Evidently Gates has forgotten everything before 1981. To summarize the state of computing on college campuses in 1973-75 (approximately the period when Gates & friends were using free time on Harvard mainframes to develop Altair BASIC, which was Microsoft's first product):
Proprietary operating systems -- in most cases the source was available, but since it only worked on one machine...
Commercial applications usually distributed as source and compiled for the target environment.
Lots of college kids busy hacking and swapping code. This was always source code -- since the hardware wasn't standardized, you had to re-compile for it, often with some tweaks. And the _fun_ was in the tweaks. (I kind of overdid it, spending so much time playing with the Star Trek program that I stopped going to classes. Eventually the college kicked me out...)
Who thought about copyright when we were having so much fun? Gates...
Of course, to play in this arena you had to somehow get access to a computer that cost more than a yard full of new Cadillacs. Either you had to be a student at a college that did not limit access to the computers, or you had to have a very tolerant employer. Microcomputers opened this up to everyone who was sufficiently interested. Within a few months of the release of Altair BASIC, hobbyist magazines were publishing hundreds of programs for it. Amd soon there were other microcomputers on the market, and everyone was adapting BASIC programs to them, and these were also swapped freely. Actually, from 1975-1981, Microsoft _was_ a major driver behind open source -- but nobody had invented a name for it yet, and this is certainly NOT what Gates was trying to claim credit for.
Just one thing Gates is correct about in the PC era -- if it hadn't been for his creation of gigantic monolithic software packages, all bound up in copyrights and security thru obscurity, it's quite possible that all those merry hackers would have simply continued doing their own thing without ever seeing themselves as a "movement" named Open Source, or a need to write the GPL so Gates and his imitators could not absorb code that had been given away freely into proprietary, closed, and undocumented programs...
Easy. First, if you lose the monitor, it's _way_ too small. So you put a buzzer in all the other pieces, and a button on the monitor to activate those buzzers.
It's a little harder with your remote, because it doesn't have any kind of receiver on it. But with Bluetooth devices, they'll receive the signal as long as you don't close them into the metal filing cabinet. 8-)
Correct. DVI Interface is a cable/connector definition for a digital interface to the monitor. I'm not sure whether Bluetooth has enough bandwidth for a monitor, but it wouldn't be a good idea for most offices anyhow. Put an RF-connected monitor into every cubicle and that's way too much RF in the air...
The mouse and keyboard are bluetooth, and that is a good idea, if it doesn't cost too much. Oh, and put a beeper in the mouse so I can hit a key and find it under the piles of paper. 8-)
Steve, my one issue with that is until the circuits are considerably more real-world, there is no basis for cost comparisons between binary and trinary circuits. An yet last time trinary was on/., there were those who didn't have any circuits to point at, yet claimed that the circuits wouldn't be significantly more complex. Since these people obviously lacked the experience to tell a practical circuit from a starting point, I figured I'd better cut them off in advance...
Just remember, in your spice modeling, try varying the transistor characteristics. It's not a real circuit until it can shrug off immense variations in the silicon. 10 degrees C temperature change can change beta by a factor of 2, for instance. Variation from wafer to wafer is even greater...
No, their inverter doesn't have an off state. It's either low (Q1 output transistor on to output almost -3V), or high (Q1 off, R2 pulls up to +3V). You need two output transistors for tristate (one in place of R2), one to turn on for high, one to turn on for low, both off for the no output current (tristated).
The cost of representing a particular number in a given base depends on a) how many digits there are in the number in that base, and b) how many digits there are in the base This assumes that the cost of increasing the base is the same as the cost of increasing the number of digits. There is no particular reason to expect this to be true. So far as I can tell, it definitely is not true when comparing the transistor count of binary, trinary, and higher bases. The step from binary to trinary is a big one -- either your gates are like analog circuits, or they are essentially a double binary gate (100% more circuitry for 50% more states/bit). The step from 3 to 4 is not nearly as big...
I dissected their inverter circuit in a different post -- in short, it won't work for the intermediate level, and in fact closely resembles a primitive ancestor of TTL binary.
This is not to say that higher bases are always and everywhere a bad idea in electronics, just that you need to be cautious when taking designs from someone who hopes someone else will build them... Transistors are becoming much cheaper than wires, and higher bases really save on wires. So does time-division multiplexing (e.g., sending the bits twice as fast on half as many wires), and at this point we better know how to do this, and can make it work more reliably at lower cost as compared to trinary. Eventually, multiplexing will hit some sort of practical speed limit, and then sending multi-level signals may be cost-effective. I just don't see any particular reason to stop at 3.
The last time ternary logic came up, I was disappointed to see no proposed schematics. Now there are schematics, but I'm still disappointed. One thing is that they designed with bipolar transistors rather than CMOS -- you cannot put more than a few thousand bipolar transistors on one chip without serious heatsinking... Beyond that, these designs lack quite a lot in speed, power consumption, and reliability as compared to even the 7400-series of TTL bipolar logic chips of the late 60's. And the first one I looked at doesn't even work.
