All that's missing is a few ten of thousands of dollars to pay some programmers to write the software.
Luckily for us, someone's already done it. It's a neat concept called Open Source. The US government paid for EMC, a linux-based CNC controller system. Using OWFS you can make user-based file systems and run multisensor digital temperature and voltage detection systems to control kilns. (I've done this.) There are a plethora of linux-based replacement PLC controller projects running out there.
To replace $300 temperature controllers from the 1980's, used on kilns and heat treatment furnaces all over the world? To replace $400 data acquisition systems from the 1970's, used on process control systems all over the world? At this price you can begin replacing industrial modems, tearing out ancient proprietary CNC controller systems on mills and lathes, retrofit large solar panel charge controller systems with these. There are industries all over the non-first-world that can't afford industrial-quality control systems. These sorts of crummy little computers have 100x the performance and flexibility of old ladder-logic programmable logic controllers, and could be turned into amazingly useful, easily-updated or replaced, manufacturing control systems.
>You care about them, they beat you up, and you keep coming back. Why?
Have you ever heard of cognitive dissonance? If people have two contradictory beliefs, to wit: I paid a lot of money for my Apple, and Apple treats me like a hated stepchild, those people resolve the contradiction by modifying one of the beliefs to match the other. Once they've paid for something, many people suffer from sunk cost syndrome and are unwilling to acknowledge that they made a bad decision. So they just decide that they aren't being treated badly after all, that whatever the company is doing is a good thing, and they just don't understand the reasons for it.
It's why soldiers look back on boot camp nostalgically, why many people believe in religion, blah blah blah. It's a huge motivator in human belief systems. And, usually, it's based on insecurity: many people would rather suffer financial, social, and emotional loss than admit they made a bad decision.
Thing is: it was a *perfect* ending. It was exactly what was supposed to happen. If you take a bunch of English comp classes, one of the things you bring home is the current idea of a well-formed story: that when you read the ending, it offers a surprise, some sort of insight, but once you've read it you know it's the way it should've ended. That's what Stephenson wrote. I just wish he could do the same thing with Snow Crash and the like. (I'd argue that in some ways he pulled off a pretty good ending with Snow Crash, actually, and with Zodiac, but neither was as pleasing as The Big U.)
Yes he does. Read "The Big U". It has a very nearly perfect ending, after being hundreds of pages of crazy raving that only a very bright writer desperately homesick for dorm life would find worthwhile. And then this wonderful ending.
I think he spent his lifetime supply of wrapping-up on that one book, and now he's stuck with the rest of his books ending like life: just sort of wandering off aimlessly.
FWIW my dad's 1964 Ford had sodium-filled exhaust valves, and I'm told many higher-end Mercedes these days have the same.
The problem comes when you're using the engine for drag-racing, and a connecting rod fails, slamming the piston into the red-hot exhaust valves. Things get very exciting.
Okay, let's make sure we're speaking with the same terminology. Hot things emit heat by three methods: conduction, convection, and radiation. That is: transfer by contact; transfer by heating some fluid that moves away from the hot source and heats other things; and by shooting out photons that are absorbed by other things, heating them up. Light *is* heat, although we generally look at heat as being the photons that are just slightly longer wavelength than what we can see.
Incandescents emit, depending on what sources you look at, something like 70% of the power they consume as radiated heat: they're blasting out *vast* quantities of infrared. Only a very small part of their total power use comes out as visible light. A hot object ejects photons over a wide frequency, as a boltzmann distribution (if I remember right.) So we see the tip of the iceberg, essentially, and all the rest is just heat. An incandescent does dump some heat through conduction -- something like 5% of its total usage, which is why light bulb sockets get hot. An LED is almost exactly the opposite. It doesn't radiate *any* IR. (Well, that's not entirely true, but the IR that's coming off it is pretty small, and is because it's hot, about which more in a sec.) But it dumps something like 30-60% of its total drawn power through conduction. The die itself gets really hot, unless it is tied to a heatsink.
But, see, here's the thing: it's producing a *lot* more visible light per unit power consumed than a lightbulb is. So you can use an LED that draws very little power and still get lots of light out of it, meaning that the standard LEDs people have played with for years don't have any problem with heat because they draw so little power that they can dissipate all the heat they're generating even though they're encapsulated in plastic.
That's not the case with new so-called ultra-ultra-ultrabright LED's. (I kid you not: I've seen them advertised that way.) They're dumping out a *lot* of heat, so the die itself is bonded to an aluminum heatsink, which sometimes is sufficient, but in many cases the heatsink then has to be plated with something we can solder to on the back side, and that's bonded in turn to a bigger heatsink yet.
