The Bell's should be very careful about who they pick on.
BS.net trims down google's bandwidth.
Google intentionally further limits bandwidth to BS.net, and posts a FAQ that the slowdown is due to BS.net. "Please contact BS.net to get this fixed!" Convenient phone numbers are provided.
Google sends a bill to BS.net for $10 / user / month to return the service to normal speed.
Google (et al) has the content that BS.net's customers want. If there is no content, there is no BS.net.
In theory, you're right. There's a lot of factors that go into how efficient a CPU and an instruction set archietecture are.
The ARM, for example, added a subset of their usual instructions (Thumb mode) that used 16-bit ops instead of the usual 32 bit. On a system with slow memory, this turned about to be about 30% faster, even though the number of operations increased.
However, I'm reasonably certain that if you were to design a CISC instruction set today, it would not resemble the x86 ISA. You could certainly come up with instructions that would be more compiler friendly, be easier to pipeline, etc.
CISC is a description of an instruction set. The x86 instruction set is a CISC instruction set. The processors are therefore CISC processors.
One of the arguments in favor of RISC was that it was easier to implement various performance improvements in the CPU on a RISC platform. However, you can implement a RISC instruction set without any of those optimizations (even without a cache, etc.) And with sufficient money, you can implement most of those things on a CISC processor as well.
And in a world with huge caches on the CPU, an extra 10% or 20% increase in the chip size to deal with CISC hasn't been a big enough penalty to impact the x86 ownership of the desktop.
It does not include total sales, where AMDs market share is significantly lower. e.g. this report excludes Dell entirely. Overall, they're somewhere around 25% of total shipments.
AMD is taking marketshare away from Intel, but they are still a much smaller player.
You mention in one of the replies that the company funding the research (Microsoft) had control over publication.
This implies that they could have funded 10 different studies (and perhaps did, even with your company), and only the one favorable to the company was published.
In the case of this study, if they'd funded doing the same study 10 different times, with 10 different small sets of administrators (posssibly being performed by the same research company, possibly being performed by 10 different independent researchers), if even one shows results favorable to Microsoft, that would be the only published results. That's like paying 10 different doctors to perform a drug study on 2 patients, and only publishing the results where the 2 patients recovered, failing to admit that the other 18 died.
Alternately, the might have funded you to research 5 different tasks, but only one favored Microsoft, and the others were buried.
If you aren't paid to publish the results, regardless of who they favor, your efforts are, in my opinion nearly useless. (Especially with the caveats introduced by small sample sizes of administrators.)
If you get a "free" or heavily discounted cell phone, you're stuck with a contract with large cancellation penalties. You're just amortizing the cost of the cell phone into your bill. (Personally, I wish they'd just offer financing on the cell phone and lower the price of service.)
And you might be surprised at how little some of those cell phones actually cost to make, since MSRP is pretty meaningless.
Re:Thats the whole point of the "puzzler"
on
Java Puzzlers
·
· Score: 1
I've worked on a C compiler for a DSP (Oak) that had sizeof(everything) == 16 bits.
But I think sizeof(char) is by definition "1" and everything else is a multiple thereof.
"(In response to 2) The volume of lost tax revenue on home sales alone would intitute a huge amount of maoney you will need to recoup. That for the times you actuall keep the money. If you sell a house, and roll the money immediatly(30 days) into a new purchase, you don't pay taxes. I don't know what the upper limit is."
There are no gains taxes on the sale of homes up to $250k / single or $500k / couple even if you don't reinvest it if you've lived in it for 2 out of the last 5 years. After that, it is usually at long term cap gains, or 15%.
Not much lost revenue here. (And all new homes will be taxed at the sales tax rate.)
"3. Every head of household will receive a monthly 'rebate' check" do you reallt think people making less the 15,000 dollars spend 6000 a year in taxes on 'basic necissities'? or does rent count into this?
The premise of this rebate is that no one should pay taxes on "poverty income". So they rebate the taxes you presumably will pay by multiplying the tax rate times the poverty level. This is regardless of your actual income, spending, or taxes paid.
If your income is below the poverty level, and you spend all of your income, you actually receive more money back in the rebate than you pay in taxes. If your income is at the poverty level, you pay no taxes. If your income is twice the poverty level, your taxes are effectively half the rate. etc.
This rebate makes the Fair Tax proposal at least someone progressive.
I think they'll need to add a customs/import tax to prevent rampant cross-border purchases, but otherwise the system seems no worse than what we have today.
