The problem is that computer software is relatively easy for an expert in the field to create, and there are rarely multiple ways of doing something, for compatibility reasons. It isn't like inventing a pencil with an eraser on the end, where another company can make a pencil and a separate eraser and still be compatible. Compatibility is only an issue in software, and therefore deserves consideration in patent protection.
The fact of the matter is that there are very, very few algorithms that are sufficiently novel to the average programmer to allow patents. I can think of maybe ten, and they all have to do with data compression and/or analysis (Lempel-Ziv, Huffman coding, Fast Fourier Transform, Discrete Cosine Transform... okay, you've got me... I can't come up with ten...). In the case of those, they have become parts of ubiquitous standards, such that if patents were used against those who have implemented these algorithms, it would be the end of computing as we know it.
I can't think of any patent that hasn't become ubiquitous that isn't obvious. For compatibility reasons, the non-obvious ones (i.e. the ones that don't have eighteen work-arounds because they're so simple and transparent that they should never have been patentable) become implemented everywhere, and must, by definition, be implemented everywhere.
Thus, Software patents are like a car company being allowed to patent "an apparatus that allows driving on the right side of the road", if there were such an apparatus. It effectively would have created a twenty year monopoly on all automobile manufacture, regardless of design, because no other vehicle could be compatible with existing roads. That's not what patents were designed to do. They were designed to provide a limited monopoly on a single design. The very nature of software makes this impossible, and -that- is why software patents are inherently bad.
Not all IP laws protect open source. Copyright laws protect open source. Patent laws threaten its very existence, and further make commercial software development dicey except for companies that are already well established....
I see this and it gives me hope. Every silly IP lawsuit like this helps illustrate how broken the patent process is, and further, how incredibly naive people are who believe that software should be patentable.... Eventually, our government will figure out how bone-headed software patents are and will realize that it allows random companies like Kodak (which have no real stake in the software business) to bully the primary providers of software technology over things that shouldn't have been patentable to begin with. Once that wheel starts turning, software patents will start to crumble under the weight of their own inanity.
Or just nuke and reinstall your existing OS. There's no way that's normal behavior unless something is really wrong. Best guess... probably some freak effect of an interrupt either not consistently being delivered or being frequently blocked by something else... you know, like the vertical blanking interrupt running at a quarter speed or something.
That's my guess. A lot of things happen during the vertical blanking interval or on some other similar periodic interrupt. In most OSes, this includes screen updates and mouse pointer redraws. This could be anything from a buggy driver to an IRQ conflict, or possibly even a bad trace on the motherboard (though the latter isn't anywhere near as likely).
If an OS reinstall doesn't solve the problem, there's probably something weird going on in the BIOS settings and/or the motherboard itself. Pull the BIOS battery for an hour. Try again. If that doesn't work... is your clock running slowly, too? If so, buy a new computer. If not... buy a new computer. EIther way.:-p
<rant>And speaking of IRQ conflicts... why hasn't any motherboard manufacturer broken with tradition and actually added enough distinctly addressable interrupt lines? I mean, the Mac has supported 64+ interrupts on its interrupt controller since 1995. Does it really take a decade of engineering to figure out how to cascade two interrupt controllers and add a driver to support it? Sheesh!</rant>
Sigh. Another victim of a 2004 computer crammed into a 1981 architecture....
If an LCD panel turns pink, that's not the front of the LCD panel becoming discolored. That's your backlight failing. Like any bulb, they eventually fail. Unlike most bulbs, though, they're kind enough to do so fairly slowly....
A memory card is a compilation, which is generally treated as a single work for the purposes of damages. Thus, in all likelihood, a judge would rule that it was only one copyright violation, if that.
Further, if the state in question has a law that says that the finder of lost property owns that property if not claimed, then that means any copyright to those photos, unless those photos were non-original, was likely transferred along with the physical card.
Odds are, there's no law being broken here, assuming the person did due diligence by turning it in to the police and waiting to see if someone claimed it. Otherwise, there could be all sorts of charges from copyright violation to theft... but I digress.
And of course, given that the web site is making up stories about the photos, odds are that the story about the card being found in a cab is equally made up. Just a hunch.
And they would know the person's address how? For that matter, they would know the person's -name- how?