Take a look at these lights -- looks just like a CPU fan, doesn't it? Now in part those are intended to look awesome, but I can tell you, having worked with some very similar to that, they get uncomfortably warm if they're in any sort of enclosure, like a canlight mounted in the ceiling.
So to more or less answer your question: if you put the same amount of power through an incandescent and an LED, you're going to get similar amounts of heat produced. (The LED will produce less, but not an order of magnitude less.) However, you'll get many times more visible light out of the LED. For the same brightness, an LED will produce many times less heat. But you still have to get that heat out of the LED if you want it to live a long time.
Oh, and the other question: the 'bulb' is usually just a little bleb of epoxy on top of the die, to spread the light rather than just having it shoot out like a searchlight. If it does get hot, it does so only because the die is conducting heat into it, which is evidence of bad design (not enough heatsinking.)
There are many, many LED drivers that are dimmable. Take a look at the National LM3405 or 3423: both accept a pwm signal for dimming, that can run from 10 Hz up to 200 kHz. (If it's under about 2kHz it's not a dimmer, though: it's a flasher, and it's really annoying.)
The trick is getting a replacement LED lightbulb, that fits in an existing socket, to work with an existing, cheap wall-mount dimmer. That'll be on the market within, uh... well, before mid-2009, certainly.
By the way, LED's have heat problems. In fact, that's probably their biggest drawback right now: they're flaming hot on the die itself. They don't *emit* any heat via radiation, but they dump vast quantities of heat through conduction. Current in-house LED replacement bulbs are the same size as an existing incandescent lightbulb, with this tiny little 2mm x 2mm die putting out the light, and the whole remainder of the volume being a big metal heatsink to try and soak all the heat out of that little die.
My company designs LED driver chips. All our stuff is running in the very high kilohertz or low megahertz range -- far, far above where humans can see. All the drivers we've torn apart from our competitors are likewise quite high-frequency, usually 600 kHz - 1.2 MHz, but occasionally as low as 40 kHz.
LED drivers, if they're not just using a big power resistor, are all (that I know of) current-limited, chopping power supplies. It doesn't make *sense* to build those to run in the hertz range. I'm guessing whatever you saw was something broken, or in the process of breaking, and not the original designer's intent.
For what it's worth, one reason trucks have gone to LED's is that they don't die because of vibration, like incandescent bulbs do. We don't really know how long LED's will last, but if they're correctly designed, they should still have at least 50% brightness at 50,000 hours of operation.
All the LED lighting solutions my company is building, and all the ones from our competitors that we've been buying and taking apart, are screw-in replacements for existing bulbs. Every single one, without an exception. If car companies wish to make their cars more proprietary by specifying custom LED lights, well, they're free to, but I think it'd be a stupid move.
The amiga was initially just to get 10 megs (the size of the hard drive) copied off the amiga and onto the linux machine, without using the floppy drive. (If you remember Amiga floppy drives you'll know why. If you don't, it's because I don't hate my neighbors and it scares my dog.)
The function generator, I use to produce very precise timing for a bunch of stepper motor drivers I'm using for CNC machining stuff. But, in general, it's really nice to be able to interconnect test&measurement equipment to computers: I can precisely measure the curve of current flow into a disk drive as it spins up and seeks/writes, at 50 measurements a second, with a computer interfaced to a digital multimeter. I have a plan, involving a strain gauge glued to a milling machine bit, and connected through a DMM, so I could do real-time calculation of tool deflection and not break off any more milling bits because I was making too heavy a cut. Stuff like that.
You sure about that? I think you're counting the same biased way the BSA is. I bet your printer driver program's not pirated. Ditto your keyboard-scanning program, or your web browser, or the default text editor or the program that displays the load average and the currently running processes. How about the program that shows what hardware is connected? Or the one that defragments the hard drive? There are hundreds of programs interacting to keep a computer running, and only a very few ones right on the very tip-top are still sufficiently unusual to warrant purchasing.
Now, if you're running a pirated OS, then you *are* running probably 75% pirated stuff, in a manner of speaking, and there are a lot of people doing that. But even then I'm betting it's nowhere near 1/5 or even 1/50th of the people in the US simply because computers are so cheap and installing an OS is something that most people have not the ability, motivation, or willingness to do.
Part of my point is that the BSA, and the companies the BSA represents, have a problem: every year, fewer applications need to be bought because there are free versions that work well enough. We don't expect to buy driver programs hardware -- they're free. We don't expect to buy internet connectivity programs -- they're free. Many window managers, garbage collection programs, data compression programs, codecs, bluetooth protocol stacks, are free. What the BSA is saying is that 1/5 of the OS installations, Photoshop installations, and Excel installations are pirated. That's possible. But right now my computer is running 72 processes, and only ten of those are programs that can be purchased. You can't pirate free stuff, and an increasing amount of software, even on Windows systems, is free stuff. That, in my opinion, is what the BSA is really concerned with in the long term.