But moments after it is implemented, they'll start putting special categories, exemptions, sin taxes, etc. in place, and it will go straight back to heck.
Right. My point is that the head of a team should perhaps be liable, not the individual team member. (And in general, if they followed appropriate practices and something bad happens, I assume their liability is somewhat limited. Just like a doctor in the event of a poor outcome as long as they didn't miss any data and performed enough tests...)
And while you can update software relatively cheaply, when a software bug costs your customer's 100,000 employees a day of downtime, your liability could easily run into the millions.
And for software, the comparison is somewhat like "Well, we bought a foundation designed by one company, pre-fab walls from another, had someone else wire it up, and then built our own roof". Who should get sued when it falls down? The guy who selected the parts? The guy who assembled them? The designer of the component that failed?
And there's a very big difference between holding a PE who owns a company liable and the draftsman or recent graduate engineer. It is the PE's responsibility to put a process into place that finds errors and gets them removed before a product leaves the building.
I'd have no trouble with putting some liability (to at least conform to best practices) on the head of a project team. But that person would have to have control of the design, coding, and testing of the product. (Or at least complete confidence in those other groups within the organization.)
But suggesting that the liability should fall on a developer who doesn't control her schedule, testing and validation of her product, integration of her work product with other developers, the target environment, etc. is absurd.
And many of the problems are caused by interactions of systems and/or misconfiguration of a system. In critical environments, you certify a specific combination of systems (including hardware and software versions), not that your product works in every possible combination. For example, aircraft are certified for a specific hardware and software combinations, and you generally aren't allowed to modify any of those without being recertified.
If you let anyone have access to reconfigure the system, or it is running with any different components (drivers, OS, patches, etc.), all bets for any kind of strict liability are off.
But at least requiring best (or at least very good) practices in software development to shield you from liability would be a great start.
But if you've complied with the procedures and a problem still sneaks through, you won't go to jail as I understand it.
If you intentionally sneak something in that causes the data to be misrepresented, you're liable.
If you put something in that is defective and didn't follow procedures, you're liable.
But even the shuttle software, for example, still has the occasional bug even though it is developed under some of the most stringent policies in the world and isn't an overly large application.
Well, I was shopping around for $14,000 in mulch for my homeowners association, clicked on a google ad for a local company, and they got a sale out of the deal.
$1.75 for an ad sounds like a pretty good deal for that company.
A number of the claims don't reference URLs. Claim #1 covers encoding any floating point coordinate system (not just lat-long), converting them to integers, and then converting an integer into any arbitrary alphabet (including decimal, hexadecimal, or possibly even base-256, i.e. bytes.)
And from their description down at the bottom:
[0083] It should be noted that embodiments of the present invention are generally beneficial in that they enable floating-point numbers to be converted into a compact ASCII string. Application within URL's is only one example where such compact strings are beneficial. Other application scenarios are within the scope of the present invention. The scheme for generating the compact strings can be applied in any place where one needs to transmit or store binary floating-point information, with the constraints that one has to limit some arbitrary character set and one wants to make it as compact as possible while preserving a desired level of accuracy. Relevant applications are not limited to those involving transmission over a network, but also include storage in data files, etc.
Except that the patent is for converting latitude and longitute into any base with any fixed precision.
The example is in base-30.
Base-10 is also covered by this patent.
It is not clear to me that using base-256 internal encoding in a computer is not covered under some of the individual claims. So if you convert lat-long to an integer, you might arguably violate this patent as well.
In the good old days, when the P-III was being replaced by the P-IV, you could roughly multiply the P-IV speed by 0.6 to get the comparable P-III speed. (Specifically, a 2Ghz P-IV was about the same as a 1.2Ghz P-III.)
So a P-IV at 3.6 Ghz, is roughly equal to a 2.16 Ghz P-III.
And guess what? The Pentium-M is a repackaged version of the Pentium-III core.
Maybe Intel and AMD and the rest of the world will start using something useful, like SPEC results to market their processors, instead of Hz ratings?
- Politicians bought votes in years past by adjusting the cost of living based on wage inflation, versus the previous (more reasonable) way of calculating it based on regular inflation.
Sorry, but wage indexing makes perfect sense for two reasons.
First, your contributions are based on your wage (and the cap should adjust with the same wage inflation number, although I don't know if it does.)
Second, Americans are much wealthier today than they were 40 years ago. SS has to reflect the higher standard of living that we all enjoy.
