Another truly innovative propulsion system, though. The Sky Cycle used superheated water that would flash into steam as the propellant.
Bob Truax was working way back then on a very similar project to the X-Prize projects. He had a little ship using the vernier engines from Atlas rockets, that was meant to lob somebody into space. Jeana Yeager, who built and flew the Voyager with Dick Rutan, was one of his astronaut candidates. Unfortunately, that project never reached fruition (well, perhaps it is fortunate for Ms Yeager.) But, the point is, this was more than 20 years ago. Truax never had somebody like Paul Allen behind him -- probably because he didn't have the spectacular track record that Rutan has.
CO2 emissions will likely increase over the next few decades, unless the countries that produce the most CO2 will start taking drastic action. That means the US, and to some extent, Europe and China. The 1% figure is about right.
The ULEV vehicles you trash actually eliminate far worse greenhouse gases. Methane, for example, is 100x as potent a greenhouse gas as CO2, and the advanced catalysts do eliminate almost all the hydrocarbons (like methane) from the exhaust.
These hurricanes are really Nature's way of suggesting to Floridians that their vote really matters.
Burt Rutan had nothing to do with the BD-10, that was a Jim Bede project all the way. Burt did contribute somewhat to the gloriously misbegotten BD-5, it's true.
The Pond Racer was a all-out unlimited-class racing plane. Unfortunately, air racing is an insanely dangerous business -- and the air racing history is populated largely by people who died in their planes. From the GeeBees to Tsunami, homebuilt air racers are just not for the faint of heart.
Burt's Amsoil racer, on the other hand, did manage to protect its pilot when it crashed after it was cut off by another plane on takeoff.
The Starship is every bit an example of what things can go wrong. I lay 99% of the blame on Beech's management and manufacturing, though. The prototype really did perform excellently -- but the production planes were sadly outperformed in many criteria (and especially the most commercially important criteria, like cruise speed and runway required) by convential King Airs.
As far as flying the next flight of SS1 so soon after unexplained performance on the most recent flight -- I completely agree with you. The 37th anniversary of Sputnik is not worth the risk. Spend another week evaluating the data, then fly.
Still, name another aero engineer with 10% of the achievements of Rutan -- Ok, I give you Kelly Johnson. Name another one:)
I've looked over most of the information Adobe has published, and it's not bad. It's true that a typical RAW format file is difficult to interpret. I've reverse-engineered a couple of RAW formats just for fun, (it's pretty easy if you can tell the camera to output a RAW and TIFF image of the same shot) and the Adobe propsed DNG format does have flags for most of the issues that I've come across (I have to say that there were some new ones for me, too -- the flag that specifies how closely the G in the RGRG rows compares to the G in the GBGB rows is something I've never even thought of.) It's good that Adobe has considered the possibility of more-than-three-channel cameras.
But -- I think that digital cameras are still *way* too new for this kind of standardization. Significant true innovation is happening at a frenetic pace, and if we limit RAW formats to a preconceived format we may inadvertantly (or advertantly, I suppose) squelch that innovation. Fuji's spectacular sensor with separate sensors at each pixel for dark and bright values is an example -- how would that be encoded here? One might well have a camera with vertical and horizontal polarizers on every other cell, to allow post-processessing to reduce or enhance specular highlights. Cameras could be built with psuedorandom placement of cells, to eliminate aliasing artifacts (Why not? It's not as if the semiconductor masks are laid down by hand anymore.)
In short, I think that this format could end up being a Procrustean bed that we force camera makers into, and that it's not worth it at this point.
While it's true that the so-called lossless JPEG can be used for 8-bit images, most of the RAW formats have significantly more bit depth than 8 bits. Therefore, any 8-bit encoding will necessarily lose some information, potentially extremely important information.
Most RAW formats are linear, and an 8-bit sampling of those images without gamma- or log-conversion will cause the loss of most of the picture detail.
Viewfinders in digital cameras, both still and motion picture.
Current non-reflex digital cameras have terribly lo-res displays, that don't give you nearly enough sharpness to focus the lens. Shooting sports, say, with a non-reflex camera is next to impossible.
I know that I'm in an infinitesimal minority on this issue, but I believe that reflex cameras are a horrible kludge for digital photography. The mechanical movement of the mirror is an anachronism, albiet a pretty and nice-sounding one. Lens design is severly compromised in reflex cameras, because the lens must be so far from the film or chip.
Anyway, a digital camera with a beautiful high-res digital viewfinder could be a very nice thing indeed.
I know that the Prius' engine is efficient, but is it really as efficient as a coal plant? I suppose once you take the electricity transmission and battery losses into account, that could well be true.
Coal plants could (people are talking about this seriously, although nobody's doing it yet) sequester their CO2 output -- that would be next to impossible in a vehicle.
Starting with 2004, Priuses are PZEV, as well SULEV. You're probably right, with today's electricity infrastructure, that perhaps the stock Prius is a better bet environmentally.