And there are different found property laws in different states. It's possible that "finders keepers" applies here. Whether it's right or not is another issue. Of course, if the owner doesn't come forward... well,then, I guess it doesn't matter.
I've used a lot of generic RAM in Macs---a bunch of 61/71/8100 machines, a 7600, a beige G3, a Pismo, a white iBook, a G4/450, and a DP G5 2GHz. In all that time, I've only had one single failure and that was because the chip turned out to be non-compliant with the JEDEC specifications. (It was at a higher density than the specification allowed. It didn't work in my Athlon machine, either, yet the vendor tested it and claimed that it worked.)
That's not saying that your mileage won't vary, but generally if a RAM vendor says "check compatibility", it's non-compliant garbage and should not be put into any machine at any cost. Anything else should generally work in just about any hardware you might use.
Not only cheaper to design, but much cheaper to fab. The technology for CPUs is currently at 90nm. RAM, by contrast, is moving from a 130nm to 110nm process this year, and there are talks of plans to move to 90nm next year. That means that by the time the DRAM vendors switch to a smaller process, the industry as a whole typically has had a whole year to work out the kinks for them.
It's people who do things like you describe that make me even more intent on using open source software whenever possible.
This weekend, one software company (which will remain nameless) lost a $1000 purchase because they were too paranoid to trust me. They required a hardware dongle. I chose the competitor (which was comparably priced) solely because of that lack of trust. Paranoia in a software company is a sign that the company doesn't actually believe its software is really worth what it is charging, IMHO. And when I'm out in the field, the last thing I want to worry about is some dongle deciding to die on me.
The alternative, spending hundreds of additional dollars for a sparre dongle, is equally unacceptable. (Also, if I had a dongle for every $500+ piece of software I've bought over the years, I'd have to add two more USB ports. There are limits to the number of USB devices on a bus, even with powered hubs....)
Trust goes both ways. If a company doesn't trust me to not steal their software, I don't trust them to continue supporting it, thus I'm much more likely to buy from someone who does.
And CPU serial numbers are a terrible idea. What if I want to move the software from one machine to another? What if, God forbid, my CPU dies? Some companies would try to deny me the right to continue using their software that I paid for, even though by law, it is my right to do so. The more ingrained these mechanisms get---the more complacent we become about them---the farther some unscrupulous companies will go, not only to protect their legitimate rights, but also to find ways to squeeze customers out of more money for permission to do things that are currently protected by law.
It is time for software users of the world to unite in saying that there is a line that should not---nay must not---be crossed. We must stand firm against those who would continue to iteratively erode our rights, and when a company takes things too far, we must say "no"---not only to that product, but to every product from that company from then on to discourage other companies from trying similar terrorist actions. Only by ending the cycle of abuse can our rights as consumers be protected as they should be.
If you -really- get a kernel panic once a week and aren't just trolling, there's something wrong with your girlfriend's hardware---probably bad RAM.
If you exclude panics caused by failing motherboards and device drivers that I was writing, I have only seem about four or five kernel panics in the 4 years I've been using Mac OS X. A weekly panic is an order of magnitude more failures than normal. The word "AppleCare" springs suddenly to mind.
Umm... if you don't count the blue-and-white G3, all professional PowerMacs have been either black, grey, or metal since Apple left beige behind. Only the consumer products (the iMac, iBook, and iPod) are white and/or colors.
The problem is that our government doesn't have the guts to require everyone to turn in every bill printed without those mechanisms within the next two years and exchange it for a new bill. The fact is that you don't have to counterfeit the latest bills. Instead, you counterfeit the -old- bills, which had almost no protection... no plastic strip, no watermark, etc.
Here's a novel idea. Stop printing money and rescind all paper currency. It's easy to print paper. No matter what the government does, it will always be relatively easy for people who really want to counterfeit bills to do so. It is relatively difficult, by comparison, to mint pieces of metal.
The obvious problem is that people become paranoid about losing a coin once it becomes worth something. That's why you also cap the maximum value of any piece of currency at $20. That is the largest bill that most people carry around when not on a trip anyway. That's the largest bill that most ATMs dispense. Just do away with everything with a higher face value.
