hours prior to the start of the race. One result of this subtle difference is that teams can program a general path into the vehicle and have it deviate from it only as necessary instead of just popping the DVD into the computer and having the computer do everything.
CMU tried that last year. They'd obtained custom aerial and LIDAR imagery of the route, and had a semitrailer full of people at workstations manually programming the route in the two hours before the race.
It didn't work. They crashed three times in eight miles. Notably, they plowed through a sheet metal fence DARPA had placed on the route, and which apparently wasn't in their aerial photographs.
This year, the USAF colonel running the Grand Challenge made it very clear that preprogramming wouldn't work. Many of the teams groaned when he put up the slide with the pictures of tank traps.
We just put in the waypoint file, exactly as DARPA gives it to us, and go. We don't do any manual preprocessing.
OpenCV is useful if you want to play with vision. The LK tracker is pretty good. The stereo algorithms are weak. The tool to get camera params by looking at a moving checkerboard is clever, but flakey.
Because, despite decades of work, vision processing of unstructured scenes still sucks.
There are things that work in computer vision. You can do stereo, if the image has strong edges in it. You can pick out big moving objects. You can find the horizon. You can work out your own positional movements from video. You can find faces, align, and recognize them, sort of.
You can find known objects in any orientation (which is very useful in industrial systems.)
You can follow roads.
That's so Tom's Hardware. "7 Pentium M CPUs!", and no word about the algorithms. They could have at least said more about the sensors. Actually, everybody's sensors suck. The radars can't profile terrain, the LIDAR units are only line scanners, the stereo vision systems have trouble locking up on dirt, and the vision systems are a long way from being intelligent. True 3D LIDAR is coming, but not this year. The Grand Challenge rules prohibit the use of the best available 3D LIDAR system, because it was developed with Government funding and wasn't available by August of last year.
So we have a line-scanning LIDAR on a tilt head, like CMU, which is an adequate but bulky solution..
We have two industrial Pentium 4 machines running QNX, on our Grand Challenge entry, along with five Galil programmable motor controllers. We have room for 3 CPUs, but the compute load fit on two of them, so we took the third one out.
Technically, QNX was an excellent choice, but because few people know it and many don't want to learn it, using it has made recruiting difficult.
As a musician, I take offense at the notion that I must be working in some sort of content factory, extruding music product on an assembly line.
Well, you are. The music industry is a brand-based advertising and marketing driven business. Actual performance of the music is incidental. Studio musicians are easily available who play better than you do. Post-production is where the real work is done, anyway. Being able to play accurately in real time has become an irrelevant skill, like penmanship.
There are bands which, over time, replaced all of their members while keeping the same name. It's about the brand, not the band.
Enthusiasts for biofuels seem to use a wierd analysis approach. The term "system expansion approach" appears in papers justifying ethanol production, and most of those papers are by Bruce Dale. That form of analysis doesn't seem to be used for anything but biofuels, so that's suspicious.
It was heavily criticized at the time. Earth is under attack, the Captain has the last powerful military force, and he parks the ship and goes AWOL looking for his wife. That's cowardice in the face of the enemy.
They shoot officers for that.
For a while, DHS actually had someone with a clue in that job. But Amit Yoran got fed up and quit, because he didn't have enough clout to do anything.
He'd previously been the head of Riptech, a security company he sold to Symantec.
The current head of cybersecurity is a lawyer and TV producer. But he's a placeholder.
The new guy will probably be from Microsoft.
The new head of DHS, Chertoff, actually has more of a clue than his predecessor, and he's getting flak for it. Chertoff has been saying that we need to focus on ways somebody can kill thousands of people, and not waste effort on lower-priority threats like ordinary suicide bombers. This is unpopular politically. It means doing hard things like incoming freight container and truck inspection, instead of silly stuff like the "no fly list".
Right now, most zombie machines send using some arbitrary identity. Most of them are just proxies or forwarders, not mail generators. The way the spam industry works is that you rent some zombies at SpecialHam, get a "bulletproof mail server" from Black Box Hosting in China, install Dark Mailer, and go. Dark Mailer runs on the "bulletproof mail server" and generates the messages, which are sent via your rented proxy farm.
If sender ID goes in, the software that takes over a target machine will just have to use the normal sending identity for that machine, or, more simply, transmit it back to the bulk mailer so the mailer can construct the outgoing messages accordingly.