Social Security is designed to replace 40% of your pre-retirement income. If we switch to inflation indexing, and wages increase 100% more than inflation (due to productivity increases), then we'll be replacing 20% of people's pre-retirement income. And while you can possibly live on 40% (especially with no SS tax and lower taxes in general), you can't live on 20%.
(And for better or worse, many people depend on Social Security for the majority of their retirement income.)
As to your other items, how about a balanced budget amendment, rather than going after Social Security?
However, if you can't make the HELOC payments in the future (due to unemployment, for example), they can easily forclose on your house.
If you have excessive credit card debt and don't make payments, they can ruin your credit, but you keep the house.
And if you have good credit, you may be able to get "transfer your balance for 0%" or at least to a lower rate card without putting your house on the line.
It may be the answer in your case, but it is a bad idea for some people. (Especially those who will run up a new credit card balance...)
Re:I will help YOU get a JOB! (Programming puzzles
on
Programming Puzzles
·
· Score: 1
The hard version of #8 specifies that you can't use memory proportional to the size of the list, nor can you modify the contents of the list.
There's a slick way to handle this, which I'll write in the margin.
They poll a bunch of people about the election, but also get demographic data about them.
Then they weight the results for each candidate based on the size for that population.
This is the same kind of statistical weighting the census uses for some of it's surveys, but is legally prohibited from doing for apportionment. You can find more details in the discussions a few years ago about the census if you want more info.
I went to college with Phil Gordon and played bridge with him on occasion (http://www.philgordonpoker.com/index.html)
From his bio:
Phil came to the world of professional poker via the high-tech industry. After entering college as a National Merit Scholarship Finalist at age 15, Phil graduated from Georgia Institute of Technology at 20 with a degree in Computer Science. After a 2 year stint at Lockheed Missiles and Space, Phil joined Netsys Technologies, Inc. in Palo Alto, California as its lead software engineer and first employee. Three years later, Cisco systems purchased Netsys for 95 million dollars in stock and cash.
I hate to say it, but "average" is pretty low. Including the people who ring-up your burger, clean the floors, and greet you at Walmart, you're in a moderately elite group.
30% of the people in the US don't graduate from High School.
The average score on the Math SAT is 519 and that is only of the people who even took the SAT. I suspect the vast majority of the programmers beat that number.
I'll admit, programmers don't necessarily know anything about any given domain (including selling burgers or playing poker) and sometimes won't recognize that fact. (So they can act like real jerks to people who are domain expoerts.) This applies to poker, as a domain expert who isn't a math whiz can easily beat someone who is "smart".
And programmers aren't necessarily smarter than any other college educated, white-collar co-worker, although their smarts are more likely to be focused on math and logic while other professions may have other kinds of intelligence. (Sales, for example, requires "smarts" of a completely different nature than programming. Which is why I will never own my own company.)
The "smarts" required to play poker certainly include some of the skills that are required of programmers. But to master it, you have to have other skills that may not be very common among programmers. I'm not sure if I'd take a top-notch salesman or a top-notch programmer in a poker competition. (And it would probably depend on the type of poker, as some are more odds focussed than others.)
He misses the next logical chain in the arguement.
To apply for COPYRIGHT registration on code, you should have to submit the source code. It would still be copyrighted, so people can't create derivitive works from it legally. But shipping copy-protected executable code certainly doesn't promote the progress of science and the useful arts. Having the source code available does.
Sorry, but that's not a submarine patent. A submarine patent is a specific "term of art" in the intellectual property arena.
A submarine patent referred to using tricks involving amending and revising patents to keep them from being granted for a long time. In some cases, these techniques kept the patent outstanding for decades. Since the patent wasn't granted, it wasn't visible in the patent system via searches, etc.
Then, when the patent was being violated by numerous people (using a laser in medicine!), they stoped the tricks and the patent office granted the patent. And then the owner sued everyone.
These techniques have mostly or completely been eliminated by changes in the law and filing rules.
This patent was legitimately granted, was in the patent archives, and any engineer or project manager on a product involving an LCD could have done a simple search and discovered it.
And being a physical thing, it is much easier to determine if you were violating this patent. The problem with software patents are that a project manager or an IP lawyer CAN'T easily do a search on some keywords and see if the software violates the patent. How does the IP lawyer know that you're using MD5 checksums on some portion of your data, and someone might have a patent on that?
Slashdot Mirror
The Bell's should be very careful about who they pick on.
BS.net trims down google's bandwidth.
