I did, and we got a fabulous diamond. We went for a fluorescent stone; the blue fluorescence cancels out some color so we could go with a slightly yellower stone -- and anyway, it's cool.
You can search online through many, many more diamonds than you would ever find in a jeweller, and the price is about 20% less for a comparable stone.
It was very surreal, clicking "Add to basket" for something that cost more than my first car...
... three years ago. Same scam. Tracked it down via domain registry and whois, and called the fellow at home. It was being run out of a dorm at UC Berkeley, and the guy claimed that the guess-and-email technique was to save peoples' privacy. I pointed out that he had to be lying because of the cascading emails, and he claimed that it was a bug that would be fixed ``very soon''.
Too bad to see that he's still scamming -- he was very smooth, and I hoped he'd graduate and go into something more innocuous, like pimping.
As a friend of mine pointed out while we were looking at the flight-spare Voyager plaque in the Smithsonian, none of those probes are likely to make it to the distant stars.
The most likely place for Pioneer 10 to be in a few hundred years is Washington, D.C. -- hanging from the ceiling in the Smithsonian...
To those who can't read (or write) them, regular expressions look like line noise. But once you learn to read them you can condense whole paragraphs of spaghetti conditionals into a single, clear (to the initiated), terse line.
For manipulating strings of characters, they are probably the single most important innovation of the last 20 years.
Actually, the plutonium isn't the issue. It's the much faster decaying fission daughters, and they'll largely be gone by then. The plutonium is poisonous, sure, but it's not especially radioactive; and as others have pointed out the bulk of Pu decays (in the absence of a chain reaction) are just alpha particles. Hell, you can buy alpha emitters at Sav-On (Americium-241).
Actually, the engine does charge the battery. Regen braking does too -- that's MG2 (the one on the drive shaft). But if the battery is low, the engine will run even with the car stopped, driving MG1 (the one opposite the engine on the sun gear doohickey) to make electricity and charge the battery.
The motor isn't the problem. It's the batteries. There's not nearly as much energy per unit mass in batteries as there is in gasoline, and driving up (say) a vertical mile just isn't an option in most pure EVs. A weekend outing from 5,000 feet (Boulder, CO) to 9,700 feet (Hessi township) -- almost a vertical mile -- is pretty common around here.
The Prius isn't alternative-fuel -- just alternative-powertrain. It has loads of cool, geeky features that work together to make... (drumroll) an ordinary 4-door family car. While I don't have a family (yet) it's nice to be able to throw some friends in the back. Basically, they've used the high-efficiency powertrain to give you a conventional 4-door car with Geo-Metro-like efficiency.
I've put about 11,000 miles on my Prius in the first year of ownership. About half of them are highway miles on roadtrips; about 10% of them
are short (1-2 mile) hops in town. Its lifetime
fuel economy is about 48MPG. Range is about 500-600 miles. On the highway it consistently
gets over 45 MPG (and I'm not gentle on the throttle -- 70-80 MPH on the level, and I floor it when I'm crossing the Rockies -- I live in Colorado). In town it gets 35-38 for the first mile or two, until the engine is warm -- then more like 48-52.
The Prius has no transmission at all -- just a second differential that shunts power
between two electric motors/generators and the
engine. (How it works).
It's all drive-by-wire: the gas pedal
is just a rheostat connected to the drive computer.
The engine has a lot of cool stuff to it: an off-center crankshaft, variable compression
ratio, and (ISTR) noncircular pistons. Because of the differential it runs at more or less whatever speed the computer wants, regardless of how fast you're going.
For me (in Colorado) a pure EV was right out because of the low energy density of batteries -- it's hard to climb mountains in
a pure EV. The Prius battery is used for load leveling on the engine (gas engines run best when the load is conditioned). Climbing over about 2000 feet vertical at freeway speeds drains the battery, but the computer handles it gracefully and the car just slows down to about 55 mph (on a standard 7% freeway
grade). Conversely, coming down more than about 1000 feet of altitude will fill the battery to the top from regen braking, and again the computer does the Right Thing, using the engine as a conventional compression brake rather than blowing up the battery pack.
So even though the Prius isn't designed for mountain climbing it works acceptably under even strenuous climbing conditions.
