The Dune/A7 dispute basically amounted to this: A7 came out with a catchy tune. Dune rewrote the lyrics, and performed the music themselves. They were barred by court order from releasing their album with the song on it.
I realize my second example isn't real (and, for that matter, I usually use Vim or LyX myself...). And I also realize this isn't traditionally the area of copyright. If only it were that simple! It could be argued easily enough that the.doc format, by virtue of containing various symbols which are copyrighted by Microsoft (like header tags, or control codes, etc), is itself Microsoft's property.
This isn't as contrived an argument as it sounds either. It was decided a while back that a typeface could not be the subject of copyright, because that could lead to some trouble with freedom of press. If this were the end of it, then font files would be free of copyright. Unfortunately, PostScript is a programming language, and PostScript fonts rather resemble a program.
So while things printed with a font, or a scanning or even bitmap translation of the font, are free of copyright, the font file itself is copyrightable (the font names, of course, are trademarkable, but that just led to a lot of new names for old fonts).
The age of majority in New Brunswick is, in fact, 19.
It does vary between the provinces though.
There is no contract involved in ADSL here. You just call in, they deliver the equipment, and they start billing you. If you fail to pay, they cut you off. Otherwise, they can't really do anything.
I forgot to mention that the age of majority does vary between the provinces. The Age of Majority Act of the province of New Brunswick does state that it's 19, though.
The problem with this alternative is simply that there is a different *type* of scarcity in "Intellectual Property".
In the real world, if Sony makes TV's that are too expensive, I can go buy an RCA, or an el-cheapo Walmart one. Each unit may very well be functionally identical. In fact, it's quite possible that Walmart's subcontractor liked Sony's TV so much that they duplicated the design. Now you have a real alternative!
The problem with IP laws is basically that it comes down to owning knowledge. You can't make an alternative without breaking laws, because you "own" the idea itself (not just verbatim copying either.. The recent problems between the bands Dune and A7 over the song "Heaven" (which is nearly identical but performed seperately) demonstrate this).
The more ideas removed from the public domain, the harder it is to make an alternative. Imagine if Microsoft decided to sue OpenOffice because they "own" the Microsoft.doc format? Where's my alternative?
Actually, copyright law doesn't really work like traditional property. Financial gain is normally a factor in copyright infringement. Because P2P applications have no financial gain for the people doing the actual distribution, the whole idea is on shakey ground. (The debate of what constitutes "fair use" is now a hot topic, because of this).
To use your analogy for traditional infringement, P2P services would be more like walking into a bank, slipping the teller a note saying that you have a gun, and then walking back out. It's debatable whether a crime of any sort has been committed.
I'm just glad to live in Canada. Faced with rampant music swapping, the government imposed a levy on blank CD media. While annoying, it gives us grounds for a compulsory license for the music. If you download music, in Canada, with the intent to burn it to a regular audio CD, more than likely (this is still untested) no crime has been committed.
Silly argument. Recently I found myself out of work, and rather poor. I drastically cut back on my food intake, expenditures of almost any sort, but I kept my ADSL.
Good thing I did too. After months of old-fashioned jobseeking failed to turn up any leads, a contact on the 'net found me a job doing PHP for a local startup.
If you want people to make something of themselves and stop being a drain on the system, they have to have the resources to educate themselves, and to find employment.
A computer is a much better investment for a low-income family than cable TV or console game systems (which both seem much more acceptable by most people's standards).
I don't know about there, but here (in Canada), having ADSL doesn't require a contract. It doesn't require that the person be of majority either. They'll merrily disconnect your service if you stop paying, but otherwise, the money of a minor is as good as anyone's.
In fact, I got my first ADSL in my own name when I was 17 (the age of majority here is 19).
That comment is mostly FUD. The amount of signal reflection from the thin foil in a CD should pose little threat to the magnetron. A cavity magnetron can just as easily receive microwave signals as send them, so unless you're amplifying the input power somehow (like, by storing energy and outputting in pulses...), then this should prove harmless.
