[A sparkgap transmitter] is a narrowband transmitter that just happens to be rather inefficient.
Exactly. The article's assertion that UWB is similar to spark gaps is ludicrously wrong. A spark gap is simply a low-gain amplifier, connected to a dissipative resonator. It's exactly like a quartz crystal oscillator, just much lower quality.
If you want to take a drug that no one knows how it works, go for it. However, I will not be taking a drug that scientists are clueless about.
The primary mode of action of many drugs is unknown. (In fact, nobody has been able to explain how a single general anesthetic works!)
And that's the good part: side effects are almost completely unknown. They're found by administering the drug and seeing what happens. Sometimes there are no side effects, sometimes you get a dry mouth, sometimes your fingernails fall off, sometimes all your mitochondria die, etc. There's a reason drug trials are carefully planned and monitored.
There is a reason you get sleepy. Your body and mind needs to rest and recharge and sleepiness is the signal to do so.
Completely wrong. There are some people who *never* sleep, yet do not suffer from bad effects. If the need for sleep where a fundamental characteristic of higher neural activity, they would die in childhood, but they do not.
It is entirely possible (and I would say likely) that sleep is caused by a clock mechanism that tries to modulate consciousness. It is plausible that the "turn off" signal from the clock becomes hyperactive if it isn't satisfied and wreaks havoc on the rest of the brain, which is perfectly capable of staying online. If you could block or reset this hypothetical clock with a drug, you could stay awake forever.
Re:black holes etc.
on
Quark Stars
·
· Score: 3, Informative
What would happen if you start dumping an huge amount of electrons in a black hole?
The electric field near the event horizon would grow larger and larger. At some point, electron-positron pairs would start "precipitating" from the vacuum. The positrons would be attracted into the negative black hole and move its net charge toward zero. The electrons would be repelled away from the black hole. Incidentally, the mass of the black hole would be decreased in the process.
Why? One way of looking at the vacuum is that it is filled with virtual particles. A group of virtual particles can "borrow" energy to spring into existence, and then annihilate after a short period of time, returning the borrowed energy to the vacuum. The time scale they are allowed to exist is governed by Heisenberg's uncertainty relation. (E*t>=h-bar.) For massive particles like electrons, it's a short period of time.
If, during their short existence, the electric field can do more work on the particles than their borrowed energy, the "debt" to the vacuum can be "repaid", and the particles can become real.
It is advertised to protect the pristine cleanliness of your perfect digital signal.
That's just silly. As long as you operate your microwave oven with the door closed, there won't be much problem.
S/PDIF doesn't transfer any power for devices, just signal (which, on the wire, would be a low power analog signal putting out no more interference than an RCA cable carrying a line level).
I haven't read the specification, but it probably carries an approx. 2.4MHz digital signal. Which means it will have significant sinewave harmonics out to 10-20MHz. Which means that a single poorly soldered connection at an RCA connector can screw up radio and TV reception, even if it still seems to work.
That's for the good stuff. If the equipment is poorly designed, it could easily have digital noise and harmonics out to hundreds of MHz. If you've done much entertainment center stuff, you've probably come across a cheap CD player or similar that screws up TV reception--I've come across several. Hostile equipment like that is why I recommend using good coax cables for digital. (Good doesn't necessarily mean expensive, but I try to avoid the cheapest cables.) It's far easier to prevent noise problems than it is to diagnose and fix them after they occur.
So you are telling me that a 'Corporate Citizen' is not made up of it's investors and and employees,...
Yes. A corporation is a legal person distinct from its officers, employees, agents, and investors.
It is, in fact, possible for a corporation to exist without any employees, officers, and owners. Example: a terrible accident could kill every employee and officer of a small company at the same time, whereupon a court could take over operation of the company in order to discharge its debts and obligations.
They took unarmored planes into close combat against highly-defended targets, often crashing their planes into those targets, and only had enough fuel for a one-way trip. The mission was certain death. If that doesn't make them suicides, I don't know what the word means.
Firstly the cryptanalyst may well have more luck breaking the combined layered cipher than trying to break both individually The layered cipher may well be weaker!