The interior is roomy and holds five people with no trouble. The trunk is adequate. The ride is quiet, and the gas mileage sure doesn't stink. Cornering is very good: tight turn radius and surprising traction given the high-mileage tires. Clearance is adequate but low: it's 4 inches under load, though the bottom 2 inches is just a flexible plastic air dam -- so you can get over 5" high obstacles without killing the car.
The Insight gets better gas mileage, looks cooler, and has better acceleration when you actually want it (though I
imagine Prius ROM mods will come out one day that boost the acceleration -- the computer really does use conservative settings), but it's also really tiny -- the Insight is more of a "geek sports car". Toyota went out of their way to make the
Prius look-and-feel like a basic (if plush) family car, and they succeeded.
It's been said before, it'll probably be said again. If you don't like ICANN, just point your DNS client to OpenNIC instead. Democratic name service the way it should be.
It's easy to find and easy to use -- from ICANN-space, try http://www.opennic.net
...or, more accurately, a "Kaypro". Remember Kaypro? The guys who made cheap Osborne clones in the 1980s? They were selling luggable computers like hotcakes, and they couldn't make 'em fast enough. A lot of the manufacturing was done in circus tents because they couldn't move into real buildings fast enough. Sure there was pilferage, but it didn't matter -- they just slapped together more computers and sent 'em out the door. Money was everywhere, investors loved 'em, each model was better than the last!
Then they announced their latest and greatest, six months ahead of release. And everyone decided to wait and buy the new machine when it came out. And there was no more money to make Kaypros. And the business just folded up.
At about $1.20/GB, hard drives are about the same price as CD media and about five times less expensive than DVD media. You can buy an NFS/SMB networked appliance complete with 1TB of disks for around $2,000 -- the price of a high end tape drive and certainly less than a DVD jukebox. And you get the terabyte of storage for free. And you can access it REALLY FAST because it's a hard drive array.
I just finished costing out a 3-petabyte database for a NASA project, and by far the cheapest way to back up data is to write them to hard drives, unplug the hard drives, and stick them in the closet. It's not an archival solution but archival media cost so much more and are so small compared to hard drives, it's ridiculous. For archival stuff we're holding out to see whether Blu-Ray takes off.
Yep -- though there's new, non-radioactive fiestaware (apparently it's still a viable brand!) available in a department store near you. I forget which chain I saw it in -- maybe Macy's...
.. and safe to have around, so long as you don't eat them (these ones are alpha emitters; alpha particles can't penetrate a sheet of paper). They're also unregulated (in retail quantities) so you don't have to get NRC approval to have them.
Polonium: You can buy photographic negative brushes that contain polonium, from good camera shops. The polonium gives off alpha particles that help to discharge static from the negatives as you brush them. $10-$20.
Americium: Smoke detectors contain Americium-241. A tiny speck of it is in the detector head -- the roughly cylindrical gizmo that looks like a stamped-metal flying saucer. $9
Uranium: pitchblende is comparatively easy to find, and of course the infamous 1970s Fiesta Ware is still to be found (though getting more difficult).
I remember teaching my 12-year-old cousin to extract cube roots in her head. Smart girl! The next year she hit the pubescent wall and suddenly math wasn't cool anymore. Damn.
In case anyone actually reads this far down:
HOW TO EXTRACT CUBE ROOTS (In your head if you want)
(1) Guess the cube root. As badly as you like -- 1 is a good place to start for most small numbers. If you have something like <foo>x10^exp, then try 1x10^(exp/3).
(2) Square the guess.
(3) Divide the original number by the square.
(4) Your next guess can be any number between the quotient and your last guess; it is guaranteed to be closer to the answer than your last guess.
(5) Repeat as necessary.
Or, for those in a hurry, you can remember the magic three logarithms ( log 2 = 0.3010, log 3 = 0.4771, log 7 = 0.8451 ); using those three and about 10 seconds you can find the logarithm of any number at all! Then divide the log by three and raise 10 to the quotient.
... at all levels. In the early 90s I TA'd a course in statistical mechanics at Stanford. We got to the inevitable part where you have to calculate the expected wait time before all of the air in the room accidentally ends up under the desk. It turns out to be something like 10^130 seconds -- a very, very long time. The most common answer was "too long for my calculator", because after all most calculators can only go up to 9E99.