Google intentionally further limits bandwidth to BS.net, and posts a FAQ that the slowdown is due to BS.net. "Please contact BS.net to get this fixed!" Convenient phone numbers are provided.
Google sends a bill to BS.net for $10 / user / month to return the service to normal speed.
Google (et al) has the content that BS.net's customers want. If there is no content, there is no BS.net.
In theory, you're right. There's a lot of factors that go into how efficient a CPU and an instruction set archietecture are.
The ARM, for example, added a subset of their usual instructions (Thumb mode) that used 16-bit ops instead of the usual 32 bit. On a system with slow memory, this turned about to be about 30% faster, even though the number of operations increased.
However, I'm reasonably certain that if you were to design a CISC instruction set today, it would not resemble the x86 ISA. You could certainly come up with instructions that would be more compiler friendly, be easier to pipeline, etc.
CISC is a description of an instruction set. The x86 instruction set is a CISC instruction set. The processors are therefore CISC processors.
One of the arguments in favor of RISC was that it was easier to implement various performance improvements in the CPU on a RISC platform. However, you can implement a RISC instruction set without any of those optimizations (even without a cache, etc.) And with sufficient money, you can implement most of those things on a CISC processor as well.
And in a world with huge caches on the CPU, an extra 10% or 20% increase in the chip size to deal with CISC hasn't been a big enough penalty to impact the x86 ownership of the desktop.
Note that this only applies to retail sales.
It does not include total sales, where AMDs market share is significantly lower. e.g. this report excludes Dell entirely. Overall, they're somewhere around 25% of total shipments.
AMD is taking marketshare away from Intel, but they are still a much smaller player.
You mention in one of the replies that the company funding the research (Microsoft) had control over publication.
This implies that they could have funded 10 different studies (and perhaps did, even with your company), and only the one favorable to the company was published.
In the case of this study, if they'd funded doing the same study 10 different times, with 10 different small sets of administrators (posssibly being performed by the same research company, possibly being performed by 10 different independent researchers), if even one shows results favorable to Microsoft, that would be the only published results. That's like paying 10 different doctors to perform a drug study on 2 patients, and only publishing the results where the 2 patients recovered, failing to admit that the other 18 died.
Alternately, the might have funded you to research 5 different tasks, but only one favored Microsoft, and the others were buried.
If you aren't paid to publish the results, regardless of who they favor, your efforts are, in my opinion nearly useless. (Especially with the caveats introduced by small sample sizes of administrators.)
Well, not quite.
If you get a "free" or heavily discounted cell phone, you're stuck with a contract with large cancellation penalties. You're just amortizing the cost of the cell phone into your bill. (Personally, I wish they'd just offer financing on the cell phone and lower the price of service.)
And you might be surprised at how little some of those cell phones actually cost to make, since MSRP is pretty meaningless.
I've worked on a C compiler for a DSP (Oak) that had sizeof(everything) == 16 bits.
But I think sizeof(char) is by definition "1" and everything else is a multiple thereof.
"(In response to 2) The volume of lost tax revenue on home sales alone would intitute a huge amount of maoney you will need to recoup. That for the times you actuall keep the money. If you sell a house, and roll the money immediatly(30 days) into a new purchase, you don't pay taxes. I don't know what the upper limit is."
There are no gains taxes on the sale of homes up to $250k / single or $500k / couple even if you don't reinvest it if you've lived in it for 2 out of the last 5 years. After that, it is usually at long term cap gains, or 15%.
Not much lost revenue here. (And all new homes will be taxed at the sales tax rate.)
"3. Every head of household will receive a monthly 'rebate' check" do you reallt think people making less the 15,000 dollars spend 6000 a year in taxes on 'basic necissities'? or does rent count into this?
The premise of this rebate is that no one should pay taxes on "poverty income". So they rebate the taxes you presumably will pay by multiplying the tax rate times the poverty level. This is regardless of your actual income, spending, or taxes paid.
If your income is below the poverty level, and you spend all of your income, you actually receive more money back in the rebate than you pay in taxes. If your income is at the poverty level, you pay no taxes. If your income is twice the poverty level, your taxes are effectively half the rate. etc.
This rebate makes the Fair Tax proposal at least someone progressive.
I think they'll need to add a customs/import tax to prevent rampant cross-border purchases, but otherwise the system seems no worse than what we have today.
But moments after it is implemented, they'll start putting special categories, exemptions, sin taxes, etc. in place, and it will go straight back to heck.