If the subciphers are independently keyed, the overall cipher is at least as strong as the weakest subcipher.
Brute force... what does that mean? 56-bit breakable today. 128-bit breakable tomorrow. 256-bit breakable... when there are more than 2^256 electrons in the universe!
Key length means nothing if you can find analytical attacks. A primary worry is automated theorem provers, which can pull structure out of provably chaotic systems--witness the equation that calculates an arbitrary binary digit of pi without calculating any other digits. If strong AIs are developed, who knows what could be analyzed...
I think it's likely that vulnerabilities will be discovered in some ciphers as better analytical tools become available. Layering ciphers can mitigate this problem, and it doesn't cost much.
A "secure algorithm" is basically one such that it does not matter whether the hacker has access to the algorithm or not.
The point of layering is not to keep the algorithm secret, it's to protect against cryptanalytical breakthroughs. Even if the cryptanalyst gets very lucky (or is very smart) and completely breaks one of the subciphers, the other subciphers still protect your data.
EULAs are a legal fiction and have no force or validity whatsoever.
That's true only for "shrinkwrap" licenses. I.e., licenses that are only seen after purchase.
If you agree to the EULA contract before paying money and receiving the product, it is enforceable. Be very, very careful when buying software from vendor websites.
Using S/PDIF over the shittiest RCA cable you can find, with large amounts of interference, you still won't lose a single bit of data in the transfer.
True, but that's not the point. The fancy shielding is to keep the digital noise from leaking into analog audio and video signals, as well as nearby radio receivers.
With MySQL you can start with SELECT * FROM my_table without having to worry about the extra stuff PostgreSQL offers.
I disagree. If you are reading along with a tutorial, none of PostgreSQL's advanced features will bite you. You have to explicity use the features to run into them, and table inheritance is not the sort of thing you do accidentally, especially if you're using a generic "SQL For Dummies" type book.
Getting used to it's more unique features like table inheritance and virtually infinite extensibility, however, will bite you in the ass when moving on more than knowing how to survive without transactions will.
As long as you aren't trying to read straight through the reference manual, you won't come across the advanced non-portable features until you know enough to ignore them.
Seriously, this is just another business cycle. Lots of people jump in, overbuild for the current market, the market crashes, the survivors consolidate, demand marches onward, business picks up, lots of people jump in,... Happens to everything: wheat, memory chips, telecom, you name it.
Bullshit. Nobody held a gun to their heads and forced them to start working there. Nobody raped their daughter until they signed the contract with the non-compete and IP clauses. No, they were fully informed about the restrictions, and judged that the salary was worth it.
isn't a "non-competition" clause the antithesis of the free market?
They are selling promises about their own behavior. You can't get any freer than that. Judging from your tone and incoherence, you probably have a "solution" to this "problem" that involes a nanny government keeping people from doing things "for their own good".
I suspect that it's because companies are by nature trying to get as much profit as possible. This includes, I think, in their view, possible monies to be made off software licences etc.
Nope, it's all about lawyers trying to justify their paychecks by putting as many lawyerly terms and conditions in everything they can lay their hands on.
I don't think that capitalism without checks is a good idea. This only encourages workers to become depersonalized and disinterested in their work.
<boggle> Working for one of these idiot companies is not slavery to capitalism, it is a conscious, willing choice made freely by the employee. Capitalism means they are free to go anywhere else if they want to, or start their own company.
Imagine if this happened in the civilian world- CompanyX modifies GPLed GNU Emacs and puts it up for sale- but before a customer can purchase it, they have to sign a separate contract promising to never redistribute the source code. It's a blatant violation.
Wrong. Only restrictive contracts between the licensor and licensee are prohibited. Here's how to get around it:
Form a shell company that never touches the code. Since we're Star Wars fans, let's call the shell company Chewco (after Chewbacca).
Licensor publicly says "We give code to people who have secrecy contracts with Chewco." Since they never say "We don't give code to people who don't have secrecy contracts", it isn't a "further restriction" according to the GPL.
If anybody without a secrecy contract asks for code, they say "We choose not to give you code at this time for private reasons."