How annoying. You'd think they'd just switch to calculating the logarithm of the answer, or divide by 10^75, or something. But, no, "very big" was enough for most. These were Stanford students, too -- supposedly the cream of the (western half of the) nation's crop of students...
The problem isn't that children are a road-block, it's that standards change drastically and much more quickly than the law can.
Twenty years ago the debate was about books like Catcher in the Rye, which contains the word "fuck" and innumerable copies of the word "crap". Then it was about Judy Blume and books like "Are you there, God? It's me, Margaret", which deals with onset of menstruation. Both of these books were banned from public libraries at different times because people regarded them as obscene and didn't want their children viewing them with public resources.
I know very few people who think that Holden Caulfield (the main character in CitR) is too evil for their children, and AYTGIMM is, well, a childrens' classic that most folks think really does belong in public libraries.
Clearly, there are limits to what most children ``should'' be exposed to -- but (A) the limits are slippery and (B) the law is a clumsy tool. CIPA was too restrictive -- the equivalent of a complete library book ban, without the financial benefit of not having to pay to put the banned books in the library.
Well, so they're available virtually everywhere anyway in college, but STDs are still rampant on many college campuses. Even if you foolishly believe she'll stay celibate, it's better for her to have 'em on-hand and give 'em away than to not have 'em around if something, er, pops up.
I am particularly concerned about the "Content Protection Status Report" filed by the MPAA on 25 April 2002, which speaks of an alleged need to ``plug the analog hole'' (their words) in the distribution of copyrighted works such as movies and music.
The MPAA would like to implement universal analog watermarking of their music and movies, and would like to encourage legislation that requires all A/D converting devices to detect (and, presumably, respond to) their watermark signal. This idea is fundamentally flawed, because it seeks to solve a global problem with a local solution. The MPAA's tract contains at least three hidden assumptions that are wrong: (1a) that it is possible to prevent _all_ illicit production of digital copies of analog works, or (1b) that reducing illicit A/D conversion will also prevent illicit distribution of those works; (2) that it is reasonable to require manufacturers of A/D converters -- extremely general devices with applications throughout society -- to include useless (to most applications) detection circuitry; and (3) that it is right to prevent citizens from enjoying the fair use of copyrighted audio and video information that they own. In short, the MPAA idea is wrong because it cannot work; because the restrictions they propose would be overbroad, imposing impossibly difficult restrictions on an entire industry; and because the restrictions would violate the quid pro quo exchange of rights that is the basis of copyright law.
(1) The MPAA proposal cannot work
The MPAA hopes to solve a problem -- the allegedly widespread copying and illicit distribution of audio and video works -- by attacking a slightly different one. The current proposal seeks not to prevent digital copying but to prevent the conversion of analog signals to digital ones, presumably preventing widespread copying by making it more difficult to convert hard-to-distribute analog signals into easy-to-distribute digital ones. The problem is that, once a single digital copy is made, that digital copy may easily be spread far and wide via file sharing or the world wide web. Raising the difficulty of the initial conversion will not prevent the production of high-quality digital copies.
A very similar case now exists with ``pirated'' digital movies which are often recorded in the theater by people with digital video cameras. Considerable effort is required to generate a digital movie by copying it in the theater, but (as the recent pirated pre-release of the _Star_Wars_ Episode II movie shows) once a single copy is available tens or hundreds of thousands of copies may easily be disseminated.
Even if, as the MPAA suggests, ``analog watermarks'' become universal in movies and audio streams, and A/D conversion devices that do not recognize those watermarks become illegal, such A/D conversion will still continue both within the United States (by scofflaws) and outside of the United States (by foreign nationals who are not restrained by U.S. law). But because of the power of the digital distribution medium, even a tiny number of people can let the digital cat out of the analog bag proposed by the MPAA, nullifying the goal that they hope to achieve.
(2) The MPAA proposal unduly restricts an entire industry unrelated to the publishing industry
Analog-to-digital converters are simple, general purpose circuits with uses at every level of society. For example, digital still or video cameras, nearly all modern automobiles, all cellular telephones, all digital telephone answering machines, and most PC computers contain A-D converters that are capable of digitizing music content. A-D converters have too many specialty applications to mention, both as individual modular integrated circuits and as complete appliances.