The net impact will be that anything costing over a couple of hundred dollars will end up being purchased via check or plastic. Well, that's the case anyway, for the most part. Couple this with a nationwide ban on profiteering from ATMs, and the problem of people on trips goes away, leaving the only people severely impacted being those doing illegal transactions (drugs, money laundering, ordering a hit on someone, etc.). And frankly, causing inconvenience for those folks is a good thing, right?
The governments of the world all have one thing in common.... When something goes wrong, they try to patch around the problem with the minimum necessary effort... kind of like a bad sysadmin. As a result, things keep going wrong because they aren't solving the fundamental problem. The fundamental problem is that paper currency is a bad idea. We all need to get over it and move on.:-)
Nuclear power doesn't make greenhouse gasses, therefore it is the logical choice.
But that's complete and utter horseshit. Nuclear power generates -huge- amounts of water vapor. Guess what the number one greenhouse gas in our atmosphere is? Water vapor causes 60% of the world's greenhouse effect.
We can't just pretend that energy expended on this planet won't cause changes to our environment. By its very nature, introducing -any- energy into the system, whether through burning something or through fission, causes global environmental change.
The real question is what the nature of that change will be. We need a heck of a lot better information than the pseudo-science that seems to dominate environmental studies these days. There's a very real possibility that the ecosystem as a whole is self-correcting, and that nothing we do (within limits) will have any lasting impact. There's also a very real possibility that in a few hundred years, Earth will look like Mars (frozen) or Venus (melting point of lead). The point is not that we don't know for certain, but that we don't seem to have even the slightest idea.
Unless I hear otherwise, I'm going to assume that whatever we use for energy is going to screw up our environment, eventually we're all going to die, and we should make the most of it while we can. But I digress....
There is exactly one way to destroy terrorism: education. Whether it is the secular education that we tend to focus upon in the U.S. or religious education, the net impact is the same. Educated people generally do not go off and fly airplanes into buildings.
What we have in parts of the world are basically complete bozos with charisma, who twist the words of the Quran and turn it into a message of hate and violence, when in reality it is about justice, mercy, and love.
Many of the world's major religions have gone through a period of "let's blow something up"... the Davidic Empire, the Crusades, the St. Bartholomew's Day Massacre, the last few decades in Ireland... the list is seemingly endless. More often than not, these conflicts of religion have turned to bloodshed because a few zealots were able to convince the largely-uneducated masses to commit barbaric acts.
The only way to stop terrorism in its tracks is to cut off the supply of easy marks... to usher in a new era of prosperity and education so that no man, woman, or child will feel so utterly helpless that he or she feels the only way out is through taking the life of another. Only then can we truly be free.
Actually, that's not entirely true. In fact, it's backwards, and thus the conclusions are quite wrong.
First, Edison's method involved vertical deflection. It was abandoned long ago, not because of patents, but because the quality sucked and the reliability sucked worse.
Vertical needle deflection is ineffectual for two reasons. First, dust tends to gather in the bottoms of the grooves. Second, the needle sits at the bottom of the groove and is weighted to press down on the bottom. With vertical deflection, the very act of deflecting the needle ends up re-etching the recording surface over time, rendering the record useless.
The modern side-to-side deflection was the obvious solution to the problem, since a sharp needle no longer needed to be pushed downwards under pressure against the actual recording surface, but could instead be deflected horizontally against much lower resistance.
Modern (stereo) records actually combine the two methods. They have dual slanted needle deflections. Each channel is encoded on one of those slanted axes, thus giving you stereo separation by virtue of the axes being roughly perpendicular to each other. And, of course, the bottom of the groove isn't a recording surface, so to a large extent, you still get the long-term viability of a horizontal deflection.
We now return you to your regularly scheduled patent discussion.
GW: It is not a rumor, it is a fact. That twelve-year-old boy does have weapons of mass deeeestruction. We must protect ourselves against this threat to our nation's security.
Boy: Yeesh. I knew Eminem sucked, but I destroyed that CD years ago.
GW: We have invaded Greenfield, TN. Most of the locals cheered as we toppled the bank sign, but were a little puzzled why the military vehicles went through their stop light and didn't slow down for the speed trap that followed. In any case, we have secured the household and are searching for the Eminem CDs of mass distraction.
Anchor: I thought you said there were weapons of mass destruction in his house.
GW: I will launch a full-scale investigation into the intelligence failure that broght on this terrible tragedy.....