Anybody can get a secrecy contract with Chewco. They give money and promise secrecy, and Chewco gives them a nice flower arrangement (or some other valuable consideration) in return. This is a legal, enforceable contract. Since Chewco is not a licensor, they are not affected by the GPL. The GPL does not prohibit the licensee from being restricted, it merely keeps the licensor from imposing the restriction as a condition of licensing, so it's OK on that count too.
1) Olber's paradox: According to plasma theory, the universe is filamentary. We live in a relatively dense area of space, so there will be more light nearby than far away.
Olber's paradox implies that either 1) the universe is finite, 2) it was created in the (cosmologically) recent past, 3) or that physics has a complex mechanism for creation and destruction of matter and energy that has somehow never been observed.
In case #1, the universe cannot be finite unless we are at its very center with extremely high precision. This is very unlikely, verging on divine intervention.
For case #3, the necessary physics to rework the cosmos on a grand scale has never been observed. Thus we can rule it out.
This leaves case #2: creation in the recent past, followed by expansion.
2) Cosmic microwave background: Plasma sources can absorb and reemit microwaves. Since the direction is random, it will quickly reach uniformity.
This is completely wrong. For two regions of space to reach thermal equilibrium, they must have been exchanging photons for hundreds of times the lightspeed travel time between them. For the opposite sides of the visible universe--at least 10 billion light years--they would have had to be thermalizing with each other for at least hundreds of billions of years. Since no objects appear to be that old, we can rule out that possibility.
As for the others, the Big Bang does no better.
A hot/dense genesis + expansion makes many quantitative, testable predictions. It predicts that the universe should have been in good equilibrium in the past, and thus that its composition should be fairly uniform in all directions: this is observed. It predicts that photons that brought about thermal equilibrium should have been expanded into a low-energy radiation field with high uniformity and a perfect Planckian spectrum: this is observed. It predicts that if you look a particular distance in any direction, that the galaxies you see will have the same apparent age: they do. It predicts that nearby galaxies should look older, while distant galaxies should look younger: they do. It predicts that isotopes should have particular relative abundances from the epoch of thermonuclear equilibrium: they do. It predicts that stellar and fluorescent nebula light will not be highly polarized: it isn't. (In contrast, plasma cosmology predicts extremely large magnetic fields that would strongly polarize nearly all light sources.)
The magnetic field problem is one of the more damning flaws in plasma cosmology. The theory holds that stars and galaxies are moved as strongly by electromagnetism as by gravity. Such magnetic fields would have a variety of trivially observable consequences: those consequences are not observed even slightly. Moreover, galactic collisions would be *spectactular* events, akin to a solar flare on a galactic scale, and no observed galaxy collision has shown appreciable magnetic phenomena.
Plasma cosmology is simply unphysical on every scale, predicting things that are not observed, and failing to predict things that are observed.
Some motherboards have a bios option, "Spread Spectrum", which rounds off the square wave clock pulses, so that noise from sharp changes is lessened, IIRC.
No, it modulates the frequency of the system clocks.
It does decrease performance, and stop some cards functioning altogether.
Performance is unchanged, and anything that breaks is garbage and should be discarded anyway.
I haven't looked into Alfven's theories, but Lerner is completely full of shit. His book The Big Bang Never Happened routinely presents only partial information in a attempt to mislead the reader; it's unscientific garbage. Lerner's misleading reasoning and selective reporting are so pervasive that the man can only be regarded as a sleazy con artist or deluded crank.
Overall the plasma cosmology theories are scientific failures. They cannot explain Olber's paradox ("Why is the sky dark?"), which is a pretty damning failure. They do not predict the existence of the cosmic microwave background, nor explain it's spectrum and uniformity. They do not explain the Hubble redshift. They do not explain why distant galaxies look different than nearby galaxies. They do not predict or explain the ratios of the primordial elements.
A good electrician will be able to hook up a meter to a few sample servers and get the exact amount of juice they pull. Use the GREATER of that number and the name plate rating on the computer.