Preventing analog to digital conversion of copyrighted material would require every such circuit to have a watermark detector and corresponding digital signal processing capability. That would impose undue burden on the manufacturers of such devices.
(3) The goals of the MPAA proposal are flawed and contrary to the spirit of copyright law
The MPAA seeks to prevent certain types of access to copyrighted analog audio and video works, but doing so would also impopse sweeping restrictions on perfectly legitimate activities using those copyrighted works.
For example, once a piece of music is digitized it is possible and (with modern software) easy to analyze exactly nuances of voice and timing that are otherwise very difficult to discern. Furthermore, music is often stored much more compactly in compressed digital form than in the original CD form. By copying and compressing the music on CDs, a music lover can store the equivalent of 2,000 CDs in a single hard drive with less volume than 10 CDs (in jewel cases), reducing the need for large racks of CDs. By digitizing and compressing music signals, users can transfer audio signals to much more mobile and convenient devices for travel, remix sequences of audio tracks, enhance particular aspects of the sound, and generally make much better use of their copyrighted material than they could without doing so.
Similarly, broadcast video is often digitized by the modern digital equivalent of a VCR -- a digital personal video recorder such as those made by TiVo. This enables much easier time shifting and storage of video than is possible with a conventional analog VCR. These uses of broadcast video have been upheld by the Supreme Court for analog recordings, and digital storage is simply a more effective way to engage in these legal uses of broadcast material.
Conclusions
In general it is not wise to restrict the _tools_ required to engage in an activity rather than the activity itself. Doing so requires legislators -- who are wise but not inhumanly so -- to anticipate every possible use of the tool. For example, an unmodified hi-fi stereo amplifier can be used to fill a room with music, as a high precision AC power supply for delicate equipment, for degaussing video monitors, as a sound synthesizer, as a PA -- all legal. But it can also be used for public performance of copyrighted music, for telephone wiretapping, for eavesdropping, or to make unlicensed LF radio broadcasts -- all illegal. Yet we distribute hi-fi stereos without requiring them to have subsystems that prevent them being attached to the telephone network or to a large external antenna, and without limiting the amount of output power (which would prevent using them for public performance).
A/D conversion of electrical signals and even of audio and video signals are general enough activities that the MPAA's ideas are not only infeasible but also just plain wrong.
I didn't post early enough, so this'll probably be invisible, but what the heck.
The device can't work because the main source of heat conduction in metals is electrical conduction. "Huh?" you ask. Well, heat is just incoherent kinetic energy. Metals and other conductors conduct electricity because electrons are free to wander all over the material. But those electrons don't just carry electric charge -- they also carry kinetic energy. Even incoherent kinetic energy.
With a few exceptions (like diamond), things that are strongly electrically conductive tend to be strongly heat conductive as well, because the electrons carry heat around as they shunt around the material. That's the fundamental limit for Peltier coolers: the better the Peltier cooler conducts electricity, the more freely the electrons can travel and the more freely they exchange heat between the two sides of the dissimilar-metal junction. Contrariwise, the more poorly the junction conducts heat, the harder it is to push enough electrons through it -- and the more inefficient it gets! So Peltier cooler designers are stuck between a rock (thermal conductivity causing losses) and a hard place (resistivity causing heating).
The problem with this device is that it still suffers from the same problem that Peltier junctions do: the charge carriers are electrons, and they have to be able to travel freely through the device. Short of some kind of Maxwell's Daemon kind of filter in the short tunneling gap, they have the same problem because the electrons that tunnel through the gap will themselves carry heat.
Their PDF document implies that the electron-carried heat is minor compared to something called "direct conduction", but in fact the reverse is generally true in metals, and they don't seem to discuss much the heat carried by the charge carriers. (I admit I just skimmed the document)
I use Opera. I downloaded the ad-containing version to start with and then, when I realized how k3w1 it was, I sent in the $30 they asked for. It's the fastest and smallest of the full-featured browsers, and that counts for a lot.
Ham packet radio and even CB are subjected to the FCC's rather stringent requirements against profanity and obscenity. I remember this being a big deal when I daydreamed about setting up a packet-radio ISP link in the early 1990s -- even sort-of-innocuous newsfroups like rec.nude could get you into trouble. I'm not sure what's different now with 802.11b.