I'm not what you'd call an early adopter. That particular machine has a GPIB card in it to talk to my 1978 function generator, and a serial connection to my 1989 Amiga, which between them have dozens of other not-pirated programs running every time they're powered up.
(I swear if I could get my 1964 oscillator interfaced to my computer I'd jump at the chance...)
I have three linux machines at home. Every time I fire one up I run several hundred 'programs', including X, Qt, the TCP/IP stack, flash, firefox, amarok, ipchains. My two little headless linux computers, one disguised as a DSL modem and the other as a firewall, likewise run at least dozens of programs. I know there are tens of thousands of computers hosting websites all over the world that, likewise, are running dozens of 100% free programs. For their 1 out of 5 statistic to be right, within the United States, there must be a dozen people running nothing but pirated software just to make up for me. I know nine other people who are, likewise, running multiple computers, including several Windows machines, that have 100% legit/free programs on them. So now we're up to a hundred or so people running nothing but pirated software just to make up for me and my nine friends.
Are there vast underground barracks filled with armies of illegal software users in Ohio and Florida? Is China outsourcing its goldfarming to the ghettos of East LA?
That's the fundamental, egalitarian mindset that produced the statement "all men are created equal." We all love to think that, and it'd be nice if it were true. But it isn't. You could spend more than 10,000 hours trying to get my girlfriend to be good at math. In that time she could certainly learn huge amounts of exotic math concepts, but she'd never be able to reliably add two numbers because she is physically unable to consistently recognize numerals and keep them in order. I used to be a bike racer, and I was pretty good at it. But no matter how much training I did now, at 40, I would never be able to race on a professional level again because my mitochondria just aren't up to it anymore. And then there's our friend Z, who is a really nice guy. The thing is: he's just not very bright. He only got his GED when he was 24, and that was really difficult for him. No matter how much you worked with him, no matter how much training, he can't even consistently drive a car well. He's only sort of literate, despite years of work with therapists and teachers. He's just, well, kind of retarded.
There are jobs which are deterministic: trash collection. You go to all the houses and pick up the trash. Pilot: you fly the plane and deal with storms. In some ways, even many doctors have deterministic jobs with vastly more complicated flowcharts, that end up at one of several points. But even those jobs have challenges that are simply beyond some people. But jobs that rely on innovation, are harder yet, because there isn't a learnable process that results in innovation. Extrapolation, and more specifically extrapolating ideas that are not only right but useful, is currently still a matter of having a prepared mind and a talent for seeing things first. I don't think that's something that can be taught to the vast majority of humanity.
>The article says that he was doing 62 MPH according to the radar gun. The GPS says 45. If the GPS was right, why was the gun wrong? Bad calibration? Operator error? Dyslexia?
A very cute girl I used to date told me one way radar guns read wrong because she'd experienced it a lot of times on the military base where her father was an officer. MP's who wanted to pull her over and talk to her would drive at 50 mph towards a wall and trigger the radar gun, then leave the 50mph reading on the gun and wait for her to drive by. These days, guns are interlinked into the police car so they subtract out the car's speed, giving an accurate reading when both the gun and target are moving, and I believe they have timestamping, so this kid probably had something else happen. One possibility is that it had particularly spectacular wheels. By which I mean, I was once radar-clocked on my bicycle doing 60 mph on flat ground. My bike's carbon fiber but with steel spokes, which are travelling at 2x the speed of the bike at the top of the wheel.
>As such, I've lost some of the nerves in my left hand from a thermite burn. However, you'd better bet I'm careful with pretty much anything explosive now. And, in the grand scheme of things, the small bit of nerve damage was worth a deeply ingrained caution for all things explosive and hot.
In my family I was the adventurer and my brother, younger, watched and learned. I jumped off stairs on my bike and broke some ribs. He learned to just ride down them rather than trying jumps. When I was dating (well, okay, I still am) I dated some really attractive, crazy women and had huge fights and all sorts of problems. He met a nice, not-crazy, regular woman and married her immediately. The one exception to this rule is where he didn't get to watch me be stupid: chemistry. I was always really careful, wore safety glasses and gloves, and used small quantities of everything. Perhaps as a result, he was the one who blew a brass cartridge reloaded with more exciting stuff through his arm.