Ignore nameplate ratings on big devices like computers and monitors. They're usually overrated. Ignore measurements. Speaking as an electrical engineer who designs computer peripherals, getting true worst-case measurements is very, very difficult. You have to exercise the hard drive heads, CPU cores, RAM busses, and I/O busses fully, and that's near impossible. Switching power supplies also draw more current as the voltage goes down. If you make the measurement when the line voltage is 130V, the equipment will draw 20% more current at 110V.
For little things like KVMs, modems, inkjet printers, etc. you can safely use the nameplate ratings.
For big things, determine how many machines you would ever conceivably want in the room. Choose the biggest, baddest equipment you could possibly want. 1U dual-proccesor machines, arrays of 15000 rpm hard drives, a desk full of 21 inch monitors, you name it. Then go to the manufacturers web sites and find the nameplate ratings for the various things, and add 'em all up. The total will be a number you won't easily outgrow.
Be sure to account for start-up loads. You don't want to trip a breaker by turning everything on at once. Hard drives draw a lot of power while they're spinning up, monitors while degaussing, laser printers while warming up the fusion rollers. This is just an educated guess, but use a factor of 2 for hard drives, and 5 for monitors. Read the specs for the laser printers very, very carefully and find the worst-case.
Also most wiring in commercial spaces is done in conduit and more than 3 wires in a conduit requires that the wires be derated and not all electricians pay attention to that (again per the NEC).
I'd go even farther. When many surge protectors divert a surge, they divert it into the ground wire. This causes a brief, high voltage spike on that circuit's ground relative to the other circuits in the room. The longer the ground wire is, the larger the spike. This spike can do nasty things to serial lines, KVM cables, and so forth that connect machines on different circuits.
So if the building breaker box is farther than, say, 50 feet from the server room, I'd have a small breaker box installed in the server room. Also this lets you recover from a tripper breaker without getting the main breaker box unlocked.
If you can afford it, have a couple of separate circuits run from the main breaker box. This gives you someplace to plug in coffee pots and vacuum cleaners without disturbing the electronics.
If the room gets its own air conditioner, make sure that has a dedicated circuit from the main breaker box.
If you can afford it, have a big industrial surge protector installed at this breaker box. Also the breaker box is a good grounding point for surge protectors on your external data lines.
The net effect is that you should plan on the electrician using #10 THHN in any conduits.
This is excellent advice. The electrical code is based on safe operation of motors and heaters. Bigger wires make your electronics more reliable by reducing voltage droop.
Also, computers often don't draw sine wave current. They draw less current at the beginning and end of the AC cycle, and more in the middle. This means the peak current is larger than the sinewave loads envisioned by the electrical codes.
Computers often need good grounding systems, so I would also require a separate ground wire to be run in the conduits even though it is usually not done since the conduit can act as a ground.
More excellent advice. Conduit is completely unacceptable for grounding computers. A grounding wire is cheaper than the cost of a single computer crash caused by a poor ground.
You will also want to make sure the racks are grounded and you may even wish to consider putting a wire mesh beneath the floor and grounding that.
Have an electrician tie all the racks and other metal stuff together with big ground wires. This will help protect the rest of your equipment if one of the devices has a ground fault. It'll also help reduce static electricity by giving you lots of big grounded metal things to touch. Wire is cheap compared to the cost of a single failure.
Finally, if possible, require that the communications cables be run in over sized conduit as well. It makes expansion much easier in the future and also provides a measure of RF shielding.
Conduit does make running wires much easier. If there is no other wiring or fluorescent lights within a few feet, I'd use nonmetallic conduit, as metallic conduit can actually act as an antenna for picking up radio waves and coupling them into your data cables. OTOH if there are AC lines parallel to the run, metallic conduit is probably better, and be sure to make the electrician ground the conduit properly.
The SuperKamiokande... detector recently suffered an accident and is currently out of commission.
"Accident" as in "half its photomultiplier tubes imploded". It will be some time before SuperK is back online.
Will different densities affect how the neutrinos travel (making aiming a difficulty)?
They ought to fly straight. (Not that that makes aiming easy!) The parameter of most interest is just the distance between the source and the detector. It is postulated that neutrinos can switch flavors in flight. The more distance they fly, the more chance they have of switching.