No, actually, it's Apple's fault for not foreseeing similar types of problem.
Software should never crash, no matter what input you put into it. That's taught in every first year programming class I've ever seen, and it goes double for software that controls peripherals.
Apple has done shit like this before. In 1988, I owned a floppy with the interesting property that inserting it in a Macintosh while the Mac was running would crash the mac.
Yes, Sony are being horrible by breaking the CD standard. No, that doesn't absolve Apple for writing crappy driver software.
The town of Lovelock, NV is on I-80 and used to be bracketed by city approach signs exactly 69 miles out: "LOVELOCK 69". I believe that all of those signs perished when the Stanford Band Shack was demolished three years ago.
The authorities eventually got wise and the signs now read "LOVELOCK 70".
Slashdot Mirror
You can search online through many, many more diamonds than you would ever find in a jeweller, and the price is about 20% less for a comparable stone.
It was very surreal, clicking "Add to basket" for something that cost more than my first car...
Too bad to see that he's still scamming -- he was very smooth, and I hoped he'd graduate and go into something more innocuous, like pimping.
The most likely place for Pioneer 10 to be in a few hundred years is Washington, D.C. -- hanging from the ceiling in the Smithsonian...
Where are moderator points when you need 'em?
To those who can't read (or write) them, regular
expressions look like line noise. But once you learn to read them you can condense whole paragraphs of spaghetti conditionals into a single, clear (to the initiated), terse line.
For manipulating strings of characters, they are probably the single most important innovation of the last 20 years.
Welcome to the federation.
Actually, the plutonium isn't the issue. It's the much faster decaying fission daughters, and they'll largely be gone by then. The plutonium is poisonous, sure, but it's not especially radioactive; and as others have pointed out the bulk of Pu decays (in the absence of a chain reaction) are just alpha particles. Hell, you can buy alpha emitters at Sav-On (Americium-241).
Actually, the engine does charge the battery. Regen braking does too -- that's
MG2 (the one on the drive shaft). But if the
battery is low, the engine will run even with
the car stopped, driving MG1 (the one opposite
the engine on the sun gear doohickey) to make
electricity and charge the battery.
The motor isn't the problem. It's the batteries. There's not nearly as much energy per unit mass in batteries as there is in gasoline, and driving up (say) a vertical mile just isn't an option in most pure EVs. A
weekend outing from 5,000 feet (Boulder, CO) to
9,700 feet (Hessi township) -- almost a vertical
mile -- is pretty common around here.
I've put about 11,000 miles on my Prius in the first year of ownership. About half of them are highway miles on roadtrips; about 10% of them are short (1-2 mile) hops in town. Its lifetime fuel economy is about 48MPG. Range is about 500-600 miles. On the highway it consistently gets over 45 MPG (and I'm not gentle on the throttle -- 70-80 MPH on the level, and I floor it when I'm crossing the Rockies -- I live in Colorado). In town it gets 35-38 for the first mile or two, until the engine is warm -- then more like 48-52.
The Prius has no transmission at all -- just a second differential that shunts power between two electric motors/generators and the engine. (How it works). It's all drive-by-wire: the gas pedal is just a rheostat connected to the drive computer.
The engine has a lot of cool stuff to it: an off-center crankshaft, variable compression ratio, and (ISTR) noncircular pistons. Because of the differential it runs at more or less whatever speed the computer wants, regardless of how fast you're going.
For me (in Colorado) a pure EV was right out because of the low energy density of batteries -- it's hard to climb mountains in a pure EV. The Prius battery is used for load leveling on the engine (gas engines run best when the load is conditioned). Climbing over about 2000 feet vertical at freeway speeds drains the battery, but the computer handles it gracefully and the car just slows down to about 55 mph (on a standard 7% freeway grade). Conversely, coming down more than about 1000 feet of altitude will fill the battery to the top from regen braking, and again the computer does the Right Thing, using the engine as a conventional compression brake rather than blowing up the battery pack.