Their ternary inverter is simply a two-transistor inverting _analog_ amplifier running on +/-3V supplies. If the input is -, Q2 turns on, bringing the base of Q1 low, turning Q1 off, so R2 pulls the output (which isn't explicitly shown) to the + rail. If the input is +, Q2 is off, and apparently this circuit depends on leakage to then bias Q1 on. This brings the output almost to the - rail. So it would work as a binary inverter. It's not nearly as good as a
TI 7404 (see page 2). The major difference is that R2 was replaced by a transistor, which turns on for high. This speeds up the low-to-high transition, since you get the full output current of the transistor until the output node is charged up. It also saves power, because one
output transistor is always off and the other always on, so when not switching only leakage currents flow at the output. (This two transistor output is called a "totem pole", and CMOS similarly depends on transistor pairs, one always off so little current flows.) Two more intermediate transistors are added, to control the top transistor on the totem pole and to reduce the resistor count. (On-chip, resistors are not cheaper than transistors.) But if you used it as a binary circuit, trinary.cc's inverter is basically the stripped-down ancestor of the 7404 circuit.
As a trinary circuit, it also has to take a 0V input and output 0V. This inverter does not do this reliably. It probably could be made to work by adjusting the resistor values until 0.0V in gave 0.0V out, but warm or cool the transistors a few degrees, and the amplifier bias will shift so that the output swings to the + or - rail. When you are trying to put the mid-level through it, you are running it like an analog amplifier, and analog amplifiers are unstable without negative feedback.
Nor would adding a few transistors and a negative feedback loop to stabilize it make it work well enough. A trinary inverter should take an input that is not right at any logic level, decide which level is closest, and output the corresponding nomimal voltage. For highs and lows (2 and 0), it does that, since it pins the output to the opposite rail. But even if you can be sure that 0.0V in = 0.0V out, with a circuit that is basically an analog amp, -0.1V in will give more than +0.1V out. So a chain of gates would allow the logic levels to get worse at each gate, until the mid-level became misinterpreted as + or -. To restore the mid-level would take a much more complicated circuit. I lay no claims to being a good designer at the transistor level, but I can't see any possibilities that are not nearly twice as complex as the corresponding binary circuits.
I think that he is offering to promise this bounty to a specific person or team, before they show working code. This shuts others out -- so he prefers someone who has demonstrated skills in the required area.
if there is some motivation to offer a bounty like this? Either he has money to give away (and this is certainly a worthy cause), or he's looking for a cheap way to get the code to make his employer's winmodem Linux-compatible and expand the market for it...
Note that the generic code he asks for does not make a complete modem program -- there is also going to have to be some hardware specific code. If you can't get the manufacturer's specs, that will be quite difficult, but to the guys that built the hardware, it's the easy part.
Doesn't it only have RNA, not DNA. Most viruses have DNA. Some, like HIV, have only RNA. RNA viruses have to carry a "reverse transcriptase" enzyme that translates the RNA code to DNA, so then the normal cell machinery can start reading the DNA and cranking out copies of virus RNA and proteins.
There's an analogy between *NA and programming:
DNA is source code
RNA is the intermediate linkable code
Proteins are the executable object code.
On the other hand, I tried using IE5 to view that page on Win98 & it's now been locked up for about 10 minutes from it. I didn't wait that long, but same here. It's not crashing IE5, but it's not sending through the pictures either, and somehow it has suppressed both the progress bar-graph and the IE5 timeout to a "page not available" sort of message.
So use the links above. They work fine (so far)...
AFAIK, no good way of producing the carbon nanotubes (buckytubes) for these cells has yet been discovered. They zap a lot of carbon to convert just a few percent to nanotubes, then try to sort the few right-sized tubes out of the mess. So if it can be produced at all, it's going to be very expensive.
my contract as a programmer at work means that my employer owns the stuff I program. This is entirely reasonable.
I'm not familiar with the UK teeny pop bands you cite, but this is not a common arrangement in the USA; bands that admittedly perform other people's work generally are paid only for performing live, and mostly in small venues. This is about the actual creators of the songs, who also performed on the vast majority of the CD's in stores:
You get a fixed salary. If your job requires travel, your employer pays the expenses. Your employer supplies the computer you work on. In most cases, "recording artists" get a small cut of the gross, minus expenses, including fees for using the company's studio. Quite often, this amounts to zero.
In brief, you are an employee. They are independent contractors. They could still come under work for hire if the company paid their expenses and "supervised" their work. But the companiew were too greedy about expenses, and even if a team of marketing droids actually wrote the songs, they certainly aren't going to put that on the public record...
It's a lot simpler than that. If it's work for hire, the copyright belongs to the company for as long as it runs. If it's not, the copyright goes back to the artist after a certain time. So all those moldie "golden oldies" from the 60's playing on the radio now belong to the artists (or their heirs, since many of these artists are now decomposing), not to the record companies. Obviously they think some of the current generation of noise-makers are going to be golden moldies in 2030...