My point being: yes, experience is a very good teacher, but *vicarious* experience, aka observation, is even *better*, because afterwards you can still count to ten with bifocal vision, if you know what I mean.
>But that begs the question exactly what was the USA thinking when they deployed those missiles to Turkey so close to the Soviet border? Did they really think the Soviets wouldn't react?
I don't know if this is true, but in "Dark Sun", a book documenting the development of the hydrogen bomb, Richard Rhodes claims that from roughly 1948 to when Powers got shot down in 1960, the United States had surveillance aircraft flying over the USSR 24 hours a day, seven days a week, and in the late '50's the US ran practice bombing runs of several dozen bombers over major western USSR cities during daylight. I also recall reading, possibly in Ben Rich's "Skunk Works", that even after the Powers shoot-down, the USSR's main/only defense against spy overflights was to just put up as many aircraft as possible, flying beneath the US aircraft, to try and physically obscure the ground.
Likewise, again from the Rhodes book, Curtis LeMay was openly calling for a pre-emptive, wide-spread nuclear strike on the USSR because he felt that it was the only reasonable defense. Lemay was the Strategic Air Command general who had the authority to launch nuclear weapons.
So, yeah, I'd say the US perception of USSR threat was a little different than the USSR perception of US threat. I'm completely amazed we didn't end up having a nuclear war, and I think the only reason we didn't was because the USSR was astoundingly reluctant to react to the repeated US aggressions. If they'd done 1/10 the things we did, I seriously doubt we would've shown the restraint they did.
That's not unusual: the cheapest test subjects are always the people doing the research. When Mary Roach was writing "Bonk" -- a book devoted to medical research on sexuality -- she wanted to see the process, not just the results, of a doctor running an MRI and imaging people having sex. The people who did such research said there was absolutely no way she could watch unless she supplied the test subjects. (She and her husband went. Uproarious, really interesting book to read.) Likewise, in undergrad microbiology courses, it would be unbelievably expensive and difficult to hand out human blood samples to students to have them, say, look at live blood cells under a microscope, so students routinely draw samples from themselves, because that way there's almost no risk of contaminated blood. My ex-gf built ultrasound transducers for medical imaging. When the company was getting started, everyone who worked there was a man (engineering, yaaaay.) So their first breast imaging setup, they hired a hooker, because they couldn't find a cheap test subject. The prostate imaging? they flipped a coin, or drew straws, as to who got to be on which end of the test equipment.
Well... the height of the wave is proportional to the depth of the water below. If you're in a boat out over deep ocean you won't notice it, but if you're in the harbor where the devastation is happening you sure will. If the eyewitnesses were in somewhat shallow water they'd probably see a lot of excitement.
Personally, I like SL because people can build beautiful artwork -- reproductions of 1850's-era Japanese teahouses, for instance -- that I can wander through without having to, y'know, fly to Japan.
It's also nice that since I've assiduously ignored all the morons and loudmouths and hung out patiently, I've met and become friends with a mechanical engineer who teaches computational flow dynamics and who answers lots of questions; a DJ who has great taste in music and has introduced me to a dozen bands I would never have heard of; the bass player for a punk rock band that I've now gotten to see play live; and someone who is building his own plane (which I plan to do) and can show me, with SL models, what he's having problems with construction-wise with his RL plane. Again, without having to fly to California or Iowa or Georgia or Florida.
It is vaguely possible I could have met all those people in various places online, if I had the patience and tenacity to go looking. On SL I didn't have to: I just kept hanging out and they met me.
And, by the way, I've never paid a dime to use SL. Why would I? That's stupid. It's free if you don't want to 'own' a piece of 'land', whatever THAT means.
Hey, a fellow MOOer! I've been on Lambda since about 1993. I think SL is somewhat like Lambda. Since there's a lower cost of entry, skillwise, there are a lot more people and a lot stupider people. But the programming is definitely reminiscent of moocode. It just sucks that LSL is so limited. If they were using moocode it'd be an enormous help: it was so nice to write display functions and statistics functions and then write code that called those to do data manipulation and presentation, and then use THAT to build database-like stuff. LSL is so... limited. But some people have built some fabulous stuff with their tiny little toolset.
The few wall warts I've poked around with were just a little transformer and a rectifier/cap: no attempt at active regulation. I've started building my own on fussy things, using a wart rated 3v or so higher and a linear regulator soldered to a small bit of copper sheet for a heatsink, with a 1000uF cap across the linear's output. Overkill, but for $7 worth of parts I don't ever have to worry about the quality of the output again.
Slashdot Mirror
All that's missing is a few ten of thousands of dollars to pay some programmers to write the software.