A group of people called PriusPlus have just about completed a plug-in modification of one of their 2004 Priuses. It is a great car, and a great way to save energy -- at least a good way to save gasoline.
The normal Prius uses its battery pack to help acceleration, hill climbing, and to power accessories. The battery pack is recharged by the gas engine and by regenerative braking. Every place except North America, the Prius has an EV button, which turns the car into a pure electric car -- but only for a mile or two before the battery reaches a state-of-charge (SOC) that is too low. The Prius battery back is designed to last an extremely long time (warranteed for 150,000 miles), and one way Toyota assures that is by limiting the SOC to a small range, from about 25% full to 80% full.
Priusplus is adding a separate "traction" battery, that works with the normal Prius drivetrain, to provide a long-distance EV mode. In their first proof-of-concept car (which should be finished this weekend) it uses 12 motorcycle Lead-Acid batteries, and it should go about 20 or 30 miles on an overnight (or overday) charge. Using far superiour Lithium Ion batteries, they should get about 80 miles for a battery pack that costs about $5,000 or so (although current Lithium cells are quite small indeed, requiring a rediculous number of batteries wired into a large pack)
If I could go, say, even 40 miles on a charge, I wouldn't use the gas motor in my Prius except to climb very steep hills during the week. I'd effectively get well over 100 mpg (Electricity costs, even in California, give a price-per-mile of about 2 cents. Unfortunately, at this point, the cost for the traction battery (because it is more deeply cycled it doesn't last as long) probably adds another few cents/mile.
PriusPlus is hoping to display there car at a show here in Los Angeles at the end of the month, and is attempting to persuade Toyota that this is a car they should build. Once people are educated about the benefits of hybrid technology, it should be a small step to show them the further benefits of plugging them in.
I watch as many as I can, they're pretty easy to see from Los Angeles, on a clear day (and most of them are!)
I've seen the Delta II launch of both Ikonos and Gravity Probe B. Both of these were daylight launches, and would have been impossible to see if they didn't have extremely tight (and nicely publicized by Aviation Week) launch windows. If you know where to look, they're quite nice to see. The arc of the rocket as it bends over and smoothly accelerates to the south is math in motion, just beautiful.
I've also watched a couple of the Minuteman launches testing missle defense systems. Again, these were well publicized events. On my street in Calabasas for the last one, everybody was out in their lawn chairs waiting for it. It did not disappoint. Compared to a satellite launch, the Minuteman gets out of the atmosphere in a hurry, and the solid fuel exhaust blooms into a huge flower-shaped colorful cloud once it is in space.
Still, it's basically impossible for me to see the beauty in a Minuteman launch. It's a goddamn ICBM, its only purpose to kill millions of people.
Any you idiots picking on the maintainer of the site -- get real. Read the site, there is absolutely nothing there than any third-grader couldn't figure out with ten minutes, a road atlas and a blunt crayon. There are a few other good Vandenberg launch sites out there, too, like this one
I've fantasized about burying a cellphone near Vandenberg, and set it up to call me when it feels the vibration of a launch. It'd be cool, cheap, and easy. Obviously the solar charger and antenna would have to be above ground. The problem with most Vandenberg launches is that you don't know when they are going to happen -- but if you knew they were firing you could just step outside and see.
Thad Beier
Re:Looooong article, but worth the read
on
Make Money Fast
·
· Score: 1
I agree with the above poster, that the article is worth reading -- it's a good story, well written. Typically Canadian:)
I think RFID tags are a good idea, but are overkill. A unique holographic image that hashes to a printed value on the face of the bill should do the job. Building unique coded holograms isn't really too hard. The hash algorithm that interprets them could be kept in a tamperproof chip. Obviously the bill checker would just say "Yes" or "No" -- you wouldn't be able to ask it what the hash of a particular hologram was.
Thad
Problem with Open Source in VFX, and a solution
on
SIGGraph and Open Source
·
· Score: 3, Informative
The problem is patents.
It's an unmitigated disaster. If I was to release the color correction tools I use at Hammerhead as Open Source software, for instance, there is no small chance that I would be sued by somebody, or more likely several somebodies, for infringements on their color correction patents. This kind of stuff is patented out the wazoo, and (unfortunately) the only thing that keeps the patent monster at bay is the fact that everybody does work secretly. That, and the fact that Hammerhead is so small that it's not worth suing.
Note well that studios are sued over almost every successful movie they make by people alleging the most tenuous copyright infrigement. A typical example is here.
Publishing open source software does have a tremendous advantage, though, in that it is a perfect vector for publishing information that could be used as prior art when trying to defend against other patents -- so open-source is a two-edged sword (or maybe a sword that is honed sharp at both ends.)
Perhaps, just perhaps, there is a solution. It might not be impossible to have anonymous open-source, with guaranteed anonymity provided much the same way the Cypherpunks' MixMaster remailer network works. That way, one could contribute to open-source projects, and share the benefits of your work with others, without exposing yourself to patent suits.