The legal brief covers these points. Basically, because the power relationship between individuals and big companies is well known to be imbalanced (which is a nice way of saying that, although you are free to go down the street to another company, by some amazing coincidence you are likely to find them offering the exact same contract), and because the clause in the Constitution authorizing copyright laws says the copyright belongs to the author, copyright laws place limits on how much of his rights the author can sign away. Unless it is "work for hire", the author can assign all rights to the work _for a limited time_, but the author remains the owner of the copyright and eventually the rights will come back to him. (Since copyright law was first developed around printed works, "author" is generically used to cover any sort of creative person...)
Work for hire implies that the company is the true creator, not the author, and so the company owns the copyright. This can come about in two ways. One is if the author is a regular employee. For instance, a newspaper reporter gets a salary and is told what stories to cover, and so unless his employment contract specifically gives him rights in the stories, they belong to the newspaper. This doesn't apply to record contracts -- the lead artists are certainly not paid a salary or by the hour. The other is if the company contracts with a free-lancer to do a particular work. E.g., the newspaper asks Hunter Thompson to write "Fear and Loathing at the World Series", so many $ for so many words, plus expenses. That is, the company specifies the work to be done, the payment for the work, and pays expenses. So if a hollywood producer hires Barry Manilow to write a theme song for a movie, specifying elements that have to be included, that is probably going to be work for hire.
The typical recording session is nothing like that. The artist writes the songs they want (maybe Britney's songs are really written by a team of marketing experts, but the record company is hardly going to publicly admit that!), payments to the artists depend on the success of the CD, and expenses are deducted from their pay. No "supervision" by the company, no fixed pay, and no coverage of expenses; IANAL, but no way is that work for hire.
They admit "Of course, the most important component of a CD is the artist's effort in developing that music." But they carefully avoid citing actual numbers, so as not to mention that the artist only gets about 10% of the retail price. The artist is really important, and they pay him almost as much as the janitor...
My best guess at the numbers they didn't list:
Stores & distributors: 50%
CD production: 5%
Royalties: 10%
Profits, advertising, fancy offices & big salaries for the record company execs, bribing disk jockeys, bailing their stars out of jail, and covering their losses on bad music: 35%
Finally, "Each year, of the approximately 27,000 new releases that hit the market, the major labels release about 7,000 new CD titles and after production, recording, promotion and distribution costs, most never sell enough to recover these costs, let alone make a profit. In the end, less than 10% are profitable, and in effect, it's these recordings that finance all the rest."
Let's see, the major labels bring out 7,000 new CD's and less than 700 are profitable. On the other hand, small labels generally don't get the big stars (or can't afford to create synthetic stars through advertising), but somehow can afford to put out 20,000 CD's. Do the major-label guys have a serious lack of taste, or what?
You are correct that snail mail also makes it quite easy to send annoying anonymous mail, but there is one big difference between e-spam and junk mail: the junk mailer pays the full cost of delivery. Spammers pay less than half of the (much smaller) cost of mass emailings; they rely on intermediate servers to pass their stuff on for free, and finally it clogs up the bandwidth that recipients paid for. It's not bad at the office, where I get a share of a T1 line, but at home where 56K is the only affordable connection available (neither Verizon nor the cable company being ready for the 21st century), any spam that gets past the filters is a major annoyance.
On the other hand, why would I complain if someone pays to have free paper for lighting the wood-stove delivered to my home? 8-)
If you want to count each individual photon, photomultiplier tubes are the only choice.
In a PMT, a photon hitting the first plate releases an electron. The first plate (cathode) is negatively charged, so the electron flies off towards the less-negative 2nd plate, picking up enough energy to knock several electrons loose. These hit the third plate, knocking out more electrons, and so on. After many plates, the pulse of electrons is large enough to be easily measured, so they are collected and output on a wire at the back of the tube (anode). You can either measure the average current to determine photons/seconds, or detect each pulse to determine when each photon arrived. The super-K uses the latter method, since it has to compare photon arrival times to find the position of the event which created a burst of photons.
The PMT has very high gain and a remarkably good signal to noise ratio. "Gain" is the number of electrons out for one freed electron in, and you just add plates (and increase the overall voltage) until you get what you need. "Noise" would be an electron spontaneously flying off from the cathode, and this is pretty rare.
Solid-state detectors also start with a photon energizing one electron to jump somewhere it wouldn't normally go. Then you need an amplifier. It's possible to build solid-state circuits that will amplify a single electron to a measurable pulse, but to make it that sensitive you must also make it possible for electrons to just tunnel through the first amplifier stage on their own, and this is indistinguishable from detected photons. So it's hard to sort out the signal from the noise.
It would be simple to pressurize the enclosures A photomultiplier is a vacuum tube, 20 inches long in this case. It can't be pressurized, and one end has to be transparent. So whether it's the tube itself or a glass enclosure around it, you've got to have a big glass vessel that can support the pressure. Glass is actually pretty strong in compression, so if the shape is rounded for arch-like load transfer (a sphere or a cylinder with rounded ends), the tubes would hold up quite well to static pressure. Shock is a whole different matter.