Luckily for us, someone's already done it. It's a neat concept called Open Source.
The US government paid for EMC, a linux-based CNC controller system.
Using OWFS you can make user-based file systems and run multisensor digital temperature and voltage detection systems to control kilns. (I've done this.)
There are a plethora of linux-based replacement PLC controller projects running out there.
To replace $300 temperature controllers from the 1980's, used on kilns and heat treatment furnaces all over the world?
To replace $400 data acquisition systems from the 1970's, used on process control systems all over the world?
At this price you can begin replacing industrial modems, tearing out ancient proprietary CNC controller systems on mills and lathes, retrofit large solar panel charge controller systems with these.
There are industries all over the non-first-world that can't afford industrial-quality control systems. These sorts of crummy little computers have 100x the performance and flexibility of old ladder-logic programmable logic controllers, and could be turned into amazingly useful, easily-updated or replaced, manufacturing control systems.
>You care about them, they beat you up, and you keep coming back. Why?
Have you ever heard of cognitive dissonance?
If people have two contradictory beliefs, to wit: I paid a lot of money for my Apple, and Apple treats me like a hated stepchild, those people resolve the contradiction by modifying one of the beliefs to match the other.
Once they've paid for something, many people suffer from sunk cost syndrome and are unwilling to acknowledge that they made a bad decision. So they just decide that they aren't being treated badly after all, that whatever the company is doing is a good thing, and they just don't understand the reasons for it.
It's why soldiers look back on boot camp nostalgically, why many people believe in religion, blah blah blah. It's a huge motivator in human belief systems. And, usually, it's based on insecurity: many people would rather suffer financial, social, and emotional loss than admit they made a bad decision.
Thing is: it was a *perfect* ending. It was exactly what was supposed to happen.
If you take a bunch of English comp classes, one of the things you bring home is the current idea of a well-formed story: that when you read the ending, it offers a surprise, some sort of insight, but once you've read it you know it's the way it should've ended.
That's what Stephenson wrote.
I just wish he could do the same thing with Snow Crash and the like. (I'd argue that in some ways he pulled off a pretty good ending with Snow Crash, actually, and with Zodiac, but neither was as pleasing as The Big U.)
Yes he does.
Read "The Big U".
It has a very nearly perfect ending, after being hundreds of pages of crazy raving that only a very bright writer desperately homesick for dorm life would find worthwhile.
And then this wonderful ending.
I think he spent his lifetime supply of wrapping-up on that one book, and now he's stuck with the rest of his books ending like life: just sort of wandering off aimlessly.
FWIW my dad's 1964 Ford had sodium-filled exhaust valves, and I'm told many higher-end Mercedes these days have the same.
The problem comes when you're using the engine for drag-racing, and a connecting rod fails, slamming the piston into the red-hot exhaust valves. Things get very exciting.
Okay, let's make sure we're speaking with the same terminology.
Hot things emit heat by three methods: conduction, convection, and radiation. That is: transfer by contact; transfer by heating some fluid that moves away from the hot source and heats other things; and by shooting out photons that are absorbed by other things, heating them up. Light *is* heat, although we generally look at heat as being the photons that are just slightly longer wavelength than what we can see.
Incandescents emit, depending on what sources you look at, something like 70% of the power they consume as radiated heat: they're blasting out *vast* quantities of infrared. Only a very small part of their total power use comes out as visible light. A hot object ejects photons over a wide frequency, as a boltzmann distribution (if I remember right.) So we see the tip of the iceberg, essentially, and all the rest is just heat. An incandescent does dump some heat through conduction -- something like 5% of its total usage, which is why light bulb sockets get hot.
An LED is almost exactly the opposite. It doesn't radiate *any* IR. (Well, that's not entirely true, but the IR that's coming off it is pretty small, and is because it's hot, about which more in a sec.) But it dumps something like 30-60% of its total drawn power through conduction. The die itself gets really hot, unless it is tied to a heatsink.
But, see, here's the thing: it's producing a *lot* more visible light per unit power consumed than a lightbulb is. So you can use an LED that draws very little power and still get lots of light out of it, meaning that the standard LEDs people have played with for years don't have any problem with heat because they draw so little power that they can dissipate all the heat they're generating even though they're encapsulated in plastic.
That's not the case with new so-called ultra-ultra-ultrabright LED's. (I kid you not: I've seen them advertised that way.) They're dumping out a *lot* of heat, so the die itself is bonded to an aluminum heatsink, which sometimes is sufficient, but in many cases the heatsink then has to be plated with something we can solder to on the back side, and that's bonded in turn to a bigger heatsink yet.