I'm not sure how one would do this, and the network of visual effects studios might be too small -- and the coding styles of the few hundred programmers might be too distinctive -- to have this work, but it could be interesting.
Well, mostly right. Larry Gritz did write BMRT, the Blue Moon Rendering Tools, and he did go to Pixar and was one of the leading people on the RenderMan team for quite a few years.
He left to form ExLuna, and ExLuna was then bought by Nvidia. At Nvidia, Gritz and the rest of his team are behind the new Gelato hardware-assisted high-quality rendering product. It's pretty cool. [pun reluctantly admitted to]
It's all a very long and interesting story, that unfortunately will not fit within the margin here.
As I recall (and I'm certain that others will correct you further) there is no closed-form solution for the three-body problem. The shapes of the orbits cannot be written down as a simple equation -- where (neglecting relativity) two orbiting bodies trace perfect ellipses.
On the other hand, you can calculate a solution to the three-body problem to any level of accuracy that you are interested in, without much effort. Yes, it's an approximation, but so is any calculation.
As usual, there is more to the story. Apple brought my company in on a project back in the mid 80's when they bought the Cray. While we had to sign an NDA in blood, I doubt anybody will mind me talking about it now, almost 20 years later.
Apple was trying to design a new cpu chip. It would have had vector processing capabilities not all that different from the Cray, so they bought the Cray both to do circuit simulations on the chip and as a model for their own design.
The chip was going to be a 100 MHz chip (an astonishing speed for the time) with a four-pipleline vector processing unit.
They considered (but eventually declined to) hire us to develop some kind of 3D desktop for the Mac. The idea was this would distinguish the Mac further from other computing systems, but they wouldn't be able to emulate the interface because they didn't have the horsepower.
Anyway, that's the Apple-Cray story as I understand it. I'm sure that there is a lot more to the story than I know, of course.
Thad Beier
Separating Linux users from Windows users
on
The Spyware Inferno
·
· Score: 3, Insightful
HTML doesn't have a 'rant' tag, but consider the following as such.
I personally cannot imagine having spyware on my machine, and I similarly cannot imagine any Linux user tolerating it. Most Linux users chose it, in large part, because of the control it gives you over everything that your computer does. Having your computer hijacked by advertisers is antithetical to that concept.
But I watch Windows users tolerate truly mindboggling amounts of adware/spamware/malware. The typical windows users tolerate 100 times what I would consider completely unacceptable.
I know it's elitist to say this, but what happens is that Windows users will make the tradeoff of malware to allow them to steal music and other content. They don't protest, because deep down they know what they're doing is wrong.
Linux users, typically, have no such guilt and therefore don't tolerate that kind of intrusion onto their computer.
Thad
Re:RGBCMY is more marketing factoid than it isreal
on
RGB to become RGBCMY
·
· Score: 5, Informative
As I recall, a linear combination of RGB can express any possible color -- if you allow for negative amounts of the components. A really bright yellow might be 1 R + 1 G -.2 B for example.
That's still a linear combination, but just one that's not particular useful in the real world of phosphors and filters.
Greg Brandeau of Pixar had a birds-of-a-feather meeting to discuss the use of Linux in visual effects and animation. It was well attended and Greg made a great presentation.
The main issue that he wanted to address was the need for visual effects studios to present as united a front as possible, so that people who write the software that we use (typically called ISVs) will be able to do Linux ports with some confidence. Greg points out that historically people had used Red Hat 7.x, and that companies like Nothing Real and Alias were able to make Linux versions of their programs without too much trouble. But now, most studios find Red Hat Enterprise licenseing unattractive, and everybody is choosing a different platform. Greg's poll found some companies running Fedora Core 2 (Pixar is doing this, as is my company,) others are using Suse and some are using RHE 3.0.
The problem we face is that if we balkanize too much, it will be impossible for ISVs to support us. Even though all the big visual effects and animation studios use Linux extensively, it is still a very small part of the market for these tools. Alias says that Linux is a distant third place in the number of licenses, after Windows and OS X. (Interestingly, the second place Linux distribution for Alias customers is TurboLinux, because of its strong Asian language support.)
Several vendors were present at the meeting. NVidia said that they really try hard to qualify on every Linux distribution -- they test on dozens of different distributions (some don't work). Intel, on the other hand, said that any past the top two or three just cannot be supported. In particular, because Linux is so customizable, it is often impossible to replicate bugs and problems, because it's unknown what the customer's environment is.
A general solution presented by Greg, which seemed reasonable, was to suggest to the ISVs and IHVs that they qualify for two systems, say RHE and Suse 9.1. Studios would have a stock system of one of those systems, and would test support issues on those stock systems to see if the problem was with the vendors stuff or was due to customization done at the studio. There would be a need to adjust what these two systems might be, over time, as things evolve. Greg suggested that Pixar would be happy to sponsor a web site and discussions on an ongoing (every six month, say) basis to update these reference systems.
John Carey gave a talk on the differnce between a distribution and a operating system -- basically presenting the challenge that ISVs face when trying to write to the Linux market.