Slashdot Mirror
And that's the good part: side effects are almost completely unknown. They're found by administering the drug and seeing what happens. Sometimes there are no side effects, sometimes you get a dry mouth, sometimes your fingernails fall off, sometimes all your mitochondria die, etc. There's a reason drug trials are carefully planned and monitored.
Completely wrong. There are some people who *never* sleep, yet do not suffer from bad effects. If the need for sleep where a fundamental characteristic of higher neural activity, they would die in childhood, but they do not.It is entirely possible (and I would say likely) that sleep is caused by a clock mechanism that tries to modulate consciousness. It is plausible that the "turn off" signal from the clock becomes hyperactive if it isn't satisfied and wreaks havoc on the rest of the brain, which is perfectly capable of staying online. If you could block or reset this hypothetical clock with a drug, you could stay awake forever.
Why? One way of looking at the vacuum is that it is filled with virtual particles. A group of virtual particles can "borrow" energy to spring into existence, and then annihilate after a short period of time, returning the borrowed energy to the vacuum. The time scale they are allowed to exist is governed by Heisenberg's uncertainty relation. (E*t>=h-bar.) For massive particles like electrons, it's a short period of time.
If, during their short existence, the electric field can do more work on the particles than their borrowed energy, the "debt" to the vacuum can be "repaid", and the particles can become real.
That's for the good stuff. If the equipment is poorly designed, it could easily have digital noise and harmonics out to hundreds of MHz. If you've done much entertainment center stuff, you've probably come across a cheap CD player or similar that screws up TV reception--I've come across several. Hostile equipment like that is why I recommend using good coax cables for digital. (Good doesn't necessarily mean expensive, but I try to avoid the cheapest cables.) It's far easier to prevent noise problems than it is to diagnose and fix them after they occur.
The blurb on the front page didn't have a single misspelling!
It is, in fact, possible for a corporation to exist without any employees, officers, and owners. Example: a terrible accident could kill every employee and officer of a small company at the same time, whereupon a court could take over operation of the company in order to discharge its debts and obligations.
They took unarmored planes into close combat against highly-defended targets, often crashing their planes into those targets, and only had enough fuel for a one-way trip. The mission was certain death. If that doesn't make them suicides, I don't know what the word means.
The main propulsion system of every vehicle is a weapon. Military and civil defense planners need to learn this lesson.
It's an *easy* decision: the Twin Towers should be rebuilt. In Riyadh.I think it's likely that vulnerabilities will be discovered in some ciphers as better analytical tools become available. Layering ciphers can mitigate this problem, and it doesn't cost much.
If you agree to the EULA contract before paying money and receiving the product, it is enforceable. Be very, very careful when buying software from vendor websites.
LOL! That's one of the funniest things I've ever heard.
Seriously, this is just another business cycle. Lots of people jump in, overbuild for the current market, the market crashes, the survivors consolidate, demand marches onward, business picks up, lots of people jump in, ... Happens to everything: wheat, memory chips, telecom, you name it.
OK, OK, I stole Chewco from Enron, but they were through with it anyway. ;-)
In case #1, the universe cannot be finite unless we are at its very center with extremely high precision. This is very unlikely, verging on divine intervention.
For case #3, the necessary physics to rework the cosmos on a grand scale has never been observed. Thus we can rule it out.
This leaves case #2: creation in the recent past, followed by expansion.