Take a look at these lights -- looks just like a CPU fan, doesn't it? Now in part those are intended to look awesome, but I can tell you, having worked with some very similar to that, they get uncomfortably warm if they're in any sort of enclosure, like a canlight mounted in the ceiling.
So to more or less answer your question: if you put the same amount of power through an incandescent and an LED, you're going to get similar amounts of heat produced. (The LED will produce less, but not an order of magnitude less.) However, you'll get many times more visible light out of the LED. For the same brightness, an LED will produce many times less heat. But you still have to get that heat out of the LED if you want it to live a long time.
Oh, and the other question: the 'bulb' is usually just a little bleb of epoxy on top of the die, to spread the light rather than just having it shoot out like a searchlight. If it does get hot, it does so only because the die is conducting heat into it, which is evidence of bad design (not enough heatsinking.)
There are many, many LED drivers that are dimmable. Take a look at the National LM3405 or 3423: both accept a pwm signal for dimming, that can run from 10 Hz up to 200 kHz. (If it's under about 2kHz it's not a dimmer, though: it's a flasher, and it's really annoying.)
The trick is getting a replacement LED lightbulb, that fits in an existing socket, to work with an existing, cheap wall-mount dimmer. That'll be on the market within, uh... well, before mid-2009, certainly.
By the way, LED's have heat problems. In fact, that's probably their biggest drawback right now: they're flaming hot on the die itself. They don't *emit* any heat via radiation, but they dump vast quantities of heat through conduction. Current in-house LED replacement bulbs are the same size as an existing incandescent lightbulb, with this tiny little 2mm x 2mm die putting out the light, and the whole remainder of the volume being a big metal heatsink to try and soak all the heat out of that little die.
My company designs LED driver chips. All our stuff is running in the very high kilohertz or low megahertz range -- far, far above where humans can see. All the drivers we've torn apart from our competitors are likewise quite high-frequency, usually 600 kHz - 1.2 MHz, but occasionally as low as 40 kHz.
LED drivers, if they're not just using a big power resistor, are all (that I know of) current-limited, chopping power supplies. It doesn't make *sense* to build those to run in the hertz range. I'm guessing whatever you saw was something broken, or in the process of breaking, and not the original designer's intent.
For what it's worth, one reason trucks have gone to LED's is that they don't die because of vibration, like incandescent bulbs do. We don't really know how long LED's will last, but if they're correctly designed, they should still have at least 50% brightness at 50,000 hours of operation.
All the LED lighting solutions my company is building, and all the ones from our competitors that we've been buying and taking apart, are screw-in replacements for existing bulbs. Every single one, without an exception. If car companies wish to make their cars more proprietary by specifying custom LED lights, well, they're free to, but I think it'd be a stupid move.
The amiga was initially just to get 10 megs (the size of the hard drive) copied off the amiga and onto the linux machine, without using the floppy drive. (If you remember Amiga floppy drives you'll know why. If you don't, it's because I don't hate my neighbors and it scares my dog.)
The function generator, I use to produce very precise timing for a bunch of stepper motor drivers I'm using for CNC machining stuff. But, in general, it's really nice to be able to interconnect test&measurement equipment to computers: I can precisely measure the curve of current flow into a disk drive as it spins up and seeks/writes, at 50 measurements a second, with a computer interfaced to a digital multimeter. I have a plan, involving a strain gauge glued to a milling machine bit, and connected through a DMM, so I could do real-time calculation of tool deflection and not break off any more milling bits because I was making too heavy a cut. Stuff like that.
You sure about that? I think you're counting the same biased way the BSA is. I bet your printer driver program's not pirated. Ditto your keyboard-scanning program, or your web browser, or the default text editor or the program that displays the load average and the currently running processes. How about the program that shows what hardware is connected? Or the one that defragments the hard drive?
There are hundreds of programs interacting to keep a computer running, and only a very few ones right on the very tip-top are still sufficiently unusual to warrant purchasing.
Now, if you're running a pirated OS, then you *are* running probably 75% pirated stuff, in a manner of speaking, and there are a lot of people doing that. But even then I'm betting it's nowhere near 1/5 or even 1/50th of the people in the US simply because computers are so cheap and installing an OS is something that most people have not the ability, motivation, or willingness to do.
Part of my point is that the BSA, and the companies the BSA represents, have a problem: every year, fewer applications need to be bought because there are free versions that work well enough. We don't expect to buy driver programs hardware -- they're free. We don't expect to buy internet connectivity programs -- they're free. Many window managers, garbage collection programs, data compression programs, codecs, bluetooth protocol stacks, are free.