The guy from Intel (I forget his name) suggested that Linux Standards Base (LSB) 2.0 is an attempt to address these problems -- after all, they are not unique to the visual effects industry. While he recognizes that LSB 2.0 is not a panacea, it will go a significant part of the way to help standardize Linux from the ISVs point of view.
Dan Wexler of NVidia noted that it's even harder than other people thought. He is working on abusing graphics cards in unconventional ways, and has found that he has significant problems with combinations of motherboards, graphics cards, and memory. Perhaps the reference systems mentioned above will have to specify hardware as well as software configurations.
Andy Maltz from the Academy of Motion Picture Arts and Sciences made a presentation as well, talking about the Academy's revivified Science and Technical Council. He suggested that the Academy was very interested in helping solve technical problems, and providing a forum to share solutions between studios.
All in all, it was a good meeting. My thanks to Greg for putting it together, and maintaining relentless good cheer through some difficult questions and issues.
Sorry, I said paraffin in the American sense of the word. We use the word paraffin to mean "candle wax". I understand that it means something different in the rest of the world.
Hybrid engines have a solid fuel and a liquid oxidizer. The article says that the DaVinci rocket uses Nitrous Oxide (a liquid if kept under moderate pressure) as the oxidizer, and something other than rubber (I'm guessing candle wax) as the fuel.
TFA says that the DaVinci group is planning on using a hybrid rocket engine, using nitrous oxide as the oxidizer (as Rutan's SpaceShipOne does) but using something other than synthetic rubber as the fuel. That does make a little bit of sense, as after Rutan's group settled on their rocket engine design, there has been some spectacular research out of Stanford using paraffin (in the American sense of the word) as hybrid rocket fuel. Paraffin has the nice property that it as it gets hot it turns from a solid into a very free-flowing liquid -- which lets it burn very quickly (something that rubber-burning hybrid motors have a hard time with -- the Rutan engine has four separate channels through the fuel to allow it to burn quickly, this leads to the possibility of blowing chunks of propellant.)
Unfortunately, though, the DaVinci website says that their ship will use Kerosene/LOX as the propellant and oxidizer. They have pictures of the engines, including some test firings, on the web site.
You just don't change engine technology at this point in the project.
The only possibilities are that these people are 1) insane or 2) scam artists. It's too bad, it would be spectacularly great if they were on the up-and-up...but...it doesn't appear that they are.
kfhickel says > Umm, RTFA, dude: SCO says it discovered the e-mails in a mountain of documents IBM produced in discovery related to SCO's lawsuit against IBM over the Linux operating system.
Nevertheless, you cannot go on fishing expeditions through discovery documents for alleged infractions that weren't part of your original complaint. Furthermore, and far more egregiously, you cannot share with the world information you found in confidence during discovery.
SCO has really screwed the pooch here. Darl is going to be found in contempt.
That said, it is hard to hold back when reporters come calling. Dan Lyons, in particular, is good at wheedling info that probably shouldn't be shared.
1) DES has been around a long time. People have attacked it for years, with every new and old technique of cryptanalysis. DES was created by IBM with help (no, really!) from NSA -- it was NSA that proposed adjustements in the S-Boxes that made DES more resistant to differential cryptanalysis. DES has proven to be secure, except for the obvious key-length problem, in the very best way you can prove an algorithm secure -- by having the best minds on the planet beat on it mercilessly for decades.
2) DES hardware exists, and is inexpensive and relatively secure. Using current hardware to impliment triple DES is easy.
3) It's upward compatible with existing systems. Using the same key three times yeilds is the same as doing regular 56-bit DES (The second DES is usually set up in decrypt mode.) One could well argue that the interoperability with single DES is not really a win, though -- that it allows users to be insecure.
While AES is great, and has been vetted as well as can be expected in the few years since its invention, triple DES is not a bad alternative.
There have been many reports recently of virus writers attempting to blackmail companies. Having this virus, an obvious DDoS attack on Google, happen the same day that Google announced the price of its IPO shares is just what you would expect if the Google didn't pay the blackmail.
I don't know how we'll ever be able to test this hypothesis, but I think that something stinks here.
It is a wonderful feature of fonts that the shape of the glyphs themselves is well-nigh uncopyrightable. Which is the reason that Postscript fonts are programs -- the programs *are* copyrightable. The names of the fonts, as well, can be protected by trademark and potentially by copyright.
But there's nothing to prevent one from making a beautiful font that looks extremely like Palatino or Times Roman for Linux -- except that it would take a significant amount of effort.
I worked with the management of Bitstream back when they were creating their font library back in the mid eighties. They hires famous font designers to recreate things that looked a lot like familiar fonts. They had something called "Swiss" which looked a lot like the normal "Helvetica" (Helvetica is latin for Switzerland). They hired Zapf to create their version of Palatino. Bitstream, of course, made beautiful fonts on their own, too -- most notably the Lucida family.
Slashdot Mirror
The Sky Cycle was Bob Truax, not Burt Rutan.