But when you have 21,000 pieces of glass, it shouldn't come as a surprise when one gets broken. Why didn't they have baffles or shock-resistant enclosures around the tubes to prevent chain-reactions?
Wrong. In software, when the engineering is done, the job is done. After you design hardware, you've also got to have it built, and if you're earning enough to be able to pay that out of your own pocket, you won't have the spare time to design your own laptop... I design, build, and program functional test fixtures, so I'm pretty experienced in hand-built single-copy electronics. Simply getting parts can be difficult for a solo worker; about 10% of all the parts made are available in small quantities at distributors like Digikey, for up to 10 times the large-quantity price, but for the rest, the only chance of getting just one is to persuade them to send you a free sample. Since I work in an electronics factory, I have two big advantages when it comes to parts procurement: thousands of kinds of parts are already in the plant on thousand-piece reels, and if I need something we don't have, I can generally get a free sample by, umm, not making it clear that this is a one-off project and big orders aren't going to follow if the part works out...
So IF you don't care about size, weight, and speed, I can do a one-off build for 5 or 10 times as much as the unit cost in mass production, using off-the-shelf embedded controller PCB's (overpriced, but much cheaper than custom), wirewrapping whatever custom circuitry is needed, and buying a case that's more than big enough to hold it all. Sometimes that's $500 just for the case, because cases that big aren't exactly commodity items. (If you are good enough at metal-working, you can make your own case from about $100 in metal stock -- but people who have managed to learn this as well as electronics are pretty rare.) I can also layout a custom PCB, but I'm not good enough at layout to design a board that will work at 20MHz, or to compress even a 486-equivalent motherboard into laptop size. And to get a big, moderate-density, low-speed PCB made with good quality costs several thousand for the setup at the manufacturer. (There are etch-your-own-board kits, but with the quality I've seen coming out of these, I'd rather have good wirewrap.)
we're no longer taught hardware as well as we're taught software. I'm not sure "no longer" is true, but it's certainly true that engineering schools are not teaching the real-world aspects of getting from a schematic to working hardware. On the other hand, I don't think it's practical to learn it all anymore. For instance, if you really want to learn board layout, (1) get a job on a production line for a while, (2) take the IPC classes (these are aimed more at draftsmen than engineers, but it's all that's available), and (3) go to work doing board layouts at some company that has the good sense to have someone with 10 years experience watching over you. By the time you finish this, you're going to be 5 years out of date in other areas. Most board designers don't go through anything like this, and it sure shows when we have to try to produce their lousy layouts... Likewise, most engineers drawing schematics or laying out boards are quite ignorant of test requirements, until they've had a dozen boards thrown back at them as impossible to test at production speeds...
I don't think it was "if you build it, they will come" so much as salesmanship ran wild -- the assumption that you could sell a non-product on the web for no money and still make money somehow... "We're giving it away for free, but we dominate the market."
Or the notion that you could use the web to sell low-profit-margin items like groceries. Supermarkets are expert at keeping all costs down, or they don't survive, because people will go to a different store just because milk is 1 cent lower. Dot-coms weren't into extreme cost-cutting, never mind that UPS delivery alone costs more than the price at a supermarket.
"Culp compares the practice of publishing vulnerabilities to shouting "Fire" in a crowded movie theater. What he forgets is that there actually is a fire, the vulnerabilities exist regardless. Blaming the person who disclosed the vulnerability is like imprisoning the person who first saw the flames."
According to the article, they were refilling the tank after draining it when the glass bulbs started imploding in a chain reaction. The materials for neutrino detectors have to be _extremely_ pure, so I don't see any chance of having unintended impurities in sufficient concentration to react with glass.
On the other hand, extremely pure water is itself a strong solvent for many things. That may be why they had to periodically change the water in the first place, that it gradually picked up impurities from the tank. But dissolving glass doesn't seem likely unless they really had the wrong kind of glass to start with.
The one thing that's clear is the "chain reaction." The photomultipliers are glass vacuum turbes 20 inches long. In air, one of these breaking would be like a cherry bomb or maybe even a hand-grenade. Deep under water it would be much more violent due to the higher pressure, also water transmits shock waves better than air. So unless the bulbs were well protected, any accident or defective bulb that broke one would start a shock wave that might break more, etc.
I have two theories about why the researchers are reportedly closed mouthed about two simple questions: What broke the first one, and why weren't they protected against a foreseeable chain reaction implosion. Perhaps "Yoshi dropped a wrench, and we didn't think of that" is just too embarrassing. Or, if Japanese scientists are anything like the American ones I've known, more likely they said just what happened, in jargon that is utterly incomprehensible to the reporter, repeated until he gave up. And they weren't even trying to obfuscate!
Slashdot Mirror
It's proper to ask for ID's. But that isn't the complaint here -- it's that they've made it well-nigh impossible to reach them at all.