The real danger comes from the arcing in the CD itself. This can pose a bit of a hazard of the CD bursting into flames. After a time, though, the CD pits and takes on characteristics of a microwave antenna. You probably want to stop it at this point, or the CD will begin heating rapidly, and the reflected energy has a chance to rise (although, because of the low current carrying capacity of foil, not usually to levels dangerous to the magnetron).
Something like a fork, however, is more likely to damage the microwave. On top of the sparks streaming from the points, the charge-arc cycle can lead to a high-powered pulse waveform, which can cause the amplification required to damage a magnetron.
It's not impossible. In fact, it has been done several times (an example is here).
While building it on a breadboard is looking more difficult, discrete logic is still quite usable (with, for example, wire wrapping). In fact, in recent years, advanced families of TTL (or hybrid CMOS) (like ABT, etc) have boosted speed to a point where discrete logic can easily manage these clock rates (33MHz isn't that bad anyway).
That said, ISA and the standard 8-bit microprocessor busses were both much easier to interface to.
I was going to mod you up, but I decided to reply instead.
The limitation you refer to has very little to do with managing crosstalk. What actually happened is that the phone company was maximizing the number of channels they could fit on a single wire by using a technology called Frequency Division Multiplexing (FDM).
Since an average voice conversation has a bandwidth of about 2KHz, they built in a low-pass filter with a cutoff somewhere in the vicinity of 3KHz. This means they can heterodyne the channels, each (roughly)3KHz wide, onto a single wire.
Now, this means that the data rate (in terms of zero crossings per second (the original meaning of baud) is limited to about 2400. The "high speed" modems, all the way up to 56K, have a baud rate of 2400. This is a hard limit due to the phone company hardware.
What changed is the number of bits per baud. A 56K modem might use as many as 24 bits per baud, assuming the line is clear enough. The number of bits per baud is capped by the noise floor of the signal, which is also why you won't always connect at 56K (noisier lines can't handle the resolution).
In the move to digital networks, the same total channel datarate was designed into the switching systems. I'm not entirely sure the sampling rate and quantization parameters of these systems, though.
Well, the thing is that the device is a simple ARM-based single-board computer with two ethernet ports and a modem. It runs uCLinux. Anything you can cross-compile and cram into the flash, it will support.
It should be relatively simple to set up, say, PPPoE on one of the ports, a connection to a LAN on the other. Then write a simple script that watches the PPPoE connection status, and dials the modem if the PPPoE has been unable to connect for a certain amount of time.
The device would obviously have control of its own default gateway, which would solve that issue. Would the failover be transparent? Not quite, open connections would break. But it is simple, and would provide service through a DSL blackout.
Shameless Plug
on
GTK+ TTY Port
·
· Score: 3, Interesting
Ooh, more TTY stuff. I love stuff that runs on a TTY. aalib rocks. I have mplayer set up for movies, I have the original port for Quake 1... I run links for webbrowsing. I use centericq for my chat.
Even better, I wrote aavga2 to run Quake2 on aalib!
Now that Gtk+ is moving to TTY as well, maybe I can get rid of X entirely? *grin*
Hey, I resent that! At least one of us is logical:P
One of my previous partners (on my suggestion) bought me a lab white sapphire heart pendant. It looked big in the picture... it was ENORMOUS when I got it (it's about 3/4" across). It's still my favorite jewel, and it REALLY turns heads when I wear it on something black.
Substituting similar size and quality, the natural wins out. Comparable *price*, I'll take the lab gem any day!
... but I've dated 3 of them at different times. At this point, I can identify pretty much all of the common gems, and some of the rare ones. That sort of dishonesty would get you the boot, and quick.
But I've stayed away from diamonds, for the most part. Moissanite is prettier anyway. "ooh, sparkly".