I would content that the lack of the floppy drive in Macs is more closely likened to the lack of a cigarette lighter in most automobiles. (Most have a DC outlet with a cover plug, and the lighter is an optional add-on.)
First, both are minor parts of the vehicle's functionality. Second, they're features that most people don't want or need. Third, they're features that the people who do want them really shouldn't be using in the first place, as they are harmful to your health (or at least the reliability of your data and/or your ability to concentrate on the road).
And we didn't abandon SCSI. Your firewire hard drive uses SBP2. That's SCSI commands sent over a serial tunnel. Want to know what your CD/DVD drive uses? ATAPI. That's SCSI packets encapsulated into ATA requests.
SCSI is like the force. "If you kill me, you will only make me stronger." And so, in effect, by abandoning the actual parallel SCSI physical transport (which really sucked, IMHO), the SCSI protocol (which doesn't) lives on in nearly every computer built in the past several years, PC or Mac.
Yes, but some would argue that a common way to steal a car is to tow it, then disarm the alarm when you get to your chop shop. Motion detectors aren't inherently bad, it's just that no company has built one that detects only the right variety of motion---that is the motion of the tires/axle in a circular fashion....:-)
And we have the right to tell them to go screw themselves if they don't want to make it available under reasonable terms. The problem is that people did that and suddenly they respond with "Nobody's buying movies! Everybody must be stealing them!"
No, my rights end where they injure others. My rights to watch a DVD on a Linux box do not injure the movie industry, therefore those rights are inalienable. Those who say otherwise are the greatest threat to the freedom of our country and our world. We must stand firm.
As Ray Bradbury put it in Fahrenheit 451:
I saw the way things were going, a long time back. I said nothing. I'm one of the innocents who could have spoken up and out when no one would listen to the 'guilty,' but I did not speak and thus became guilty myself. And when finally they set the structure to burn the books, using the firemen, I grunted a few times and subsided, for there were no others grunting or yelling with me, by then.
T1: number of bytes in T2
T2: number of times to repeat "Byte"
Or, for stereo:
T1 T2 xx xx xx Byte
04 01 93 BF 60 00
I do believe that's 6-7 bytes shorter than the header on a MIDI file, thank you very much.
Re:I try to avoid them altogether.
on
Fake ATM Fraud Expose
·
· Score: 2, Interesting
It takes less than a dollar worth of materials and a matter of seconds to capture a fingerprint off of... pretty much anything. Voice identification can be captured with a tape recorder just as quickly. With the exception of retina scans, biometrics add a trivial amount of protection, and frankly, I don't want anything resembling a laser anywhere near my eyes.
What I want to see is something that reads neruoelectric signatures. For the initial version, you'd think about your favorite food while leaning your head against a sensor pad. Of course, that could be captured, but that's just phase 1.
Phase 2 is to look at am image shown on the screen. When you sign up for an account, they'd do this once and store the neural impulses generated. From then on, they would show you the image and send the neural signature to the bank. The bank would compare the results and authorize the transaction, and would send a new image to display. You would see the second image, and the neural impulses generated by this second image would be sent back to the bank to store for the next time you tried to make a transaction.
The key requirements are that each transaction could require confirming the neural signature generated by any one or several of the prior images and the images sent for generating new signatures must be taken from a large enough database to get a high degree of variation. Finally, there must be expiration for old images, as one would expect one's reaction to an image to drift over time. Thus, an account unused in 90 days would be frozen until in-person verification could take place.
In such a case, in the unlikely event that someone were able to steal access to someone's account by taking enough prior neural signatures, they would still have to generate a new neural signature for the new image, which would mean that either it would be completely fictitious (which could probably be detected), a copy of some prior signature (which would definitely be detected and an alarm would sound), or would be the signature generated by the criminal, which could then be used as positive identification once that person gets caught.
Slashdot Mirror
The fact of the matter is that there are very, very few algorithms that are sufficiently novel to the average programmer to allow patents. I can think of maybe ten, and they all have to do with data compression and/or analysis (Lempel-Ziv, Huffman coding, Fast Fourier Transform, Discrete Cosine Transform... okay, you've got me... I can't come up with ten...). In the case of those, they have become parts of ubiquitous standards, such that if patents were used against those who have implemented these algorithms, it would be the end of computing as we know it.