Evidently Gates has forgotten everything before 1981. To summarize the state of computing on college campuses in 1973-75 (approximately the period when Gates & friends were using free time on Harvard mainframes to develop Altair BASIC, which was Microsoft's first product):
Proprietary operating systems -- in most cases the source was available, but since it only worked on one machine...
Commercial applications usually distributed as source and compiled for the target environment.
Lots of college kids busy hacking and swapping code. This was always source code -- since the hardware wasn't standardized, you had to re-compile for it, often with some tweaks. And the _fun_ was in the tweaks. (I kind of overdid it, spending so much time playing with the Star Trek program that I stopped going to classes. Eventually the college kicked me out...)
Who thought about copyright when we were having so much fun? Gates...
Of course, to play in this arena you had to somehow get access to a computer that cost more than a yard full of new Cadillacs. Either you had to be a student at a college that did not limit access to the computers, or you had to have a very tolerant employer. Microcomputers opened this up to everyone who was sufficiently interested. Within a few months of the release of Altair BASIC, hobbyist magazines were publishing hundreds of programs for it. Amd soon there were other microcomputers on the market, and everyone was adapting BASIC programs to them, and these were also swapped freely. Actually, from 1975-1981, Microsoft _was_ a major driver behind open source -- but nobody had invented a name for it yet, and this is certainly NOT what Gates was trying to claim credit for.
Just one thing Gates is correct about in the PC era -- if it hadn't been for his creation of gigantic monolithic software packages, all bound up in copyrights and security thru obscurity, it's quite possible that all those merry hackers would have simply continued doing their own thing without ever seeing themselves as a "movement" named Open Source, or a need to write the GPL so Gates and his imitators could not absorb code that had been given away freely into proprietary, closed, and undocumented programs...
Easy. First, if you lose the monitor, it's _way_ too small. So you put a buzzer in all the other pieces, and a button on the monitor to activate those buzzers.
It's a little harder with your remote, because it doesn't have any kind of receiver on it. But with Bluetooth devices, they'll receive the signal as long as you don't close them into the metal filing cabinet. 8-)
Offtopic!!! Hey, this was the one that got me started laughing out loud.
Correct. DVI Interface is a cable/connector definition for a digital interface to the monitor. I'm not sure whether Bluetooth has enough bandwidth for a monitor, but it wouldn't be a good idea for most offices anyhow. Put an RF-connected monitor into every cubicle and that's way too much RF in the air...
The mouse and keyboard are bluetooth, and that is a good idea, if it doesn't cost too much. Oh, and put a beeper in the mouse so I can hit a key and find it under the piles of paper. 8-)
Steve, my one issue with that is until the circuits are considerably more real-world, there is no basis for cost comparisons between binary and trinary circuits. An yet last time trinary was on /., there were those who didn't have any circuits to point at, yet claimed that the circuits wouldn't be significantly more complex. Since these people obviously lacked the experience to tell a practical circuit from a starting point, I figured I'd better cut them off in advance...
Just remember, in your spice modeling, try varying the transistor characteristics. It's not a real circuit until it can shrug off immense variations in the silicon. 10 degrees C temperature change can change beta by a factor of 2, for instance. Variation from wafer to wafer is even greater...
No, their inverter doesn't have an off state. It's either low (Q1 output transistor on to output almost -3V), or high (Q1 off, R2 pulls up to +3V). You need two output transistors for tristate (one in place of R2), one to turn on for high, one to turn on for low, both off for the no output current (tristated).
The cost of representing a particular number in a given base depends on a) how many digits there are in the number in that base, and b) how many digits there are in the base This assumes that the cost of increasing the base is the same as the cost of increasing the number of digits. There is no particular reason to expect this to be true. So far as I can tell, it definitely is not true when comparing the transistor count of binary, trinary, and higher bases. The step from binary to trinary is a big one -- either your gates are like analog circuits, or they are essentially a double binary gate (100% more circuitry for 50% more states/bit). The step from 3 to 4 is not nearly as big...
I dissected their inverter circuit in a different post -- in short, it won't work for the intermediate level, and in fact closely resembles a primitive ancestor of TTL binary.
This is not to say that higher bases are always and everywhere a bad idea in electronics, just that you need to be cautious when taking designs from someone who hopes someone else will build them... Transistors are becoming much cheaper than wires, and higher bases really save on wires. So does time-division multiplexing (e.g., sending the bits twice as fast on half as many wires), and at this point we better know how to do this, and can make it work more reliably at lower cost as compared to trinary. Eventually, multiplexing will hit some sort of practical speed limit, and then sending multi-level signals may be cost-effective. I just don't see any particular reason to stop at 3.
The last time ternary logic came up, I was disappointed to see no proposed schematics. Now there are schematics, but I'm still disappointed. One thing is that they designed with bipolar transistors rather than CMOS -- you cannot put more than a few thousand bipolar transistors on one chip without serious heatsinking... Beyond that, these designs lack quite a lot in speed, power consumption, and reliability as compared to even the 7400-series of TTL bipolar logic chips of the late 60's. And the first one I looked at doesn't even work.