"I wired this single coil up to a big resistor (somewhere around 1 ohm I think) and cranked it hard by hand. At about 104 rpm, under load, I had 6 Volts AC and 6 Amps into the resistor (36 Watts). Im not sure if I'm figuring all this right - but I think that with 12 coils wired in 3 phase Star configuration... I should be close to 400 watts @ 100 rpm."
Unclear what he's measuring with, but it it's a typical DVM, the voltages and currents are peak values, not average. If he's getting good sine waves out, multiply by 0.707. For worse waveforms, the value is lower. He's probably getting under 300 watts out, which seems about right for the machine he's built.
What's making this work is rare-earth magnets. The field strengths you get today from rare-earth magnets are so high that even this simple design will sort of work. But good designs with rare-earth magnets do far better.
What's scary is that his turbine has a wooden hub and no overspeed protection. Most wind turbines are built to feather or turn sideways ot the wind in overspeed conditions. This thing will probably throw a blade in a big storm.
Real Goods will sell you a 3KW watt wind turbine for $5000, or a 400 watt unit for $900. So spending $700 on magnets alone to build your own is not a win. Real Goods units deliver more power from a 4' rotor than this guy gets from a 17' rotor, and they tilt to a safe position in high winds.
It looks real enough to me, but not cost-effective.
The works themselves are "public domain" but the performance of those works is covered by copyright.
There is no private copyright here. The BBC is a unit of the British Government, and its musicians are thus Government employees. There's "crown copyright", but that's limited to certain types of materials.
In the US, you can download many performances by the United States Marine Band. There can be no copyright on those performances. Works by the U.S. Government are not subject to copyright protection
Free, legal MP3 here!
Automatic memory management, as in reference counted smart pointers, are in boost right now, and will be in the standard.
They've been in Boost for years. But you can't really do reference counted smart pointers safely in C++ without language changes. It's been tried many times. The basic issue revolves around the need to extract a raw pointer from a smart pointer to get anything done. The auto_ptr mess illustrates the problems. The auto_ptr spec has been changed three times, and it still isn't satisfactory. Various solutions have been proposed, but they either require language changes or don't work reliably.
Many of the safety problems in C++ could be fixed without performance penalties. But safety is a low priority of the C++ committee, and this is with Strostrup's consent. That's the problem. Every day, millions of people suffer from the unreliable software that results.
C++ is in deep trouble, and Strostrup is in denial about it.
C++ is the only major language with extensive abstraction but without memory safety. All other major languages are either memory-safe or don't hide the underlying machinery. (Java, C#, VB, Perl, Python, etc. all have automatic memory management. Some use garbage collection; some use reference counts. C is unsafe, but hides nothing.)
This fact is responsible for millions of program crashes every day. Most security holes in C++ code come from this problem. Java and C# were invented primarily to eliminate the safety problems of C++.
The open source community has generally stayed with C, where at least you can see by examination what's going on. C++ is losing market share to Java.
And Strostrup denies this is a problem.
This has happened before. Last time, it was Wirth. Wirth designed Pascal, Modula, and Modula II, but refused to admit that each had serious problems. He fought external compilation in Pascal. He fought extensions to the language.
He even fought compile-time arithmetic.
In the end, he took Pascal from a major language to a historical footnote.
Serious systems programming was once done in Pascal, but not in Wirth's version of it. The original Macintosh and Lisa software was written in nonstandard versions of Pascal. And much of the DOS era was built on Turbo Pascal. But proliferating nonstandard versions of Pascal caused another set of problems.
C++ has been in decline for years.
"Evans Data has found that the percentage of developers using C++ has steadily declined over the last six years--from 76 percent in the spring 1998 to 46 percent in fall 2004."
Strostrup also denies that.
The C++ committee has been taken over by template fanatics. Most of the committee's effort revolves around obscure template features that few will use, and which no responsible programming manager would allow on a mission-critical project. There's very little interest in language safety, and a vocal minority that insists language safety is undesirable or impossible.
All is not well in the C++ world. Claming otherwise is irresponsible.
The Qinetiq millimeter microwave scanner seems to be vaporware. Everything on their web site is just a drawing. It's not clear you can really get a good image with millimeter microwave. Detect metal, sure.
Reliably detect guns, maybe. Explosive analysis, unlikely.