Well, it sounds like even a class A vaccum tube amp is cheaper, both in parts (one single steel-drum style beam power tube), and in operation cost (a class A vaccum tube will be 50% efficient (which is aweful, really)).
Actually, while tubes can certainly be destroyed by EMP, it takes a much larger pulse to do so. A pulse that would destroy most tube equipment would also cause sparks to fly off anything metal, belt buckles to burn people, etc.
Solid state circuits (especially MOS devices, although bipolar are still vulnerable to a lesser extent) are vulnerable to voltage spikes. Vaccum tubes are vulnerable to POWER spikes. You can easily pulse 1MV through a small radio tube, as long as you do it for a very small time, without causing damage. Try that on a MOS device and it'll burn right through the insulating layer (ESD does just this).
While it doesn't require being very close to the blast to get a substantial voltage, you would probably feel the blast if you were close enough to get that kind of power from it.
A cavity magnetron, commonly found in microwave ovens and radars, is a tuned oscillator. It is a "metal can"-style vaccum tube.
It's commonly used as a microwave transciever, since it can be modulated with an external voltage, and can also (from what I recall) decode the modulation as well (the output voltage is proportional to the difference between the internal resonant signal and the incoming signal).
The one in a microwave oven is roughly 400W at 2.4GHz. That's quite a bit of power from a little tube.
Actually, quantization error (the error inherent in reducing data into a digital format, like the 16-bits you mention) is easily quantified. 16 bits gives a noise floor of about -96dB. The very best of the best vinyl has a noise floor of about -60dB.
I'm just glad you didn't try to make an argument based on time sampling, like most do. The frequency response of vinyl is aweful, full of peaks and valleys due to the medium.
It's much more likely that you enjoy the sound of distortion that comes with the analog system. I've had a couple old (decent) analog systems that put out very pleasant sound. The actual frequency response was AWEFUL!
DSP can certainly reproduce the effect you are looking for. If you accurately measure the impulse response of your audio system, it's almost trivial to make a DSP algorithm to duplicate it. In fact, if you measure the impulse response of your digital system as well, you can even add a deconvolution filter to remove any distortion caused by the analog parts of the digital system while adding back the distortion of your system.
There is a difference between something that can switch power to a 500kW load, and something rated at a power handling capacity of 500kW. A high-current switching element can likely pass thousands of amps, but may only have a power handling of, oh, 200W.
Remember, P=I^2*R. When switched fully "ON", "R" will be very, very low, causing a very low power dissipation. When switched fully "OFF", "I" will be very, very small (leakage current only). This leakage current across the large "R" still produces very small power dissipation.
Try to run one of these solid-state switching elements as a class A amplifier and see how long it takes to let the blue smoke out.
Your idea of using a class D amplifier works very well for audio frequencies, with a low-pass filter (a Chebyshev for audio, or a Bessel for something where the time domain is important) cutting off the higher harmonics. The thing to remember, though, is that standard PWM requires an oscillator frequency *many* times the sampling rate. 1MHz to drive an audio frequency signal (which has about 20KHz bandwidth) is quite common.
This causes real problems for higher frequencies, though, since IGBT's can't operate much past a MHz or two. Certainly, to broadcast in the range of 1MHz (AM radio, roughly), you would need a signal of at least 50MHz. Possible? Not easily. Even if it could be done, the high-order bandpass filter required to seperate out the higher frequencies to a point where it would be legal to transmit would get you.
It is *possible* that class D amplifiers may see their way into radio eventually, but this won't happen for quite a while.
This isn't so critical for FM radio, which commonly uses solid state components in a class C amplifier.
It's not so much the efficiency as the power handling. Here is an example.
FM radio transmitters are far cheaper than AM per watt of output power. The reason is that FM radio can get away with using a class C amplifier, and using a bandpass filter afterward to remove the harmonics. The class C is more efficient (about 75%, if I remember correctly), and dissipates less power, so a solid-state (transistor/MOSFET-based) amplifier can be used.