What the BSA is saying is that 1/5 of the OS installations, Photoshop installations, and Excel installations are pirated. That's possible. But right now my computer is running 72 processes, and only ten of those are programs that can be purchased. You can't pirate free stuff, and an increasing amount of software, even on Windows systems, is free stuff.
That, in my opinion, is what the BSA is really concerned with in the long term.
I'm not what you'd call an early adopter. That particular machine has a GPIB card in it to talk to my 1978 function generator, and a serial connection to my 1989 Amiga, which between them have dozens of other not-pirated programs running every time they're powered up.
(I swear if I could get my 1964 oscillator interfaced to my computer I'd jump at the chance...)
I have three linux machines at home. Every time I fire one up I run several hundred 'programs', including X, Qt, the TCP/IP stack, flash, firefox, amarok, ipchains. My two little headless linux computers, one disguised as a DSL modem and the other as a firewall, likewise run at least dozens of programs. I know there are tens of thousands of computers hosting websites all over the world that, likewise, are running dozens of 100% free programs.
For their 1 out of 5 statistic to be right, within the United States, there must be a dozen people running nothing but pirated software just to make up for me.
I know nine other people who are, likewise, running multiple computers, including several Windows machines, that have 100% legit/free programs on them. So now we're up to a hundred or so people running nothing but pirated software just to make up for me and my nine friends.
Are there vast underground barracks filled with armies of illegal software users in Ohio and Florida? Is China outsourcing its goldfarming to the ghettos of East LA?
That's the fundamental, egalitarian mindset that produced the statement "all men are created equal." We all love to think that, and it'd be nice if it were true.
But it isn't.
You could spend more than 10,000 hours trying to get my girlfriend to be good at math. In that time she could certainly learn huge amounts of exotic math concepts, but she'd never be able to reliably add two numbers because she is physically unable to consistently recognize numerals and keep them in order.
I used to be a bike racer, and I was pretty good at it. But no matter how much training I did now, at 40, I would never be able to race on a professional level again because my mitochondria just aren't up to it anymore.
And then there's our friend Z, who is a really nice guy. The thing is: he's just not very bright. He only got his GED when he was 24, and that was really difficult for him. No matter how much you worked with him, no matter how much training, he can't even consistently drive a car well. He's only sort of literate, despite years of work with therapists and teachers. He's just, well, kind of retarded.
There are jobs which are deterministic: trash collection. You go to all the houses and pick up the trash. Pilot: you fly the plane and deal with storms. In some ways, even many doctors have deterministic jobs with vastly more complicated flowcharts, that end up at one of several points. But even those jobs have challenges that are simply beyond some people.
But jobs that rely on innovation, are harder yet, because there isn't a learnable process that results in innovation. Extrapolation, and more specifically extrapolating ideas that are not only right but useful, is currently still a matter of having a prepared mind and a talent for seeing things first. I don't think that's something that can be taught to the vast majority of humanity.
>The article says that he was doing 62 MPH according to the radar gun. The GPS says 45. If the GPS was right, why was the gun wrong? Bad calibration? Operator error? Dyslexia?
A very cute girl I used to date told me one way radar guns read wrong because she'd experienced it a lot of times on the military base where her father was an officer. MP's who wanted to pull her over and talk to her would drive at 50 mph towards a wall and trigger the radar gun, then leave the 50mph reading on the gun and wait for her to drive by.
These days, guns are interlinked into the police car so they subtract out the car's speed, giving an accurate reading when both the gun and target are moving, and I believe they have timestamping, so this kid probably had something else happen.
One possibility is that it had particularly spectacular wheels. By which I mean, I was once radar-clocked on my bicycle doing 60 mph on flat ground. My bike's carbon fiber but with steel spokes, which are travelling at 2x the speed of the bike at the top of the wheel.
>As such, I've lost some of the nerves in my left hand from a thermite burn. However, you'd better bet I'm careful with pretty much anything explosive now. And, in the grand scheme of things, the small bit of nerve damage was worth a deeply ingrained caution for all things explosive and hot.
In my family I was the adventurer and my brother, younger, watched and learned.
I jumped off stairs on my bike and broke some ribs. He learned to just ride down them rather than trying jumps.
When I was dating (well, okay, I still am) I dated some really attractive, crazy women and had huge fights and all sorts of problems. He met a nice, not-crazy, regular woman and married her immediately.
The one exception to this rule is where he didn't get to watch me be stupid: chemistry. I was always really careful, wore safety glasses and gloves, and used small quantities of everything. Perhaps as a result, he was the one who blew a brass cartridge reloaded with more exciting stuff through his arm.