Another truly innovative propulsion system, though. The Sky Cycle used superheated water that would flash into steam as the propellant.
Bob Truax was working way back then on a very similar project to the X-Prize projects. He had a little ship using the vernier engines from Atlas rockets, that was meant to lob somebody into space. Jeana Yeager, who built and flew the Voyager with Dick Rutan, was one of his astronaut candidates. Unfortunately, that project never reached fruition (well, perhaps it is fortunate for Ms Yeager.) But, the point is, this was more than 20 years ago. Truax never had somebody like Paul Allen behind him -- probably because he didn't have the spectacular track record that Rutan has.
Thad Beier
CO2 emissions will likely increase over the next few decades, unless the countries that produce the most CO2 will start taking drastic action. That means the US, and to some extent, Europe and China. The 1% figure is about right.
The ULEV vehicles you trash actually eliminate far worse greenhouse gases. Methane, for example, is 100x as potent a greenhouse gas as CO2, and the advanced catalysts do eliminate almost all the hydrocarbons (like methane) from the exhaust.
These hurricanes are really Nature's way of suggesting to Floridians that their vote really matters.
Thad
Burt Rutan had nothing to do with the BD-10, that was a Jim Bede project all the way. Burt did contribute somewhat to the gloriously misbegotten BD-5, it's true.
:)
The Pond Racer was a all-out unlimited-class racing plane. Unfortunately, air racing is an insanely dangerous business -- and the air racing history is populated largely by people who died in their planes. From the GeeBees to Tsunami, homebuilt air racers are just not for the faint of heart.
Burt's Amsoil racer, on the other hand, did manage to protect its pilot when it crashed after it was cut off by another plane on takeoff.
The Starship is every bit an example of what things can go wrong. I lay 99% of the blame on Beech's management and manufacturing, though. The prototype really did perform excellently -- but the production planes were sadly outperformed in many criteria (and especially the most commercially important criteria, like cruise speed and runway required) by convential King Airs.
As far as flying the next flight of SS1 so soon after unexplained performance on the most recent flight -- I completely agree with you. The 37th anniversary of Sputnik is not worth the risk. Spend another week evaluating the data, then fly.
Still, name another aero engineer with 10% of the achievements of Rutan -- Ok, I give you Kelly Johnson. Name another one
Thad
I've looked over most of the information Adobe has published, and it's not bad. It's true that a typical RAW format file is difficult to interpret. I've reverse-engineered a couple of RAW formats just for fun, (it's pretty easy if you can tell the camera to output a RAW and TIFF image of the same shot) and the Adobe propsed DNG format does have flags for most of the issues that I've come across (I have to say that there were some new ones for me, too -- the flag that specifies how closely the G in the RGRG rows compares to the G in the GBGB rows is something I've never even thought of.) It's good that Adobe has considered the possibility of more-than-three-channel cameras.
But -- I think that digital cameras are still *way* too new for this kind of standardization. Significant true innovation is happening at a frenetic pace, and if we limit RAW formats to a preconceived format we may inadvertantly (or advertantly, I suppose) squelch that innovation. Fuji's spectacular sensor with separate sensors at each pixel for dark and bright values is an example -- how would that be encoded here? One might well have a camera with vertical and horizontal polarizers on every other cell, to allow post-processessing to reduce or enhance specular highlights. Cameras could be built with psuedorandom placement of cells, to eliminate aliasing artifacts (Why not? It's not as if the semiconductor masks are laid down by hand anymore.)
In short, I think that this format could end up being a Procrustean bed that we force camera makers into, and that it's not worth it at this point.
Thad
While it's true that the so-called lossless JPEG can be used for 8-bit images, most of the RAW formats have significantly more bit depth than 8 bits. Therefore, any 8-bit encoding will necessarily lose some information, potentially extremely important information.
Most RAW formats are linear, and an 8-bit sampling of those images without gamma- or log-conversion will cause the loss of most of the picture detail.
Thad
Viewfinders in digital cameras, both still and motion picture.
Current non-reflex digital cameras have terribly lo-res displays, that don't give you nearly enough sharpness to focus the lens. Shooting sports, say, with a non-reflex camera is next to impossible.
I know that I'm in an infinitesimal minority on this issue, but I believe that reflex cameras are a horrible kludge for digital photography. The mechanical movement of the mirror is an anachronism, albiet a pretty and nice-sounding one. Lens design is severly compromised in reflex cameras, because the lens must be so far from the film or chip.
Anyway, a digital camera with a beautiful high-res digital viewfinder could be a very nice thing indeed.
Thad
I know that the Prius' engine is efficient, but is it really as efficient as a coal plant? I suppose once you take the electricity transmission and battery losses into account, that could well be true.
Coal plants could (people are talking about this seriously, although nobody's doing it yet) sequester their CO2 output -- that would be next to impossible in a vehicle.