The problem is that class C isn't linear. It's not even *close* to linear. FM stores data in the frequency domain, so this doesn't matter. AM stores data in the time domain, so this distortion is unacceptable. The distortion of a class B amplifier, even, is ususally.
Because of this, huge class A amplifiers (at 50% maximum efficiency) are used for AM transmission. The other 50% of power is being dissipatd as heat! When you hear an AM station broadcasting at 10 or 20kW, that means you have 10 or 20kW of heat being dissipated in the amplifier.
This heat would fry a transistor alive. Instead, they use massive vaccum "tubes" that look more like steel drums. These giant steel vats can easily handle the dissipation of the transmission, but are much more expensive.
Yes, we *can* start treating the wires like transmission lines ("antennas" that don't broadcast, for the non-hardware types). This is what recent advances such as low-voltage differential signalling and high-speed serial interfaces like PCI-X are all about.
The reasons that we can't with current generation hardware are:
Ground Noise. Because of the way the signal is propogated at higher and higher speed and lower voltages, the noise margins start looking bad, and the amount of noise in the ground plane itself becomes unacceptable. Advances like internal ground planes (with 6 and 8-layer boards) have helped up till now, but these advances are getting harder. Differential signalling helps this by providing both a negative and positive version, which can be compared at the other end.
Crosstalk. Two wires (or traces on a PC-board), sitting next to each other, acts as a capacitor. As frequency goes up, a capacitor starts looking more and more like a short circuit, instead of the open circuit you would expect. This means that the two wires will inject signal into each other, rendering communication impossible. Differential signalling again helps this one, because you can then run signal lines in pairs, which should pick up interference in roughly the same amount (common mode), which can be ignored on the other side by the comparator.
Capacitative load. This is related to, but different than, the one above. Basically, the capacitance between the traces and the ground plane becomes more important, and dissipates more of the power sent into the traces, slowing rise-times on traces and increasing propogation delays. Low-dialectric circuit board materials are becoming more and more common to compensate.
Synchronization. Keeping a set of parallel data lines in sync becomes a lot more difficult when the data rate far exceeds the propogation delay. When you have many bits still on the wire travelling to the next chip, keeping the bits arriving simultaneously becomes nearly impossible. This has led to some of the high-speed serial busses, like PCI-X and Serial ATA. Since you have only one line, you can push bits onto it at a much higher speed without worrying about synchronizing it.
Slashdot Mirror
I realize my second example isn't real (and, for that matter, I usually use Vim or LyX myself...). And I also realize this isn't traditionally the area of copyright. If only it were that simple! It could be argued easily enough that the .doc format, by virtue of containing various symbols which are copyrighted by Microsoft (like header tags, or control codes, etc), is itself Microsoft's property.
This isn't as contrived an argument as it sounds either. It was decided a while back that a typeface could not be the subject of copyright, because that could lead to some trouble with freedom of press. If this were the end of it, then font files would be free of copyright. Unfortunately, PostScript is a programming language, and PostScript fonts rather resemble a program.
So while things printed with a font, or a scanning or even bitmap translation of the font, are free of copyright, the font file itself is copyrightable (the font names, of course, are trademarkable, but that just led to a lot of new names for old fonts).
There is no contract involved in ADSL here. You just call in, they deliver the equipment, and they start billing you. If you fail to pay, they cut you off. Otherwise, they can't really do anything.
I forgot to mention that the age of majority does vary between the provinces. The Age of Majority Act of the province of New Brunswick does state that it's 19, though.
In the real world, if Sony makes TV's that are too expensive, I can go buy an RCA, or an el-cheapo Walmart one. Each unit may very well be functionally identical. In fact, it's quite possible that Walmart's subcontractor liked Sony's TV so much that they duplicated the design. Now you have a real alternative!
The problem with IP laws is basically that it comes down to owning knowledge. You can't make an alternative without breaking laws, because you "own" the idea itself (not just verbatim copying either.. The recent problems between the bands Dune and A7 over the song "Heaven" (which is nearly identical but performed seperately) demonstrate this).