Z-backscatter technology can definitely do this, but that's a low-dose X-ray system. It's actually doing an elemental analysis, so you can look for suspicious chemicals. But it's slow, there's a modest X-ray risk, and the AS&E equipment is huge.
Rapidscan has been able to downsize backscatter equipment a bit, but it's still a slow scan.
"In a recent
study, 19 out of 20 people preferred a Secure 1000 scan to an invasive pat-down
physical search."
Nescafe Hot was developed in 1995. It was test-marketed in the Midlands of the UK in 2001, but wasn't enormously successful. Priced at £1.16 for a small coffee, it was rejected by the market.
An earlier approach appears in the Japanese self-heating sake cans. It's the same calcium oxide/water reaction as the others, but the mechanism for setting it off is clunkier.
The idea is old, but the trick was to develop a simple, reliable way to start the reaction. The Japanese sake cans required that the user push a big pointed pin through a hole in one end of the can. Early versions had the pin attached externally to the can; later versions integrated the pin into a plastic cover. Thermotic Developments worked out an improved triggering mechanism with no loose parts. That's a neat trick, since it has to be idiot-proof, survive dropped cans and cases, and be manufacturable at high speed at very low cost with a near zero defect rate.
The next generation of the technology runs hotter and can do soup and small packaged meals.
It's about time. SGI has been dead for years.
on
SGI Faces Bankruptcy
·
· Score: 2, Informative
This has been coming for a while. The only question in Silicon Valley has been why it took so long.
Around 1997, I went down to Sony Imageworks in Hollywood to talk to them about physics engines. They were almost entirely an SGI shop back then, but had just purchased some NT systems running Softimage|3D.
I was asked whether some NT software was going to be ported to SGI, and, realizing that was a dead end, replied "Resistance is useless. You will be assimilated".
Three years later, I visited again. Everything was NT except for some of the same SGI machines I'd seen three years ago.
SGI just couldn't cope with graphics becoming cheap. Around 2000, they dramatically announced some NT workstations, priced from $7000 upwards. They just didn't get it.
SGI's supercomputer side developed some interesting hardware, but there's no real market for supercomputers. It's all government, and mostly pork anyway. Lousy price/performance has forced them out of the server farm business. What's left?
In the movie world, there's the theatrical release, rated by the MPAA, and then, often, there's the unrated "Director's cut", with extra sex, violence, or long boring scenes, depending on what was cut in the first pass. Games may go that way. Would probably increase sales, too.
19 known problems cause most bugs
on
The New C Standard
·
· Score: 3, Informative
I've been trying to get that information. Yoran said that when he was head of the National Cyber Security Division at the Department of Homeland Security. He quit shortly thereafter, and his replacement is a TV producer. At this point, it's not even clear that there is a National Cyber Security Division at DHS. It's no longer in their table of organization.
Apparently, there were objections from Microsoft when DHS started to point out that Microsoft was the problem. What's left of the
"National Cyber Security Division" actually seems to be CMU's CERT plus a front man, Andy Purdy, in Washington.
Slashdot Mirror
"Thumbs up" and "Thumbs down" info can also be aggregated, to provide ratings others can use. This drives a blog system, so you can go on and discuss what's good and what sucked. That's also useful as a way to make consensus corrections to the TV schedule, since the free sources of that info can be a little off.
The fastest and most accurate commercial-button pushers get listed on a web site as high scorers.
You wish. Sand. Dirt. Snow. Good quality concrete.
Some road surfaces just don't have many sharp edges. Stereo lockup by brute force correlation won't work on them.
Actually, no. Stanford's VW Touareg was provided by VW with the ability to be driveable via a serial port. This is apparently a mod they use internally for testing.
CMU tried that last year. They'd obtained custom aerial and LIDAR imagery of the route, and had a semitrailer full of people at workstations manually programming the route in the two hours before the race.
It didn't work. They crashed three times in eight miles. Notably, they plowed through a sheet metal fence DARPA had placed on the route, and which apparently wasn't in their aerial photographs.
This year, the USAF colonel running the Grand Challenge made it very clear that preprogramming wouldn't work. Many of the teams groaned when he put up the slide with the pictures of tank traps.
We just put in the waypoint file, exactly as DARPA gives it to us, and go. We don't do any manual preprocessing.