Their ternary inverter is simply a two-transistor inverting _analog_ amplifier running on +/-3V supplies. If the input is -, Q2 turns on, bringing the base of Q1 low, turning Q1 off, so R2 pulls the output (which isn't explicitly shown) to the + rail. If the input is +, Q2 is off, and apparently this circuit depends on leakage to then bias Q1 on. This brings the output almost to the - rail. So it would work as a binary inverter. It's not nearly as good as a
TI 7404 (see page 2). The major difference is that R2 was replaced by a transistor, which turns on for high. This speeds up the low-to-high transition, since you get the full output current of the transistor until the output node is charged up. It also saves power, because one
output transistor is always off and the other always on, so when not switching only leakage currents flow at the output. (This two transistor output is called a "totem pole", and CMOS similarly depends on transistor pairs, one always off so little current flows.) Two more intermediate transistors are added, to control the top transistor on the totem pole and to reduce the resistor count. (On-chip, resistors are not cheaper than transistors.) But if you used it as a binary circuit, trinary.cc's inverter is basically the stripped-down ancestor of the 7404 circuit.
As a trinary circuit, it also has to take a 0V input and output 0V. This inverter does not do this reliably. It probably could be made to work by adjusting the resistor values until 0.0V in gave 0.0V out, but warm or cool the transistors a few degrees, and the amplifier bias will shift so that the output swings to the + or - rail. When you are trying to put the mid-level through it, you are running it like an analog amplifier, and analog amplifiers are unstable without negative feedback.
Nor would adding a few transistors and a negative feedback loop to stabilize it make it work well enough. A trinary inverter should take an input that is not right at any logic level, decide which level is closest, and output the corresponding nomimal voltage. For highs and lows (2 and 0), it does that, since it pins the output to the opposite rail. But even if you can be sure that 0.0V in = 0.0V out, with a circuit that is basically an analog amp, -0.1V in will give more than +0.1V out. So a chain of gates would allow the logic levels to get worse at each gate, until the mid-level became misinterpreted as + or -. To restore the mid-level would take a much more complicated circuit. I lay no claims to being a good designer at the transistor level, but I can't see any possibilities that are not nearly twice as complex as the corresponding binary circuits.
I think that he is offering to promise this bounty to a specific person or team, before they show working code. This shuts others out -- so he prefers someone who has demonstrated skills in the required area.
if there is some motivation to offer a bounty like this? Either he has money to give away (and this is certainly a worthy cause), or he's looking for a cheap way to get the code to make his employer's winmodem Linux-compatible and expand the market for it...
Note that the generic code he asks for does not make a complete modem program -- there is also going to have to be some hardware specific code. If you can't get the manufacturer's specs, that will be quite difficult, but to the guys that built the hardware, it's the easy part.
Doesn't it only have RNA, not DNA. Most viruses have DNA. Some, like HIV, have only RNA. RNA viruses have to carry a "reverse transcriptase" enzyme that translates the RNA code to DNA, so then the normal cell machinery can start reading the DNA and cranking out copies of virus RNA and proteins.
There's an analogy between *NA and programming:
DNA is source code
RNA is the intermediate linkable code
Proteins are the executable object code.
On the other hand, I tried using IE5 to view that page on Win98 & it's now been locked up for about 10 minutes from it. I didn't wait that long, but same here. It's not crashing IE5, but it's not sending through the pictures either, and somehow it has suppressed both the progress bar-graph and the IE5 timeout to a "page not available" sort of message.
So use the links above. They work fine (so far)...
AFAIK, no good way of producing the carbon nanotubes (buckytubes) for these cells has yet been discovered. They zap a lot of carbon to convert just a few percent to nanotubes, then try to sort the few right-sized tubes out of the mess. So if it can be produced at all, it's going to be very expensive.
my contract as a programmer at work means that my employer owns the stuff I program. This is entirely reasonable.
I'm not familiar with the UK teeny pop bands you cite, but this is not a common arrangement in the USA; bands that admittedly perform other people's work generally are paid only for performing live, and mostly in small venues. This is about the actual creators of the songs, who also performed on the vast majority of the CD's in stores:
You get a fixed salary. If your job requires travel, your employer pays the expenses. Your employer supplies the computer you work on. In most cases, "recording artists" get a small cut of the gross, minus expenses, including fees for using the company's studio. Quite often, this amounts to zero.
In brief, you are an employee. They are independent contractors. They could still come under work for hire if the company paid their expenses and "supervised" their work. But the companiew were too greedy about expenses, and even if a team of marketing droids actually wrote the songs, they certainly aren't going to put that on the public record...
It's a lot simpler than that. If it's work for hire, the copyright belongs to the company for as long as it runs. If it's not, the copyright goes back to the artist after a certain time. So all those moldie "golden oldies" from the 60's playing on the radio now belong to the artists (or their heirs, since many of these artists are now decomposing), not to the record companies. Obviously they think some of the current generation of noise-makers are going to be golden moldies in 2030...