The more ideas removed from the public domain, the harder it is to make an alternative. Imagine if Microsoft decided to sue OpenOffice because they "own" the Microsoft .doc format? Where's my alternative?
To use your analogy for traditional infringement, P2P services would be more like walking into a bank, slipping the teller a note saying that you have a gun, and then walking back out. It's debatable whether a crime of any sort has been committed.
I'm just glad to live in Canada. Faced with rampant music swapping, the government imposed a levy on blank CD media. While annoying, it gives us grounds for a compulsory license for the music. If you download music, in Canada, with the intent to burn it to a regular audio CD, more than likely (this is still untested) no crime has been committed.
Good thing I did too. After months of old-fashioned jobseeking failed to turn up any leads, a contact on the 'net found me a job doing PHP for a local startup.
If you want people to make something of themselves and stop being a drain on the system, they have to have the resources to educate themselves, and to find employment.
A computer is a much better investment for a low-income family than cable TV or console game systems (which both seem much more acceptable by most people's standards).
In fact, I got my first ADSL in my own name when I was 17 (the age of majority here is 19).
The real danger comes from the arcing in the CD itself. This can pose a bit of a hazard of the CD bursting into flames. After a time, though, the CD pits and takes on characteristics of a microwave antenna. You probably want to stop it at this point, or the CD will begin heating rapidly, and the reflected energy has a chance to rise (although, because of the low current carrying capacity of foil, not usually to levels dangerous to the magnetron).
Something like a fork, however, is more likely to damage the microwave. On top of the sparks streaming from the points, the charge-arc cycle can lead to a high-powered pulse waveform, which can cause the amplification required to damage a magnetron.
While building it on a breadboard is looking more difficult, discrete logic is still quite usable (with, for example, wire wrapping). In fact, in recent years, advanced families of TTL (or hybrid CMOS) (like ABT, etc) have boosted speed to a point where discrete logic can easily manage these clock rates (33MHz isn't that bad anyway).
That said, ISA and the standard 8-bit microprocessor busses were both much easier to interface to.
The limitation you refer to has very little to do with managing crosstalk. What actually happened is that the phone company was maximizing the number of channels they could fit on a single wire by using a technology called Frequency Division Multiplexing (FDM).
Since an average voice conversation has a bandwidth of about 2KHz, they built in a low-pass filter with a cutoff somewhere in the vicinity of 3KHz. This means they can heterodyne the channels, each (roughly)3KHz wide, onto a single wire.
Now, this means that the data rate (in terms of zero crossings per second (the original meaning of baud) is limited to about 2400. The "high speed" modems, all the way up to 56K, have a baud rate of 2400. This is a hard limit due to the phone company hardware.
What changed is the number of bits per baud. A 56K modem might use as many as 24 bits per baud, assuming the line is clear enough. The number of bits per baud is capped by the noise floor of the signal, which is also why you won't always connect at 56K (noisier lines can't handle the resolution).
In the move to digital networks, the same total channel datarate was designed into the switching systems. I'm not entirely sure the sampling rate and quantization parameters of these systems, though.
It should be relatively simple to set up, say, PPPoE on one of the ports, a connection to a LAN on the other. Then write a simple script that watches the PPPoE connection status, and dials the modem if the PPPoE has been unable to connect for a certain amount of time.
The device would obviously have control of its own default gateway, which would solve that issue. Would the failover be transparent? Not quite, open connections would break. But it is simple, and would provide service through a DSL blackout.
Even better, I wrote aavga2 to run Quake2 on aalib!
Now that Gtk+ is moving to TTY as well, maybe I can get rid of X entirely? *grin*
Yup, that would do it for me.
One of my previous partners (on my suggestion) bought me a lab white sapphire heart pendant. It looked big in the picture... it was ENORMOUS when I got it (it's about 3/4" across). It's still my favorite jewel, and it REALLY turns heads when I wear it on something black.