Right. My point is that the head of a team should perhaps be liable, not the individual team member. (And in general, if they followed appropriate practices and something bad happens, I assume their liability is somewhat limited. Just like a doctor in the event of a poor outcome as long as they didn't miss any data and performed enough tests...)
And while you can update software relatively cheaply, when a software bug costs your customer's 100,000 employees a day of downtime, your liability could easily run into the millions.
And for software, the comparison is somewhat like "Well, we bought a foundation designed by one company, pre-fab walls from another, had someone else wire it up, and then built our own roof". Who should get sued when it falls down? The guy who selected the parts? The guy who assembled them? The designer of the component that failed?
And there's a very big difference between holding a PE who owns a company liable and the draftsman or recent graduate engineer. It is the PE's responsibility to put a process into place that finds errors and gets them removed before a product leaves the building.
I'd have no trouble with putting some liability (to at least conform to best practices) on the head of a project team. But that person would have to have control of the design, coding, and testing of the product. (Or at least complete confidence in those other groups within the organization.)
But suggesting that the liability should fall on a developer who doesn't control her schedule, testing and validation of her product, integration of her work product with other developers, the target environment, etc. is absurd.
And many of the problems are caused by interactions of systems and/or misconfiguration of a system. In critical environments, you certify a specific combination of systems (including hardware and software versions), not that your product works in every possible combination. For example, aircraft are certified for a specific hardware and software combinations, and you generally aren't allowed to modify any of those without being recertified.
If you let anyone have access to reconfigure the system, or it is running with any different components (drivers, OS, patches, etc.), all bets for any kind of strict liability are off.
But at least requiring best (or at least very good) practices in software development to shield you from liability would be a great start.
But if you've complied with the procedures and a problem still sneaks through, you won't go to jail as I understand it.
If you intentionally sneak something in that causes the data to be misrepresented, you're liable.
If you put something in that is defective and didn't follow procedures, you're liable.
But even the shuttle software, for example, still has the occasional bug even though it is developed under some of the most stringent policies in the world and isn't an overly large application.
Well, I was shopping around for $14,000 in mulch for my homeowners association, clicked on a google ad for a local company, and they got a sale out of the deal.
$1.75 for an ad sounds like a pretty good deal for that company.
Something that is 8 times as long, as well as 8 times in each of the other two dimensions, is 512 times more massive.
So a crew of 280k is equal to an aircraft carrier with a crew of 560.
No big deal at all.
A number of the claims don't reference URLs. Claim #1 covers encoding any floating point coordinate system (not just lat-long), converting them to integers, and then converting an integer into any arbitrary alphabet (including decimal, hexadecimal, or possibly even base-256, i.e. bytes.)
And from their description down at the bottom:
[0083] It should be noted that embodiments of the present invention are generally beneficial in that they enable floating-point numbers to be converted into a compact ASCII string. Application within URL's is only one example where such compact strings are beneficial. Other application scenarios are within the scope of the present invention. The scheme for generating the compact strings can be applied in any place where one needs to transmit or store binary floating-point information, with the constraints that one has to limit some arbitrary character set and one wants to make it as compact as possible while preserving a desired level of accuracy. Relevant applications are not limited to those involving transmission over a network, but also include storage in data files, etc.
Note that there are claims (1, 8, and 16-21) that don't limit themselves to URLs.
Any software that uses such an encoding internally would violate this patent.
Except that the patent is for converting latitude and longitute into any base with any fixed precision.
The example is in base-30.
Base-10 is also covered by this patent.
It is not clear to me that using base-256 internal encoding in a computer is not covered under some of the individual claims. So if you convert lat-long to an integer, you might arguably violate this patent as well.
In the good old days, when the P-III was being replaced by the P-IV, you could roughly multiply the P-IV speed by 0.6 to get the comparable P-III speed. (Specifically, a 2Ghz P-IV was about the same as a 1.2Ghz P-III.)
So a P-IV at 3.6 Ghz, is roughly equal to a 2.16 Ghz P-III.
And guess what? The Pentium-M is a repackaged version of the Pentium-III core.
Maybe Intel and AMD and the rest of the world will start using something useful, like SPEC results to market their processors, instead of Hz ratings?
- Politicians bought votes in years past by adjusting the cost of living based on wage inflation, versus the previous (more reasonable) way of calculating it based on regular inflation.
Sorry, but wage indexing makes perfect sense for two reasons.