The legal brief covers these points. Basically, because the power relationship between individuals and big companies is well known to be imbalanced (which is a nice way of saying that, although you are free to go down the street to another company, by some amazing coincidence you are likely to find them offering the exact same contract), and because the clause in the Constitution authorizing copyright laws says the copyright belongs to the author, copyright laws place limits on how much of his rights the author can sign away. Unless it is "work for hire", the author can assign all rights to the work _for a limited time_, but the author remains the owner of the copyright and eventually the rights will come back to him. (Since copyright law was first developed around printed works, "author" is generically used to cover any sort of creative person...)
Work for hire implies that the company is the true creator, not the author, and so the company owns the copyright. This can come about in two ways. One is if the author is a regular employee. For instance, a newspaper reporter gets a salary and is told what stories to cover, and so unless his employment contract specifically gives him rights in the stories, they belong to the newspaper. This doesn't apply to record contracts -- the lead artists are certainly not paid a salary or by the hour. The other is if the company contracts with a free-lancer to do a particular work. E.g., the newspaper asks Hunter Thompson to write "Fear and Loathing at the World Series", so many $ for so many words, plus expenses. That is, the company specifies the work to be done, the payment for the work, and pays expenses. So if a hollywood producer hires Barry Manilow to write a theme song for a movie, specifying elements that have to be included, that is probably going to be work for hire.
The typical recording session is nothing like that. The artist writes the songs they want (maybe Britney's songs are really written by a team of marketing experts, but the record company is hardly going to publicly admit that!), payments to the artists depend on the success of the CD, and expenses are deducted from their pay. No "supervision" by the company, no fixed pay, and no coverage of expenses; IANAL, but no way is that work for hire.
They admit "Of course, the most important component of a CD is the artist's effort in developing that music." But they carefully avoid citing actual numbers, so as not to mention that the artist only gets about 10% of the retail price. The artist is really important, and they pay him almost as much as the janitor...
My best guess at the numbers they didn't list:
Stores & distributors: 50%
CD production: 5%
Royalties: 10%
Profits, advertising, fancy offices & big salaries for the record company execs, bribing disk jockeys, bailing their stars out of jail, and covering their losses on bad music: 35%
Finally, "Each year, of the approximately 27,000 new releases that hit the market, the major labels release about 7,000 new CD titles and after production, recording, promotion and distribution costs, most never sell enough to recover these costs, let alone make a profit. In the end, less than 10% are profitable, and in effect, it's these recordings that finance all the rest."
Let's see, the major labels bring out 7,000 new CD's and less than 700 are profitable. On the other hand, small labels generally don't get the big stars (or can't afford to create synthetic stars through advertising), but somehow can afford to put out 20,000 CD's. Do the major-label guys have a serious lack of taste, or what?
You are correct that snail mail also makes it quite easy to send annoying anonymous mail, but there is one big difference between e-spam and junk mail: the junk mailer pays the full cost of delivery. Spammers pay less than half of the (much smaller) cost of mass emailings; they rely on intermediate servers to pass their stuff on for free, and finally it clogs up the bandwidth that recipients paid for. It's not bad at the office, where I get a share of a T1 line, but at home where 56K is the only affordable connection available (neither Verizon nor the cable company being ready for the 21st century), any spam that gets past the filters is a major annoyance.
On the other hand, why would I complain if someone pays to have free paper for lighting the wood-stove delivered to my home? 8-)
If you want to count each individual photon, photomultiplier tubes are the only choice.
In a PMT, a photon hitting the first plate releases an electron. The first plate (cathode) is negatively charged, so the electron flies off towards the less-negative 2nd plate, picking up enough energy to knock several electrons loose. These hit the third plate, knocking out more electrons, and so on. After many plates, the pulse of electrons is large enough to be easily measured, so they are collected and output on a wire at the back of the tube (anode). You can either measure the average current to determine photons/seconds, or detect each pulse to determine when each photon arrived. The super-K uses the latter method, since it has to compare photon arrival times to find the position of the event which created a burst of photons.
The PMT has very high gain and a remarkably good signal to noise ratio. "Gain" is the number of electrons out for one freed electron in, and you just add plates (and increase the overall voltage) until you get what you need. "Noise" would be an electron spontaneously flying off from the cathode, and this is pretty rare.
Solid-state detectors also start with a photon energizing one electron to jump somewhere it wouldn't normally go. Then you need an amplifier. It's possible to build solid-state circuits that will amplify a single electron to a measurable pulse, but to make it that sensitive you must also make it possible for electrons to just tunnel through the first amplifier stage on their own, and this is indistinguishable from detected photons. So it's hard to sort out the signal from the noise.