This is completely wrong. For two regions of space to reach thermal equilibrium, they must have been exchanging photons for hundreds of times the lightspeed travel time between them. For the opposite sides of the visible universe--at least 10 billion light years--they would have had to be thermalizing with each other for at least hundreds of billions of years. Since no objects appear to be that old, we can rule out that possibility. A hot/dense genesis + expansion makes many quantitative, testable predictions. It predicts that the universe should have been in good equilibrium in the past, and thus that its composition should be fairly uniform in all directions: this is observed. It predicts that photons that brought about thermal equilibrium should have been expanded into a low-energy radiation field with high uniformity and a perfect Planckian spectrum: this is observed. It predicts that if you look a particular distance in any direction, that the galaxies you see will have the same apparent age: they do. It predicts that nearby galaxies should look older, while distant galaxies should look younger: they do. It predicts that isotopes should have particular relative abundances from the epoch of thermonuclear equilibrium: they do. It predicts that stellar and fluorescent nebula light will not be highly polarized: it isn't. (In contrast, plasma cosmology predicts extremely large magnetic fields that would strongly polarize nearly all light sources.)The magnetic field problem is one of the more damning flaws in plasma cosmology. The theory holds that stars and galaxies are moved as strongly by electromagnetism as by gravity. Such magnetic fields would have a variety of trivially observable consequences: those consequences are not observed even slightly. Moreover, galactic collisions would be *spectactular* events, akin to a solar flare on a galactic scale, and no observed galaxy collision has shown appreciable magnetic phenomena.
Plasma cosmology is simply unphysical on every scale, predicting things that are not observed, and failing to predict things that are observed.
Overall the plasma cosmology theories are scientific failures. They cannot explain Olber's paradox ("Why is the sky dark?"), which is a pretty damning failure. They do not predict the existence of the cosmic microwave background, nor explain it's spectrum and uniformity. They do not explain the Hubble redshift. They do not explain why distant galaxies look different than nearby galaxies. They do not predict or explain the ratios of the primordial elements.
For little things like KVMs, modems, inkjet printers, etc. you can safely use the nameplate ratings.
For big things, determine how many machines you would ever conceivably want in the room. Choose the biggest, baddest equipment you could possibly want. 1U dual-proccesor machines, arrays of 15000 rpm hard drives, a desk full of 21 inch monitors, you name it. Then go to the manufacturers web sites and find the nameplate ratings for the various things, and add 'em all up. The total will be a number you won't easily outgrow.
Be sure to account for start-up loads. You don't want to trip a breaker by turning everything on at once. Hard drives draw a lot of power while they're spinning up, monitors while degaussing, laser printers while warming up the fusion rollers. This is just an educated guess, but use a factor of 2 for hard drives, and 5 for monitors. Read the specs for the laser printers very, very carefully and find the worst-case.
I'd go even farther. When many surge protectors divert a surge, they divert it into the ground wire. This causes a brief, high voltage spike on that circuit's ground relative to the other circuits in the room. The longer the ground wire is, the larger the spike. This spike can do nasty things to serial lines, KVM cables, and so forth that connect machines on different circuits.So if the building breaker box is farther than, say, 50 feet from the server room, I'd have a small breaker box installed in the server room. Also this lets you recover from a tripper breaker without getting the main breaker box unlocked.
If you can afford it, have a couple of separate circuits run from the main breaker box. This gives you someplace to plug in coffee pots and vacuum cleaners without disturbing the electronics.
If the room gets its own air conditioner, make sure that has a dedicated circuit from the main breaker box.
If you can afford it, have a big industrial surge protector installed at this breaker box. Also the breaker box is a good grounding point for surge protectors on your external data lines.
This is excellent advice. The electrical code is based on safe operation of motors and heaters. Bigger wires make your electronics more reliable by reducing voltage droop.Also, computers often don't draw sine wave current. They draw less current at the beginning and end of the AC cycle, and more in the middle. This means the peak current is larger than the sinewave loads envisioned by the electrical codes.
More excellent advice. Conduit is completely unacceptable for grounding computers. A grounding wire is cheaper than the cost of a single computer crash caused by a poor ground. Have an electrician tie all the racks and other metal stuff together with big ground wires. This will help protect the rest of your equipment if one of the devices has a ground fault. It'll also help reduce static electricity by giving you lots of big grounded metal things to touch. Wire is cheap compared to the cost of a single failure. Conduit does make running wires much easier. If there is no other wiring or fluorescent lights within a few feet, I'd use nonmetallic conduit, as metallic conduit can actually act as an antenna for picking up radio waves and coupling them into your data cables. OTOH if there are AC lines parallel to the run, metallic conduit is probably better, and be sure to make the electrician ground the conduit properly.