So even though the Prius isn't designed for mountain climbing it works acceptably under even strenuous climbing conditions. The interior is roomy and holds five people with no trouble. The trunk is adequate. The ride is quiet, and the gas mileage sure doesn't stink. Cornering is very good: tight turn radius and surprising traction given the high-mileage tires. Clearance is adequate but low: it's 4 inches under load, though the bottom 2 inches is just a flexible plastic air dam -- so you can get over 5" high obstacles without killing the car.
The Insight gets better gas mileage, looks cooler, and has better acceleration when you actually want it (though I imagine Prius ROM mods will come out one day that boost the acceleration -- the computer really does use conservative settings), but it's also really tiny -- the Insight is more of a "geek sports car". Toyota went out of their way to make the Prius look-and-feel like a basic (if plush) family car, and they succeeded.
said again. If you don't like ICANN, just point
your DNS client to OpenNIC instead. Democratic
name service the way it should be.
It's easy to find and easy to use -- from ICANN-space, try http://www.opennic.net
Then they announced their latest and greatest, six months ahead of release. And everyone decided to wait and buy the new machine when it came out. And there was no more money to make Kaypros. And the business just folded up.
I just finished costing out a 3-petabyte database for a NASA project, and by far the cheapest way to back up data is to write them to hard drives, unplug the hard drives, and stick them in the closet. It's not an archival solution but archival media cost so much more and are so small compared to hard drives, it's ridiculous. For archival stuff we're holding out to see whether Blu-Ray takes off.
Yep -- though there's new, non-radioactive fiestaware (apparently it's still a viable brand!) available in a department store near you. I forget which chain I saw it in -- maybe Macy's...
.. and safe to have around, so long as you don't eat them (these ones are alpha emitters; alpha particles can't penetrate a sheet of paper). They're also unregulated (in retail quantities) so you don't have to get NRC approval to have them.
Polonium: You can buy photographic negative brushes that contain polonium, from good camera shops. The polonium gives off alpha particles that help to discharge static from the negatives as you brush them. $10-$20.
Americium: Smoke detectors contain Americium-241. A tiny speck of it is in the detector head -- the roughly cylindrical gizmo that looks like a stamped-metal flying saucer. $9
Uranium: pitchblende is comparatively easy to find, and of course the infamous 1970s Fiesta Ware is still to be found (though getting more difficult).
Likewise if you code a quicksort where a bubblesort would be more appropriate! (those constant terms can sure kill...)
Heh! Cool.
I remember teaching my 12-year-old cousin to extract cube roots in her head. Smart girl! The next year she hit the pubescent wall and suddenly math wasn't cool anymore. Damn.
In case anyone actually reads this far down:
HOW TO EXTRACT CUBE ROOTS (In your head if you want)
(1) Guess the cube root. As badly as you like -- 1 is a good place to start for most small numbers. If you have something like <foo>x10^exp, then try 1x10^(exp/3).
(2) Square the guess.
(3) Divide the original number by the square.
(4) Your next guess can be any number between the
quotient and your last guess; it is guaranteed to be closer to the answer than your last guess.
(5) Repeat as necessary.
Or, for those in a hurry, you can remember the magic three logarithms ( log 2 = 0.3010, log 3 = 0.4771, log 7 = 0.8451 ); using those three and about 10 seconds you can find the logarithm of any number at all! Then divide the log by three and raise 10 to the quotient.
... at all levels. In the early 90s I TA'd a course in statistical mechanics at Stanford. We got to the inevitable part where you have to calculate the expected wait time before all of the air in the room accidentally ends up under the desk. It turns out to be something like 10^130 seconds -- a very, very long time. The most common answer was "too long for my calculator", because after all most calculators can only go up to 9E99.
How annoying. You'd think they'd just switch to calculating the logarithm of the answer, or divide by 10^75, or something. But, no, "very big" was enough for most. These were Stanford students, too -- supposedly the cream of the (western half of the) nation's crop of students...
Twenty years ago the debate was about books like Catcher in the Rye, which contains the word "fuck" and innumerable copies of the word "crap". Then it was about Judy Blume and books like "Are you there, God? It's me, Margaret", which deals with onset of menstruation. Both of these books were banned from public libraries at different times because people regarded them as obscene and didn't want their children viewing them with public resources.
I know very few people who think that Holden Caulfield (the main character in CitR) is too evil for their children, and AYTGIMM is, well, a childrens' classic that most folks think really does belong in public libraries.