Starting with 2004, Priuses are PZEV, as well SULEV. You're probably right, with today's electricity infrastructure, that perhaps the stock Prius is a better bet environmentally.
thad
The normal Prius uses its battery pack to help acceleration, hill climbing, and to power accessories. The battery pack is recharged by the gas engine and by regenerative braking. Every place except North America, the Prius has an EV button, which turns the car into a pure electric car -- but only for a mile or two before the battery reaches a state-of-charge (SOC) that is too low. The Prius battery back is designed to last an extremely long time (warranteed for 150,000 miles), and one way Toyota assures that is by limiting the SOC to a small range, from about 25% full to 80% full.
Priusplus is adding a separate "traction" battery, that works with the normal Prius drivetrain, to provide a long-distance EV mode. In their first proof-of-concept car (which should be finished this weekend) it uses 12 motorcycle Lead-Acid batteries, and it should go about 20 or 30 miles on an overnight (or overday) charge. Using far superiour Lithium Ion batteries, they should get about 80 miles for a battery pack that costs about $5,000 or so (although current Lithium cells are quite small indeed, requiring a rediculous number of batteries wired into a large pack)
If I could go, say, even 40 miles on a charge, I wouldn't use the gas motor in my Prius except to climb very steep hills during the week. I'd effectively get well over 100 mpg (Electricity costs, even in California, give a price-per-mile of about 2 cents. Unfortunately, at this point, the cost for the traction battery (because it is more deeply cycled it doesn't last as long) probably adds another few cents/mile.
PriusPlus is hoping to display there car at a show here in Los Angeles at the end of the month, and is attempting to persuade Toyota that this is a car they should build. Once people are educated about the benefits of hybrid technology, it should be a small step to show them the further benefits of plugging them in.
I fervently hope that PriusPlus will succeed!
Thad
I watch as many as I can, they're pretty easy to see from Los Angeles, on a clear day (and most of them are!)
I've seen the Delta II launch of both Ikonos and Gravity Probe B. Both of these were daylight launches, and would have been impossible to see if they didn't have extremely tight (and nicely publicized by Aviation Week) launch windows. If you know where to look, they're quite nice to see. The arc of the rocket as it bends over and smoothly accelerates to the south is math in motion, just beautiful.
I've also watched a couple of the Minuteman launches testing missle defense systems. Again, these were well publicized events. On my street in Calabasas for the last one, everybody was out in their lawn chairs waiting for it. It did not disappoint. Compared to a satellite launch, the Minuteman gets out of the atmosphere in a hurry, and the solid fuel exhaust blooms into a huge flower-shaped colorful cloud once it is in space.
Still, it's basically impossible for me to see the beauty in a Minuteman launch. It's a goddamn ICBM, its only purpose to kill millions of people.
Any you idiots picking on the maintainer of the site -- get real. Read the site, there is absolutely nothing there than any third-grader couldn't figure out with ten minutes, a road atlas and a blunt crayon. There are a few other good Vandenberg launch sites out there, too, like
this one
I've fantasized about burying a cellphone near Vandenberg, and set it up to call me when it feels the vibration of a launch. It'd be cool, cheap, and easy. Obviously the solar charger and antenna would have to be above ground. The problem with most Vandenberg launches is that you don't know when they are going to happen -- but if you knew they were firing you could just step outside and see.
Thad Beier
I agree with the above poster, that the article is worth reading -- it's a good story, well written. Typically Canadian :)
I think RFID tags are a good idea, but are overkill. A unique holographic image that hashes to a printed value on the face of the bill should do the job. Building unique coded holograms isn't really too hard. The hash algorithm that interprets them could be kept in a tamperproof chip. Obviously the bill checker would just say "Yes" or "No" -- you wouldn't be able to ask it what the hash of a particular hologram was.
Thad
It's an unmitigated disaster. If I was to release the color correction tools I use at Hammerhead as Open Source software, for instance, there is no small chance that I would be sued by somebody, or more likely several somebodies, for infringements on their color correction patents. This kind of stuff is patented out the wazoo, and (unfortunately) the only thing that keeps the patent monster at bay is the fact that everybody does work secretly. That, and the fact that Hammerhead is so small that it's not worth suing. Note well that studios are sued over almost every successful movie they make by people alleging the most tenuous copyright infrigement. A typical example is here.
Publishing open source software does have a tremendous advantage, though, in that it is a perfect vector for publishing information that could be used as prior art when trying to defend against other patents -- so open-source is a two-edged sword (or maybe a sword that is honed sharp at both ends.)
Perhaps, just perhaps, there is a solution. It might not be impossible to have anonymous open-source, with guaranteed anonymity provided much the same way the Cypherpunks' MixMaster remailer network works. That way, one could contribute to open-source projects, and share the benefits of your work with others, without exposing yourself to patent suits.
I'm not sure how one would do this, and the network of visual effects studios might be too small -- and the coding styles of the few hundred programmers might be too distinctive -- to have this work, but it could be interesting.
Thad Beier
Well, mostly right. Larry Gritz did write BMRT, the Blue Moon Rendering Tools, and he did go to Pixar and was one of the leading people on the RenderMan team for quite a few years.