It would be simple to pressurize the enclosures A photomultiplier is a vacuum tube, 20 inches long in this case. It can't be pressurized, and one end has to be transparent. So whether it's the tube itself or a glass enclosure around it, you've got to have a big glass vessel that can support the pressure. Glass is actually pretty strong in compression, so if the shape is rounded for arch-like load transfer (a sphere or a cylinder with rounded ends), the tubes would hold up quite well to static pressure. Shock is a whole different matter.
But when you have 21,000 pieces of glass, it shouldn't come as a surprise when one gets broken. Why didn't they have baffles or shock-resistant enclosures around the tubes to prevent chain-reactions?
Wrong. In software, when the engineering is done, the job is done. After you design hardware, you've also got to have it built, and if you're earning enough to be able to pay that out of your own pocket, you won't have the spare time to design your own laptop... I design, build, and program functional test fixtures, so I'm pretty experienced in hand-built single-copy electronics. Simply getting parts can be difficult for a solo worker; about 10% of all the parts made are available in small quantities at distributors like Digikey, for up to 10 times the large-quantity price, but for the rest, the only chance of getting just one is to persuade them to send you a free sample. Since I work in an electronics factory, I have two big advantages when it comes to parts procurement: thousands of kinds of parts are already in the plant on thousand-piece reels, and if I need something we don't have, I can generally get a free sample by, umm, not making it clear that this is a one-off project and big orders aren't going to follow if the part works out...
So IF you don't care about size, weight, and speed, I can do a one-off build for 5 or 10 times as much as the unit cost in mass production, using off-the-shelf embedded controller PCB's (overpriced, but much cheaper than custom), wirewrapping whatever custom circuitry is needed, and buying a case that's more than big enough to hold it all. Sometimes that's $500 just for the case, because cases that big aren't exactly commodity items. (If you are good enough at metal-working, you can make your own case from about $100 in metal stock -- but people who have managed to learn this as well as electronics are pretty rare.) I can also layout a custom PCB, but I'm not good enough at layout to design a board that will work at 20MHz, or to compress even a 486-equivalent motherboard into laptop size. And to get a big, moderate-density, low-speed PCB made with good quality costs several thousand for the setup at the manufacturer. (There are etch-your-own-board kits, but with the quality I've seen coming out of these, I'd rather have good wirewrap.)
we're no longer taught hardware as well as we're taught software. I'm not sure "no longer" is true, but it's certainly true that engineering schools are not teaching the real-world aspects of getting from a schematic to working hardware. On the other hand, I don't think it's practical to learn it all anymore. For instance, if you really want to learn board layout, (1) get a job on a production line for a while, (2) take the IPC classes (these are aimed more at draftsmen than engineers, but it's all that's available), and (3) go to work doing board layouts at some company that has the good sense to have someone with 10 years experience watching over you. By the time you finish this, you're going to be 5 years out of date in other areas. Most board designers don't go through anything like this, and it sure shows when we have to try to produce their lousy layouts... Likewise, most engineers drawing schematics or laying out boards are quite ignorant of test requirements, until they've had a dozen boards thrown back at them as impossible to test at production speeds...
I don't think it was "if you build it, they will come" so much as salesmanship ran wild -- the assumption that you could sell a non-product on the web for no money and still make money somehow... "We're giving it away for free, but we dominate the market."
Or the notion that you could use the web to sell low-profit-margin items like groceries. Supermarkets are expert at keeping all costs down, or they don't survive, because people will go to a different store just because milk is 1 cent lower. Dot-coms weren't into extreme cost-cutting, never mind that UPS delivery alone costs more than the price at a supermarket.
"Culp compares the practice of publishing vulnerabilities to shouting "Fire" in a crowded movie theater. What he forgets is that there actually is a fire, the vulnerabilities exist regardless. Blaming the person who disclosed the vulnerability is like imprisoning the person who first saw the flames."
According to the article, they were refilling the tank after draining it when the glass bulbs started imploding in a chain reaction. The materials for neutrino detectors have to be _extremely_ pure, so I don't see any chance of having unintended impurities in sufficient concentration to react with glass.
On the other hand, extremely pure water is itself a strong solvent for many things. That may be why they had to periodically change the water in the first place, that it gradually picked up impurities from the tank. But dissolving glass doesn't seem likely unless they really had the wrong kind of glass to start with.
The one thing that's clear is the "chain reaction." The photomultipliers are glass vacuum turbes 20 inches long. In air, one of these breaking would be like a cherry bomb or maybe even a hand-grenade. Deep under water it would be much more violent due to the higher pressure, also water transmits shock waves better than air. So unless the bulbs were well protected, any accident or defective bulb that broke one would start a shock wave that might break more, etc.
I have two theories about why the researchers are reportedly closed mouthed about two simple questions: What broke the first one, and why weren't they protected against a foreseeable chain reaction implosion. Perhaps "Yoshi dropped a wrench, and we didn't think of that" is just too embarrassing. Or, if Japanese scientists are anything like the American ones I've known, more likely they said just what happened, in jargon that is utterly incomprehensible to the reporter, repeated until he gave up. And they weren't even trying to obfuscate!