My point being: yes, experience is a very good teacher, but *vicarious* experience, aka observation, is even *better*, because afterwards you can still count to ten with bifocal vision, if you know what I mean.
And to make your dog sound like a cat? Put it in the freezer for a week and run it across a bandsaw. "Meeeeeooooowwwwwwwwwww"
>But that begs the question exactly what was the USA thinking when they deployed those missiles to Turkey so close to the Soviet border? Did they really think the Soviets wouldn't react?
I don't know if this is true, but in "Dark Sun", a book documenting the development of the hydrogen bomb, Richard Rhodes claims that from roughly 1948 to when Powers got shot down in 1960, the United States had surveillance aircraft flying over the USSR 24 hours a day, seven days a week, and in the late '50's the US ran practice bombing runs of several dozen bombers over major western USSR cities during daylight. I also recall reading, possibly in Ben Rich's "Skunk Works", that even after the Powers shoot-down, the USSR's main/only defense against spy overflights was to just put up as many aircraft as possible, flying beneath the US aircraft, to try and physically obscure the ground.
Likewise, again from the Rhodes book, Curtis LeMay was openly calling for a pre-emptive, wide-spread nuclear strike on the USSR because he felt that it was the only reasonable defense. Lemay was the Strategic Air Command general who had the authority to launch nuclear weapons.
So, yeah, I'd say the US perception of USSR threat was a little different than the USSR perception of US threat. I'm completely amazed we didn't end up having a nuclear war, and I think the only reason we didn't was because the USSR was astoundingly reluctant to react to the repeated US aggressions. If they'd done 1/10 the things we did, I seriously doubt we would've shown the restraint they did.
That's not unusual: the cheapest test subjects are always the people doing the research.
When Mary Roach was writing "Bonk" -- a book devoted to medical research on sexuality -- she wanted to see the process, not just the results, of a doctor running an MRI and imaging people having sex. The people who did such research said there was absolutely no way she could watch unless she supplied the test subjects. (She and her husband went. Uproarious, really interesting book to read.)
Likewise, in undergrad microbiology courses, it would be unbelievably expensive and difficult to hand out human blood samples to students to have them, say, look at live blood cells under a microscope, so students routinely draw samples from themselves, because that way there's almost no risk of contaminated blood.
My ex-gf built ultrasound transducers for medical imaging. When the company was getting started, everyone who worked there was a man (engineering, yaaaay.) So their first breast imaging setup, they hired a hooker, because they couldn't find a cheap test subject. The prostate imaging? they flipped a coin, or drew straws, as to who got to be on which end of the test equipment.
Well... the height of the wave is proportional to the depth of the water below. If you're in a boat out over deep ocean you won't notice it, but if you're in the harbor where the devastation is happening you sure will. If the eyewitnesses were in somewhat shallow water they'd probably see a lot of excitement.
Personally, I like SL because people can build beautiful artwork -- reproductions of 1850's-era Japanese teahouses, for instance -- that I can wander through without having to, y'know, fly to Japan.
It's also nice that since I've assiduously ignored all the morons and loudmouths and hung out patiently, I've met and become friends with a mechanical engineer who teaches computational flow dynamics and who answers lots of questions; a DJ who has great taste in music and has introduced me to a dozen bands I would never have heard of; the bass player for a punk rock band that I've now gotten to see play live; and someone who is building his own plane (which I plan to do) and can show me, with SL models, what he's having problems with construction-wise with his RL plane. Again, without having to fly to California or Iowa or Georgia or Florida.
It is vaguely possible I could have met all those people in various places online, if I had the patience and tenacity to go looking. On SL I didn't have to: I just kept hanging out and they met me.
And, by the way, I've never paid a dime to use SL. Why would I? That's stupid. It's free if you don't want to 'own' a piece of 'land', whatever THAT means.
Hey, a fellow MOOer! I've been on Lambda since about 1993.
I think SL is somewhat like Lambda. Since there's a lower cost of entry, skillwise, there are a lot more people and a lot stupider people. But the programming is definitely reminiscent of moocode. It just sucks that LSL is so limited. If they were using moocode it'd be an enormous help: it was so nice to write display functions and statistics functions and then write code that called those to do data manipulation and presentation, and then use THAT to build database-like stuff. LSL is so... limited. But some people have built some fabulous stuff with their tiny little toolset.
The few wall warts I've poked around with were just a little transformer and a rectifier/cap: no attempt at active regulation. I've started building my own on fussy things, using a wart rated 3v or so higher and a linear regulator soldered to a small bit of copper sheet for a heatsink, with a 1000uF cap across the linear's output. Overkill, but for $7 worth of parts I don't ever have to worry about the quality of the output again.