He left to form ExLuna, and ExLuna was then bought by Nvidia. At Nvidia, Gritz and the rest of his team are behind the new Gelato hardware-assisted high-quality rendering product. It's pretty cool. [pun reluctantly admitted to]
It's all a very long and interesting story, that unfortunately will not fit within the margin here.
Thad Beier
As I recall (and I'm certain that others will correct you further) there is no closed-form solution for the three-body problem. The shapes of the orbits cannot be written down as a simple equation -- where (neglecting relativity) two orbiting bodies trace perfect ellipses.
On the other hand, you can calculate a solution to the three-body problem to any level of accuracy that you are interested in, without much effort. Yes, it's an approximation, but so is any calculation.
Thad
As usual, there is more to the story. Apple brought my company in on a project back in the mid 80's when they bought the Cray. While we had to sign an NDA in blood, I doubt anybody will mind me talking about it now, almost 20 years later.
Apple was trying to design a new cpu chip. It would have had vector processing capabilities not all that different from the Cray, so they bought the Cray both to do circuit simulations on the chip and as a model for their own design.
The chip was going to be a 100 MHz chip (an astonishing speed for the time) with a four-pipleline vector processing unit.
They considered (but eventually declined to) hire us to develop some kind of 3D desktop for the Mac. The idea was this would distinguish the Mac further from other computing systems, but they wouldn't be able to emulate the interface because they didn't have the horsepower.
Anyway, that's the Apple-Cray story as I understand it. I'm sure that there is a lot more to the story than I know, of course.
Thad Beier
HTML doesn't have a 'rant' tag, but consider the following as such.
I personally cannot imagine having spyware on my machine, and I similarly cannot imagine any Linux user tolerating it. Most Linux users chose it, in large part, because of the control it gives you over everything that your computer does. Having your computer hijacked by advertisers is antithetical to that concept.
But I watch Windows users tolerate truly mindboggling amounts of adware/spamware/malware. The typical windows users tolerate 100 times what I would consider completely unacceptable.
I know it's elitist to say this, but what happens is that Windows users will make the tradeoff of malware to allow them to steal music and other content. They don't protest, because deep down they know what they're doing is wrong.
Linux users, typically, have no such guilt and therefore don't tolerate that kind of intrusion onto their computer.
Thad
As I recall, a linear combination of RGB can express any possible color -- if you allow for negative amounts of the components. A really bright yellow might be 1 R + 1 G - .2 B for example.
That's still a linear combination, but just one that's not particular useful in the real world of phosphors and filters.
Thad
Greg Brandeau of Pixar had a birds-of-a-feather meeting to discuss the use of Linux in visual effects and animation. It was well attended and Greg made a great presentation.
The main issue that he wanted to address was the need for visual effects studios to present as united a front as possible, so that people who write the software that we use (typically called ISVs) will be able to do Linux ports with some confidence. Greg points out that historically people had used Red Hat 7.x, and that companies like Nothing Real and Alias were able to make Linux versions of their programs without too much trouble. But now, most studios find Red Hat Enterprise licenseing unattractive, and everybody is choosing a different platform. Greg's poll found some companies running Fedora Core 2 (Pixar is doing this, as is my company,) others are using Suse and some are using RHE 3.0.
The problem we face is that if we balkanize too much, it will be impossible for ISVs to support us. Even though all the big visual effects and animation studios use Linux extensively, it is still a very small part of the market for these tools. Alias says that Linux is a distant third place in the number of licenses, after Windows and OS X. (Interestingly, the second place Linux distribution for Alias customers is TurboLinux, because of its strong Asian language support.)
Several vendors were present at the meeting. NVidia said that they really try hard to qualify on every Linux distribution -- they test on dozens of different distributions (some don't work). Intel, on the other hand, said that any past the top two or three just cannot be supported. In particular, because Linux is so customizable, it is often impossible to replicate bugs and problems, because it's unknown what the customer's environment is.
A general solution presented by Greg, which seemed reasonable, was to suggest to the ISVs and IHVs that they qualify for two systems, say RHE and Suse 9.1. Studios would have a stock system of one of those systems, and would test support issues on those stock systems to see if the problem was with the vendors stuff or was due to customization done at the studio. There would be a need to adjust what these two systems might be, over time, as things evolve. Greg suggested that Pixar would be happy to sponsor a web site and discussions on an ongoing (every six month, say) basis to update these reference systems.
John Carey gave a talk on the differnce between a distribution and a operating system -- basically presenting the challenge that ISVs face when trying to write to the Linux market.
The guy from Intel (I forget his name) suggested that Linux Standards Base (LSB) 2.0 is an attempt to address these problems -- after all, they are not unique to the visual effects industry. While he recognizes that LSB 2.0 is not a panacea, it will go a significant part of the way to help standardize Linux from the ISVs point of view.