OpenCV is useful if you want to play with vision. The LK tracker is pretty good. The stereo algorithms are weak. The tool to get camera params by looking at a moving checkerboard is clever, but flakey.
Because, despite decades of work, vision processing of unstructured scenes still sucks.
There are things that work in computer vision. You can do stereo, if the image has strong edges in it. You can pick out big moving objects. You can find the horizon. You can work out your own positional movements from video. You can find faces, align, and recognize them, sort of. You can find known objects in any orientation (which is very useful in industrial systems.) You can follow roads.
Beyond that, not much works.
So we have a line-scanning LIDAR on a tilt head, like CMU, which is an adequate but bulky solution..
We have two industrial Pentium 4 machines running QNX, on our Grand Challenge entry, along with five Galil programmable motor controllers. We have room for 3 CPUs, but the compute load fit on two of them, so we took the third one out.
Technically, QNX was an excellent choice, but because few people know it and many don't want to learn it, using it has made recruiting difficult.
Well, you are. The music industry is a brand-based advertising and marketing driven business. Actual performance of the music is incidental. Studio musicians are easily available who play better than you do. Post-production is where the real work is done, anyway. Being able to play accurately in real time has become an irrelevant skill, like penmanship.
There are bands which, over time, replaced all of their members while keeping the same name. It's about the brand, not the band.
Enthusiasts for biofuels seem to use a wierd analysis approach. The term "system expansion approach" appears in papers justifying ethanol production, and most of those papers are by Bruce Dale. That form of analysis doesn't seem to be used for anything but biofuels, so that's suspicious.
It was heavily criticized at the time. Earth is under attack, the Captain has the last powerful military force, and he parks the ship and goes AWOL looking for his wife. That's cowardice in the face of the enemy. They shoot officers for that.
The current head of cybersecurity is a lawyer and TV producer. But he's a placeholder.
The new guy will probably be from Microsoft.
The new head of DHS, Chertoff, actually has more of a clue than his predecessor, and he's getting flak for it. Chertoff has been saying that we need to focus on ways somebody can kill thousands of people, and not waste effort on lower-priority threats like ordinary suicide bombers. This is unpopular politically. It means doing hard things like incoming freight container and truck inspection, instead of silly stuff like the "no fly list".
If sender ID goes in, the software that takes over a target machine will just have to use the normal sending identity for that machine, or, more simply, transmit it back to the bulk mailer so the mailer can construct the outgoing messages accordingly.
MX Logic reports that, as of March, 9% of spam already has valid SPF markings, and 0.83% have valid Sender ID markings. So the technology to bypass SPF and Sender ID is already deployed.
Works for some of us.
With the new high dynamic range lighting systems and 12-bit output to monitors, even more shades of black will be possible.
Unclear what he's measuring with, but it it's a typical DVM, the voltages and currents are peak values, not average. If he's getting good sine waves out, multiply by 0.707. For worse waveforms, the value is lower. He's probably getting under 300 watts out, which seems about right for the machine he's built.
What's making this work is rare-earth magnets. The field strengths you get today from rare-earth magnets are so high that even this simple design will sort of work. But good designs with rare-earth magnets do far better.
What's scary is that his turbine has a wooden hub and no overspeed protection. Most wind turbines are built to feather or turn sideways ot the wind in overspeed conditions. This thing will probably throw a blade in a big storm.
Real Goods will sell you a 3KW watt wind turbine for $5000, or a 400 watt unit for $900. So spending $700 on magnets alone to build your own is not a win. Real Goods units deliver more power from a 4' rotor than this guy gets from a 17' rotor, and they tilt to a safe position in high winds.
It looks real enough to me, but not cost-effective.
There is no private copyright here. The BBC is a unit of the British Government, and its musicians are thus Government employees. There's "crown copyright", but that's limited to certain types of materials.
In the US, you can download many performances by the United States Marine Band. There can be no copyright on those performances. Works by the U.S. Government are not subject to copyright protection Free, legal MP3 here!
If the RIAA doesn't like that, tough.
Over in the real world, it's beginning to look like the source of the Valerie Plume leak was Karl Rove.
They've been in Boost for years. But you can't really do reference counted smart pointers safely in C++ without language changes. It's been tried many times. The basic issue revolves around the need to extract a raw pointer from a smart pointer to get anything done. The auto_ptr mess illustrates the problems. The auto_ptr spec has been changed three times, and it still isn't satisfactory. Various solutions have been proposed, but they either require language changes or don't work reliably.