First, your contributions are based on your wage (and the cap should adjust with the same wage inflation number, although I don't know if it does.)
Second, Americans are much wealthier today than they were 40 years ago. SS has to reflect the higher standard of living that we all enjoy.
Social Security is designed to replace 40% of your pre-retirement income. If we switch to inflation indexing, and wages increase 100% more than inflation (due to productivity increases), then we'll be replacing 20% of people's pre-retirement income. And while you can possibly live on 40% (especially with no SS tax and lower taxes in general), you can't live on 20%.
(And for better or worse, many people depend on Social Security for the majority of their retirement income.)
As to your other items, how about a balanced budget amendment, rather than going after Social Security?
However, if you can't make the HELOC payments in the future (due to unemployment, for example), they can easily forclose on your house.
If you have excessive credit card debt and don't make payments, they can ruin your credit, but you keep the house.
And if you have good credit, you may be able to get "transfer your balance for 0%" or at least to a lower rate card without putting your house on the line.
It may be the answer in your case, but it is a bad idea for some people. (Especially those who will run up a new credit card balance...)
The hard version of #8 specifies that you can't use memory proportional to the size of the list, nor can you modify the contents of the list.
There's a slick way to handle this, which I'll write in the margin.
You're all wrong about this.
They poll a bunch of people about the election, but also get demographic data about them.
Then they weight the results for each candidate based on the size for that population.
This is the same kind of statistical weighting the census uses for some of it's surveys, but is legally prohibited from doing for apportionment. You can find more details in the discussions a few years ago about the census if you want more info.
I went to college with Phil Gordon and played bridge with him on occasion (http://www.philgordonpoker.com/index.html)
From his bio:
Phil came to the world of professional poker via the high-tech industry. After entering college as a National Merit Scholarship Finalist at age 15, Phil graduated from Georgia Institute of Technology at 20 with a degree in Computer Science. After a 2 year stint at Lockheed Missiles and Space, Phil joined Netsys Technologies, Inc. in Palo Alto, California as its lead software engineer and first employee. Three years later, Cisco systems purchased Netsys for 95 million dollars in stock and cash.
I hate to say it, but "average" is pretty low. Including the people who ring-up your burger, clean the floors, and greet you at Walmart, you're in a moderately elite group.
30% of the people in the US don't graduate from High School.
The average score on the Math SAT is 519 and that is only of the people who even took the SAT. I suspect the vast majority of the programmers beat that number.
I'll admit, programmers don't necessarily know anything about any given domain (including selling burgers or playing poker) and sometimes won't recognize that fact. (So they can act like real jerks to people who are domain expoerts.) This applies to poker, as a domain expert who isn't a math whiz can easily beat someone who is "smart".
And programmers aren't necessarily smarter than any other college educated, white-collar co-worker, although their smarts are more likely to be focused on math and logic while other professions may have other kinds of intelligence. (Sales, for example, requires "smarts" of a completely different nature than programming. Which is why I will never own my own company.)
The "smarts" required to play poker certainly include some of the skills that are required of programmers. But to master it, you have to have other skills that may not be very common among programmers. I'm not sure if I'd take a top-notch salesman or a top-notch programmer in a poker competition. (And it would probably depend on the type of poker, as some are more odds focussed than others.)
He misses the next logical chain in the arguement.
To apply for COPYRIGHT registration on code, you should have to submit the source code. It would still be copyrighted, so people can't create derivitive works from it legally. But shipping copy-protected executable code certainly doesn't promote the progress of science and the useful arts. Having the source code available does.
Sorry, but that's not a submarine patent. A submarine patent is a specific "term of art" in the intellectual property arena.
A submarine patent referred to using tricks involving amending and revising patents to keep them from being granted for a long time. In some cases, these techniques kept the patent outstanding for decades. Since the patent wasn't granted, it wasn't visible in the patent system via searches, etc.
Then, when the patent was being violated by numerous people (using a laser in medicine!), they stoped the tricks and the patent office granted the patent. And then the owner sued everyone.
These techniques have mostly or completely been eliminated by changes in the law and filing rules.
This patent was legitimately granted, was in the patent archives, and any engineer or project manager on a product involving an LCD could have done a simple search and discovered it.
And being a physical thing, it is much easier to determine if you were violating this patent. The problem with software patents are that a project manager or an IP lawyer CAN'T easily do a search on some keywords and see if the software violates the patent. How does the IP lawyer know that you're using MD5 checksums on some portion of your data, and someone might have a patent on that?