I can't think of any patent that hasn't become ubiquitous that isn't obvious. For compatibility reasons, the non-obvious ones (i.e. the ones that don't have eighteen work-arounds because they're so simple and transparent that they should never have been patentable) become implemented everywhere, and must, by definition, be implemented everywhere.
Thus, Software patents are like a car company being allowed to patent "an apparatus that allows driving on the right side of the road", if there were such an apparatus. It effectively would have created a twenty year monopoly on all automobile manufacture, regardless of design, because no other vehicle could be compatible with existing roads. That's not what patents were designed to do. They were designed to provide a limited monopoly on a single design. The very nature of software makes this impossible, and -that- is why software patents are inherently bad.
By direct action:
- OpenDoc
- Network servers in general
- Web Services
- Applescript (tell Finder to empty trash)
Without direct action:- #! notation (e.g. #!/bin/sh)
- File type and creator
- File extensions
Oh, and I haven't seen the patent numbers posted yet. 5,206,951, 5,421,012, and 5,226,161.I see this and it gives me hope. Every silly IP lawsuit like this helps illustrate how broken the patent process is, and further, how incredibly naive people are who believe that software should be patentable.... Eventually, our government will figure out how bone-headed software patents are and will realize that it allows random companies like Kodak (which have no real stake in the software business) to bully the primary providers of software technology over things that shouldn't have been patentable to begin with. Once that wheel starts turning, software patents will start to crumble under the weight of their own inanity.
Les brevets sont mort. Vive les brevets.
That's my guess. A lot of things happen during the vertical blanking interval or on some other similar periodic interrupt. In most OSes, this includes screen updates and mouse pointer redraws. This could be anything from a buggy driver to an IRQ conflict, or possibly even a bad trace on the motherboard (though the latter isn't anywhere near as likely).
If an OS reinstall doesn't solve the problem, there's probably something weird going on in the BIOS settings and/or the motherboard itself. Pull the BIOS battery for an hour. Try again. If that doesn't work... is your clock running slowly, too? If so, buy a new computer. If not... buy a new computer. EIther way. :-p
<rant>And speaking of IRQ conflicts... why hasn't any motherboard manufacturer broken with tradition and actually added enough distinctly addressable interrupt lines? I mean, the Mac has supported 64+ interrupts on its interrupt controller since 1995. Does it really take a decade of engineering to figure out how to cascade two interrupt controllers and add a driver to support it? Sheesh!</rant>
Sigh. Another victim of a 2004 computer crammed into a 1981 architecture....
Further, if the state in question has a law that says that the finder of lost property owns that property if not claimed, then that means any copyright to those photos, unless those photos were non-original, was likely transferred along with the physical card.
Odds are, there's no law being broken here, assuming the person did due diligence by turning it in to the police and waiting to see if someone claimed it. Otherwise, there could be all sorts of charges from copyright violation to theft... but I digress.
And of course, given that the web site is making up stories about the photos, odds are that the story about the card being found in a cab is equally made up. Just a hunch.
And there are different found property laws in different states. It's possible that "finders keepers" applies here. Whether it's right or not is another issue. Of course, if the owner doesn't come forward... well,then, I guess it doesn't matter.
That's not saying that your mileage won't vary, but generally if a RAM vendor says "check compatibility", it's non-compliant garbage and should not be put into any machine at any cost. Anything else should generally work in just about any hardware you might use.
Just MHO.
This weekend, one software company (which will remain nameless) lost a $1000 purchase because they were too paranoid to trust me. They required a hardware dongle. I chose the competitor (which was comparably priced) solely because of that lack of trust. Paranoia in a software company is a sign that the company doesn't actually believe its software is really worth what it is charging, IMHO. And when I'm out in the field, the last thing I want to worry about is some dongle deciding to die on me.
The alternative, spending hundreds of additional dollars for a sparre dongle, is equally unacceptable. (Also, if I had a dongle for every $500+ piece of software I've bought over the years, I'd have to add two more USB ports. There are limits to the number of USB devices on a bus, even with powered hubs....)
Trust goes both ways. If a company doesn't trust me to not steal their software, I don't trust them to continue supporting it, thus I'm much more likely to buy from someone who does.