Substituting similar size and quality, the natural wins out. Comparable *price*, I'll take the lab gem any day!
But I've stayed away from diamonds, for the most part. Moissanite is prettier anyway. "ooh, sparkly".
Well, it sounds like even a class A vaccum tube amp is cheaper, both in parts (one single steel-drum style beam power tube), and in operation cost (a class A vaccum tube will be 50% efficient (which is aweful, really)).
Solid state circuits (especially MOS devices, although bipolar are still vulnerable to a lesser extent) are vulnerable to voltage spikes. Vaccum tubes are vulnerable to POWER spikes. You can easily pulse 1MV through a small radio tube, as long as you do it for a very small time, without causing damage. Try that on a MOS device and it'll burn right through the insulating layer (ESD does just this).
While it doesn't require being very close to the blast to get a substantial voltage, you would probably feel the blast if you were close enough to get that kind of power from it.
It's commonly used as a microwave transciever, since it can be modulated with an external voltage, and can also (from what I recall) decode the modulation as well (the output voltage is proportional to the difference between the internal resonant signal and the incoming signal).
The one in a microwave oven is roughly 400W at 2.4GHz. That's quite a bit of power from a little tube.
I'm just glad you didn't try to make an argument based on time sampling, like most do. The frequency response of vinyl is aweful, full of peaks and valleys due to the medium.
It's much more likely that you enjoy the sound of distortion that comes with the analog system. I've had a couple old (decent) analog systems that put out very pleasant sound. The actual frequency response was AWEFUL!
DSP can certainly reproduce the effect you are looking for. If you accurately measure the impulse response of your audio system, it's almost trivial to make a DSP algorithm to duplicate it. In fact, if you measure the impulse response of your digital system as well, you can even add a deconvolution filter to remove any distortion caused by the analog parts of the digital system while adding back the distortion of your system.
Remember, P=I^2*R. When switched fully "ON", "R" will be very, very low, causing a very low power dissipation. When switched fully "OFF", "I" will be very, very small (leakage current only). This leakage current across the large "R" still produces very small power dissipation.
Try to run one of these solid-state switching elements as a class A amplifier and see how long it takes to let the blue smoke out.
This causes real problems for higher frequencies, though, since IGBT's can't operate much past a MHz or two. Certainly, to broadcast in the range of 1MHz (AM radio, roughly), you would need a signal of at least 50MHz. Possible? Not easily. Even if it could be done, the high-order bandpass filter required to seperate out the higher frequencies to a point where it would be legal to transmit would get you.
It is *possible* that class D amplifiers may see their way into radio eventually, but this won't happen for quite a while.
This isn't so critical for FM radio, which commonly uses solid state components in a class C amplifier.
FM radio transmitters are far cheaper than AM per watt of output power. The reason is that FM radio can get away with using a class C amplifier, and using a bandpass filter afterward to remove the harmonics. The class C is more efficient (about 75%, if I remember correctly), and dissipates less power, so a solid-state (transistor/MOSFET-based) amplifier can be used.
The problem is that class C isn't linear. It's not even *close* to linear. FM stores data in the frequency domain, so this doesn't matter. AM stores data in the time domain, so this distortion is unacceptable. The distortion of a class B amplifier, even, is ususally.
Because of this, huge class A amplifiers (at 50% maximum efficiency) are used for AM transmission. The other 50% of power is being dissipatd as heat! When you hear an AM station broadcasting at 10 or 20kW, that means you have 10 or 20kW of heat being dissipated in the amplifier.
This heat would fry a transistor alive. Instead, they use massive vaccum "tubes" that look more like steel drums. These giant steel vats can easily handle the dissipation of the transmission, but are much more expensive.
The reasons that we can't with current generation hardware are:
I guess fire isn't invented where you're from?
Although I would still expect hardware hackers tend to have a bit of an interest in the kind of software that lets them tinker. I know I do.