Many of the safety problems in C++ could be fixed without performance penalties. But safety is a low priority of the C++ committee, and this is with Strostrup's consent. That's the problem. Every day, millions of people suffer from the unreliable software that results.
C++ is the only major language with extensive abstraction but without memory safety. All other major languages are either memory-safe or don't hide the underlying machinery. (Java, C#, VB, Perl, Python, etc. all have automatic memory management. Some use garbage collection; some use reference counts. C is unsafe, but hides nothing.) This fact is responsible for millions of program crashes every day. Most security holes in C++ code come from this problem. Java and C# were invented primarily to eliminate the safety problems of C++. The open source community has generally stayed with C, where at least you can see by examination what's going on. C++ is losing market share to Java.
And Strostrup denies this is a problem.
This has happened before. Last time, it was Wirth. Wirth designed Pascal, Modula, and Modula II, but refused to admit that each had serious problems. He fought external compilation in Pascal. He fought extensions to the language. He even fought compile-time arithmetic. In the end, he took Pascal from a major language to a historical footnote.
Serious systems programming was once done in Pascal, but not in Wirth's version of it. The original Macintosh and Lisa software was written in nonstandard versions of Pascal. And much of the DOS era was built on Turbo Pascal. But proliferating nonstandard versions of Pascal caused another set of problems.
C++ has been in decline for years. "Evans Data has found that the percentage of developers using C++ has steadily declined over the last six years--from 76 percent in the spring 1998 to 46 percent in fall 2004." Strostrup also denies that.
The C++ committee has been taken over by template fanatics. Most of the committee's effort revolves around obscure template features that few will use, and which no responsible programming manager would allow on a mission-critical project. There's very little interest in language safety, and a vocal minority that insists language safety is undesirable or impossible.
All is not well in the C++ world. Claming otherwise is irresponsible.
That's been tried. "Salvage One", from 1979.
Z-backscatter technology can definitely do this, but that's a low-dose X-ray system. It's actually doing an elemental analysis, so you can look for suspicious chemicals. But it's slow, there's a modest X-ray risk, and the AS&E equipment is huge.
Rapidscan has been able to downsize backscatter equipment a bit, but it's still a slow scan.
"In a recent study, 19 out of 20 people preferred a Secure 1000 scan to an invasive pat-down physical search."
Someone took apart a Nescafe Hot can, which is the same technology Puck has licensed.
An earlier approach appears in the Japanese self-heating sake cans. It's the same calcium oxide/water reaction as the others, but the mechanism for setting it off is clunkier.
The idea is old, but the trick was to develop a simple, reliable way to start the reaction. The Japanese sake cans required that the user push a big pointed pin through a hole in one end of the can. Early versions had the pin attached externally to the can; later versions integrated the pin into a plastic cover. Thermotic Developments worked out an improved triggering mechanism with no loose parts. That's a neat trick, since it has to be idiot-proof, survive dropped cans and cases, and be manufacturable at high speed at very low cost with a near zero defect rate.
The next generation of the technology runs hotter and can do soup and small packaged meals.
Around 1997, I went down to Sony Imageworks in Hollywood to talk to them about physics engines. They were almost entirely an SGI shop back then, but had just purchased some NT systems running Softimage|3D. I was asked whether some NT software was going to be ported to SGI, and, realizing that was a dead end, replied "Resistance is useless. You will be assimilated".
Three years later, I visited again. Everything was NT except for some of the same SGI machines I'd seen three years ago.
SGI just couldn't cope with graphics becoming cheap. Around 2000, they dramatically announced some NT workstations, priced from $7000 upwards. They just didn't get it.
SGI's supercomputer side developed some interesting hardware, but there's no real market for supercomputers. It's all government, and mostly pork anyway. Lousy price/performance has forced them out of the server farm business. What's left?
In the movie world, there's the theatrical release, rated by the MPAA, and then, often, there's the unrated "Director's cut", with extra sex, violence, or long boring scenes, depending on what was cut in the first pass. Games may go that way. Would probably increase sales, too.
Apparently, there were objections from Microsoft when DHS started to point out that Microsoft was the problem. What's left of the "National Cyber Security Division" actually seems to be CMU's CERT plus a front man, Andy Purdy, in Washington.