And CPU serial numbers are a terrible idea. What if I want to move the software from one machine to another? What if, God forbid, my CPU dies? Some companies would try to deny me the right to continue using their software that I paid for, even though by law, it is my right to do so. The more ingrained these mechanisms get---the more complacent we become about them---the farther some unscrupulous companies will go, not only to protect their legitimate rights, but also to find ways to squeeze customers out of more money for permission to do things that are currently protected by law.
It is time for software users of the world to unite in saying that there is a line that should not---nay must not---be crossed. We must stand firm against those who would continue to iteratively erode our rights, and when a company takes things too far, we must say "no"---not only to that product, but to every product from that company from then on to discourage other companies from trying similar terrorist actions. Only by ending the cycle of abuse can our rights as consumers be protected as they should be.
You don't like it? Well, if you don't like it, what do you want to do?
I want to sing and dance, I want to sing and dance....
If you exclude panics caused by failing motherboards and device drivers that I was writing, I have only seem about four or five kernel panics in the 4 years I've been using Mac OS X. A weekly panic is an order of magnitude more failures than normal. The word "AppleCare" springs suddenly to mind.
Here's a novel idea. Stop printing money and rescind all paper currency. It's easy to print paper. No matter what the government does, it will always be relatively easy for people who really want to counterfeit bills to do so. It is relatively difficult, by comparison, to mint pieces of metal.
The obvious problem is that people become paranoid about losing a coin once it becomes worth something. That's why you also cap the maximum value of any piece of currency at $20. That is the largest bill that most people carry around when not on a trip anyway. That's the largest bill that most ATMs dispense. Just do away with everything with a higher face value.
The net impact will be that anything costing over a couple of hundred dollars will end up being purchased via check or plastic. Well, that's the case anyway, for the most part. Couple this with a nationwide ban on profiteering from ATMs, and the problem of people on trips goes away, leaving the only people severely impacted being those doing illegal transactions (drugs, money laundering, ordering a hit on someone, etc.). And frankly, causing inconvenience for those folks is a good thing, right?
The governments of the world all have one thing in common.... When something goes wrong, they try to patch around the problem with the minimum necessary effort... kind of like a bad sysadmin. As a result, things keep going wrong because they aren't solving the fundamental problem. The fundamental problem is that paper currency is a bad idea. We all need to get over it and move on. :-)
Just my $0.019997 (adjusted for inflation).
But that's complete and utter horseshit. Nuclear power generates -huge- amounts of water vapor. Guess what the number one greenhouse gas in our atmosphere is? Water vapor causes 60% of the world's greenhouse effect.
We can't just pretend that energy expended on this planet won't cause changes to our environment. By its very nature, introducing -any- energy into the system, whether through burning something or through fission, causes global environmental change.
The real question is what the nature of that change will be. We need a heck of a lot better information than the pseudo-science that seems to dominate environmental studies these days. There's a very real possibility that the ecosystem as a whole is self-correcting, and that nothing we do (within limits) will have any lasting impact. There's also a very real possibility that in a few hundred years, Earth will look like Mars (frozen) or Venus (melting point of lead). The point is not that we don't know for certain, but that we don't seem to have even the slightest idea.
Unless I hear otherwise, I'm going to assume that whatever we use for energy is going to screw up our environment, eventually we're all going to die, and we should make the most of it while we can. But I digress....
What we have in parts of the world are basically complete bozos with charisma, who twist the words of the Quran and turn it into a message of hate and violence, when in reality it is about justice, mercy, and love.
Many of the world's major religions have gone through a period of "let's blow something up"... the Davidic Empire, the Crusades, the St. Bartholomew's Day Massacre, the last few decades in Ireland... the list is seemingly endless. More often than not, these conflicts of religion have turned to bloodshed because a few zealots were able to convince the largely-uneducated masses to commit barbaric acts.
The only way to stop terrorism in its tracks is to cut off the supply of easy marks... to usher in a new era of prosperity and education so that no man, woman, or child will feel so utterly helpless that he or she feels the only way out is through taking the life of another. Only then can we truly be free.
First, Edison's method involved vertical deflection. It was abandoned long ago, not because of patents, but because the quality sucked and the reliability sucked worse.