Clearly, there are limits to what most children ``should'' be exposed to -- but (A) the limits are slippery and (B) the law is a clumsy tool. CIPA was too restrictive -- the equivalent of a complete library book ban, without the financial benefit of not having to pay to put the banned books in the library.
Well, so they're available virtually everywhere anyway in college, but STDs are still rampant on many college campuses. Even if you foolishly believe she'll stay celibate, it's better for her to have 'em on-hand and give 'em away than to not have 'em around if something, er, pops up.
24 May 2002
I am particularly concerned about the "Content Protection Status
Report" filed by the MPAA on 25 April 2002, which speaks of an alleged
need to ``plug the analog hole'' (their words) in the distribution of
copyrighted works such as movies and music.
The MPAA would like to implement universal analog watermarking of
their music and movies, and would like to encourage legislation that
requires all A/D converting devices to detect (and, presumably,
respond to) their watermark signal. This idea is fundamentally
flawed, because it seeks to solve a global problem with a local
solution. The MPAA's tract contains at least three hidden assumptions
that are wrong: (1a) that it is possible to prevent _all_ illicit
production of digital copies of analog works, or (1b) that reducing
illicit A/D conversion will also prevent illicit distribution of those
works; (2) that it is reasonable to require manufacturers of A/D
converters -- extremely general devices with applications throughout
society -- to include useless (to most applications) detection
circuitry; and (3) that it is right to prevent citizens from enjoying
the fair use of copyrighted audio and video information that they own.
In short, the MPAA idea is wrong because it cannot work; because the
restrictions they propose would be overbroad, imposing impossibly
difficult restrictions on an entire industry; and because the restrictions
would violate the quid pro quo exchange of rights that is the basis of
copyright law.
(1) The MPAA proposal cannot work
The MPAA hopes to solve a problem -- the allegedly widespread copying
and illicit distribution of audio and video works -- by attacking a
slightly different one. The current proposal seeks not to prevent
digital copying but to prevent the conversion of analog signals to
digital ones, presumably preventing widespread copying by making it
more difficult to convert hard-to-distribute analog signals into
easy-to-distribute digital ones. The problem is that, once a single
digital copy is made, that digital copy may easily be spread far and
wide via file sharing or the world wide web. Raising the difficulty
of the initial conversion will not prevent the production of
high-quality digital copies.
A very similar case now exists with ``pirated'' digital movies which
are often recorded in the theater by people with digital video
cameras. Considerable effort is required to generate a digital movie
by copying it in the theater, but (as the recent pirated pre-release
of the _Star_Wars_ Episode II movie shows) once a single copy is
available tens or hundreds of thousands of copies may easily be disseminated.
Even if, as the MPAA suggests, ``analog watermarks'' become universal
in movies and audio streams, and A/D conversion devices that do not
recognize those watermarks become illegal, such A/D conversion will
still continue both within the United States (by scofflaws) and
outside of the United States (by foreign nationals who are not
restrained by U.S. law). But because of the power of the digital
distribution medium, even a tiny number of people can let the digital
cat out of the analog bag proposed by the MPAA, nullifying the goal
that they hope to achieve.
(2) The MPAA proposal unduly restricts an entire industry unrelated to
the publishing industry
Analog-to-digital converters are simple, general purpose circuits with
uses at every level of society. For example, digital still or video
cameras, nearly all modern automobiles, all cellular telephones, all
digital telephone answering machines, and most PC computers contain
A-D converters that are capable of digitizing music content. A-D
converters have too many specialty applications to mention, both as
individual modular integrated circuits and as complete appliances.
Preventing analog to digital conversion of copyrighted material would
require every such circuit to have a watermark detector and
corresponding digital signal processing capability. That would impose
undue burden on the manufacturers of such devices.
(3) The goals of the MPAA proposal are flawed and contrary to the spirit
of copyright law
The MPAA seeks to prevent certain types of access to copyrighted
analog audio and video works, but doing so would also impopse sweeping
restrictions on perfectly legitimate activities using those
copyrighted works.