Dan Wexler of NVidia noted that it's even harder than other people thought. He is working on abusing graphics cards in unconventional ways, and has found that he has significant problems with combinations of motherboards, graphics cards, and memory. Perhaps the reference systems mentioned above will have to specify hardware as well as software configurations.
Andy Maltz from the Academy of Motion Picture Arts and Sciences made a presentation as well, talking about the Academy's revivified Science and Technical Council. He suggested that the Academy was very interested in helping solve technical problems, and providing a forum to share solutions between studios.
All in all, it was a good meeting. My thanks to Greg for putting it together, and maintaining relentless good cheer through some difficult questions and issues.
Thad Beier
Hmmm, but what if you could fill the wings with vacuum? Huh? How about that? Lighter than helium, even!
thad
It turns out that both the Prius and Ford Escape hybrid transaxles (the heart and soul of the car) are made by the same company, Aisin.
So, Ford actually will benefit from the years of Prius experience, to some extent.
Ford also cross-licensed the Toyota hybrid patents.
[disclaimer: I just bought two Priuses -- I love the car!]
thad
Hybrid engines have a solid fuel and a liquid oxidizer. The article says that the DaVinci rocket uses Nitrous Oxide (a liquid if kept under moderate pressure) as the oxidizer, and something other than rubber (I'm guessing candle wax) as the fuel.
Here is the article that describes the Stanford research. It's great work.
Thad
TFA says that the DaVinci group is planning on using a hybrid rocket engine, using nitrous oxide as the oxidizer (as Rutan's SpaceShipOne does) but using something other than synthetic rubber as the fuel. That does make a little bit of sense, as after Rutan's group settled on their rocket engine design, there has been some spectacular research out of Stanford using paraffin (in the American sense of the word) as hybrid rocket fuel. Paraffin has the nice property that it as it gets hot it turns from a solid into a very free-flowing liquid -- which lets it burn very quickly (something that rubber-burning hybrid motors have a hard time with -- the Rutan engine has four separate channels through the fuel to allow it to burn quickly, this leads to the possibility of blowing chunks of propellant.)
Unfortunately, though, the DaVinci website says that their ship will use Kerosene/LOX as the propellant and oxidizer. They have pictures of the engines, including some test firings, on the web site.
You just don't change engine technology at this point in the project.
The only possibilities are that these people are 1) insane or 2) scam artists. It's too bad, it would be spectacularly great if they were on the up-and-up...but...it doesn't appear that they are.
Thad
kfhickel says > Umm, RTFA, dude:
SCO says it discovered the e-mails in a mountain of documents IBM produced in discovery related to SCO's lawsuit against IBM over the Linux operating system.
Nevertheless, you cannot go on fishing expeditions through discovery documents for alleged infractions that weren't part of your original complaint. Furthermore, and far more egregiously, you cannot share with the world information you found in confidence during discovery.
SCO has really screwed the pooch here. Darl is going to be found in contempt.
That said, it is hard to hold back when reporters come calling. Dan Lyons, in particular, is good at wheedling info that probably shouldn't be shared.
Thad Beier
There are significant advantages of triple DES.
1) DES has been around a long time. People have attacked it for years, with every new and old technique of cryptanalysis. DES was created by IBM with help (no, really!) from NSA -- it was NSA that proposed adjustements in the S-Boxes that made DES more resistant to differential cryptanalysis. DES has proven to be secure, except for the obvious key-length problem, in the very best way you can prove an algorithm secure -- by having the best minds on the planet beat on it mercilessly for decades.
2) DES hardware exists, and is inexpensive and relatively secure. Using current hardware to impliment triple DES is easy.
3) It's upward compatible with existing systems. Using the same key three times yeilds is the same as doing regular 56-bit DES (The second DES is usually set up in decrypt mode.) One could well argue that the interoperability with single DES is not really a win, though -- that it allows users to be insecure.
While AES is great, and has been vetted as well as can be expected in the few years since its invention, triple DES is not a bad alternative.
thad
There have been many reports recently of virus writers attempting to blackmail companies. Having this virus, an obvious DDoS attack on Google, happen the same day that Google announced the price of its IPO shares is just what you would expect if the Google didn't pay the blackmail.
I don't know how we'll ever be able to test this hypothesis, but I think that something stinks here.
thad
It is a wonderful feature of fonts that the shape of the glyphs themselves is well-nigh uncopyrightable. Which is the reason that Postscript fonts are programs -- the programs *are* copyrightable. The names of the fonts, as well, can be protected by trademark and potentially by copyright.
But there's nothing to prevent one from making a beautiful font that looks extremely like Palatino or Times Roman for Linux -- except that it would take a significant amount of effort.
I worked with the management of Bitstream back when they were creating their font library back in the mid eighties. They hires famous font designers to recreate things that looked a lot like familiar fonts. They had something called "Swiss" which looked a lot like the normal "Helvetica" (Helvetica is latin for Switzerland). They hired Zapf to create their version of Palatino. Bitstream, of course, made beautiful fonts on their own, too -- most notably the Lucida family.
Thad Beier