Vertical needle deflection is ineffectual for two reasons. First, dust tends to gather in the bottoms of the grooves. Second, the needle sits at the bottom of the groove and is weighted to press down on the bottom. With vertical deflection, the very act of deflecting the needle ends up re-etching the recording surface over time, rendering the record useless.
The modern side-to-side deflection was the obvious solution to the problem, since a sharp needle no longer needed to be pushed downwards under pressure against the actual recording surface, but could instead be deflected horizontally against much lower resistance.
Modern (stereo) records actually combine the two methods. They have dual slanted needle deflections. Each channel is encoded on one of those slanted axes, thus giving you stereo separation by virtue of the axes being roughly perpendicular to each other. And, of course, the bottom of the groove isn't a recording surface, so to a large extent, you still get the long-term viability of a horizontal deflection.
We now return you to your regularly scheduled patent discussion.
1 2 3 4 5 6 7 8 9 10
...
11 12 13 14 15 16 17 18 19 20
21 22 23 24 25 26 27 28 29 30
31 32 33 34 35 36 37 38 39 40
91 92 93 94 95 96 97 98 99
GW: It is not a rumor, it is a fact. That twelve-year-old boy does have weapons of mass deeeestruction. We must protect ourselves against this threat to our nation's security.
Boy: Yeesh. I knew Eminem sucked, but I destroyed that CD years ago.
GW: We have invaded Greenfield, TN. Most of the locals cheered as we toppled the bank sign, but were a little puzzled why the military vehicles went through their stop light and didn't slow down for the speed trap that followed. In any case, we have secured the household and are searching for the Eminem CDs of mass distraction.
Anchor: I thought you said there were weapons of mass destruction in his house.
GW: I will launch a full-scale investigation into the intelligence failure that broght on this terrible tragedy.....
Umm... no....
First, both are minor parts of the vehicle's functionality. Second, they're features that most people don't want or need. Third, they're features that the people who do want them really shouldn't be using in the first place, as they are harmful to your health (or at least the reliability of your data and/or your ability to concentrate on the road).
And we didn't abandon SCSI. Your firewire hard drive uses SBP2. That's SCSI commands sent over a serial tunnel. Want to know what your CD/DVD drive uses? ATAPI. That's SCSI packets encapsulated into ATA requests.
SCSI is like the force. "If you kill me, you will only make me stronger." And so, in effect, by abandoning the actual parallel SCSI physical transport (which really sucked, IMHO), the SCSI protocol (which doesn't) lives on in nearly every computer built in the past several years, PC or Mac.
But I digress.
No, my rights end where they injure others. My rights to watch a DVD on a Linux box do not injure the movie industry, therefore those rights are inalienable. Those who say otherwise are the greatest threat to the freedom of our country and our world. We must stand firm.
As Ray Bradbury put it in Fahrenheit 451:
T1 T2 xx xx Byte
03 C9 DF B0 00
T1: number of bytes in T2
T2: number of times to repeat "Byte"
Or, for stereo:
T1 T2 xx xx xx Byte
04 01 93 BF 60 00
I do believe that's 6-7 bytes shorter than the header on a MIDI file, thank you very much.
What I want to see is something that reads neruoelectric signatures. For the initial version, you'd think about your favorite food while leaning your head against a sensor pad. Of course, that could be captured, but that's just phase 1.
Phase 2 is to look at am image shown on the screen. When you sign up for an account, they'd do this once and store the neural impulses generated. From then on, they would show you the image and send the neural signature to the bank. The bank would compare the results and authorize the transaction, and would send a new image to display. You would see the second image, and the neural impulses generated by this second image would be sent back to the bank to store for the next time you tried to make a transaction.
The key requirements are that each transaction could require confirming the neural signature generated by any one or several of the prior images and the images sent for generating new signatures must be taken from a large enough database to get a high degree of variation. Finally, there must be expiration for old images, as one would expect one's reaction to an image to drift over time. Thus, an account unused in 90 days would be frozen until in-person verification could take place.
In such a case, in the unlikely event that someone were able to steal access to someone's account by taking enough prior neural signatures, they would still have to generate a new neural signature for the new image, which would mean that either it would be completely fictitious (which could probably be detected), a copy of some prior signature (which would definitely be detected and an alarm would sound), or would be the signature generated by the criminal, which could then be used as positive identification once that person gets caught.
Sound like fun? :-)