For example, once a piece of music is digitized it is possible and
(with modern software) easy to analyze exactly nuances of voice and
timing that are otherwise very difficult to discern. Furthermore,
music is often stored much more compactly in compressed digital form
than in the original CD form. By copying and compressing the music on
CDs, a music lover can store the equivalent of 2,000 CDs in a single
hard drive with less volume than 10 CDs (in jewel cases), reducing the
need for large racks of CDs. By digitizing and compressing music
signals, users can transfer audio signals to much more mobile and
convenient devices for travel, remix sequences of audio tracks, enhance
particular aspects of the sound, and generally make much better use of
their copyrighted material than they could without doing so.
Similarly, broadcast video is often digitized by the modern digital
equivalent of a VCR -- a digital personal video recorder such as those
made by TiVo. This enables much easier time shifting and storage of
video than is possible with a conventional analog VCR. These uses of
broadcast video have been upheld by the Supreme Court for analog recordings,
and digital storage is simply a more effective way to engage in these
legal uses of broadcast material.
Conclusions
In general it is not wise to restrict the _tools_ required to engage
in an activity rather than the activity itself. Doing so requires
legislators -- who are wise but not inhumanly so -- to anticipate
every possible use of the tool. For example, an unmodified hi-fi
stereo amplifier can be used to fill a room with music, as a high
precision AC power supply for delicate equipment, for degaussing video
monitors, as a sound synthesizer, as a PA -- all legal. But it can
also be used for public performance of copyrighted music, for
telephone wiretapping, for eavesdropping, or to make unlicensed
LF radio broadcasts -- all illegal. Yet we distribute hi-fi stereos
without requiring them to have subsystems that prevent them being
attached to the telephone network or to a large external antenna, and
without limiting the amount of output power (which would prevent using
them for public performance).
A/D conversion of electrical signals and even of audio and video signals
are general enough activities that the MPAA's ideas are not only infeasible
but also just plain wrong.
I didn't post early enough, so this'll probably
be invisible, but what the heck.
The device can't work because the main source of heat conduction in metals is electrical conduction. "Huh?" you ask. Well, heat is just incoherent kinetic energy. Metals and other conductors conduct electricity because electrons are free to wander all over the material. But those electrons don't just carry electric charge -- they also carry kinetic energy. Even incoherent kinetic energy.
With a few exceptions (like diamond), things that are strongly electrically conductive tend to be strongly heat conductive as well, because the electrons carry heat around as they shunt around the material. That's the fundamental limit for Peltier coolers: the better the Peltier cooler conducts electricity, the more freely the electrons can travel and the more freely they exchange heat between the two sides of the dissimilar-metal junction.
Contrariwise, the more poorly the junction conducts heat, the harder it is to push enough electrons through it -- and the more inefficient it gets! So Peltier cooler designers are stuck between a rock (thermal conductivity causing losses) and a hard place (resistivity causing heating).
The problem with this device is that it still suffers from the same problem that Peltier junctions do: the charge carriers are electrons, and they have to be able to travel freely through the device. Short of some kind of Maxwell's Daemon kind of filter in the short tunneling gap, they have the same problem because the electrons that tunnel through the gap will themselves carry heat.
Their PDF document implies that the electron-carried heat is minor compared to something called "direct conduction", but in fact the reverse is generally true in metals, and they don't seem to discuss much the heat carried by the charge carriers. (I admit I just skimmed the document)
I use Opera. I downloaded the ad-containing version to start with and then, when I realized how k3w1 it was, I sent in the $30 they asked for. It's the fastest and smallest of the full-featured browsers, and that counts for a lot.
Ham packet radio and even CB are subjected to the FCC's rather stringent requirements against profanity and obscenity. I remember this being a big deal when I daydreamed about setting up a packet-radio ISP link in the early 1990s -- even sort-of-innocuous newsfroups like rec.nude could get you into trouble. I'm not sure what's different now with 802.11b.
Software should never crash, no matter what input you put into it. That's taught in every first year programming class I've ever seen, and it goes double for software that controls peripherals.
Apple has done shit like this before. In 1988, I owned a floppy with the interesting property that inserting it in a Macintosh while the Mac was running would crash the mac.
Yes, Sony are being horrible by breaking the CD standard. No, that doesn't absolve Apple for writing crappy driver software.
The authorities eventually got wise and the signs now read "LOVELOCK 70".