You're right, but the point is that there are limits on rights with regard to property. You own the land your housse is on, but that doesn't permit you to dump tons of toxic chemicals on your lawn because they seep down and screw everybody. The extent to which you own the sky above your house is limited by your national laws and ceases completely at ~60 miles above sea level. Even what you can use your house for is limited by zoning restrictions.
It sounds like I'm making a case for the ever-expanding IP rights, but I'm not. The extent of intellectual property used to be limited by law (both statue and precedent). The publisher of a book did not have the right to limit resale or many reproductions - "fair use". When I sell my house, I can't add a clause stating the purchaser is entitled to dump toxic chemicals, because laws limiting my control over the product supersede the contract. For the record, I can't insert a clause forbidding resale of the house. Similar restrictions apply to IP.
This comment makes an important point: The only way we can learn about security is by studying security problems.
In a adversarial environment like computer security, you can't be any good if you only understand one side of the game. Even if you are a "good guy" you must understand how to be a "bad guy" to be worth anything. It's impossible to write antivirus software or truly understand viruses without looking at the code for them. It's impossible to develop a good cryptosystem if you don't have a detailed understanding why previous systems are bad.
Many people don't quite get how a buffer overflow works (or why they should check buffer limits in their code) until someone describes how the attack works in painstaking detail. This person will now check their buffer limits, but they also know how to write a buffer overflow attack if they are maliciously inclined - a net gain in my book.
In more general terms, the Army trains people who will never do anything except defend their position in how to attack. Law schools don't break criminal law into classes on prosecution and defense, and police study methods used by criminals. But hey, Microsoft says software is too complex for this traditional process of learning how to defend.
i.e.: I would like to use the linux RPM system for software deployment, but need to make a rational decision to go with the standard Solaris Package system since my 20 system admins refuse to learn anything new. Again, these are real world realities beyond the academic/hobbyist/startup environment.
Then you're not doing a good job explaining the benefits of RPM to them, or you have people who will fossilize when their preffered solution for a particular task becomes obsolete. I'm always amazed at people who let these considerations completely outweigh technical issues. At the point where solely internal politics forces you to choice something you believe is inferior, you have a problem.
"Like previous versions of NT, Win2K comes in several product levels, this time: Professional, Server, Advanced server, and Datacenter Server. The difference between all these versions are minor however, with the same excutable binary being used for all versions. When the system is installed, the product type is recorded in the registry..., the default parameters are tuned differently on Professonial to favor interactive work over batch work, althogh these can be changed if desired."
-Andrew Tannebaum, Modern Operating Systems, 2000. Probably one of the smartest and most objective people in the world when it comes to OSes, and he's not too fond of Linux.
Funny, couldn't agree more. I'm a Unix guy who donates my time to adminster a win98/2k network for a nonprofit. I've found that I can do anything for them I need to, given enough time and books (which is my nominal compensation, a stack of MCSE books) - it's not a question of knowing exactly "what", but of understanding how to figure out "what".
I wonder how many can actually name their congressman.
Most people can't. That's why the ACLU, NRA, Christian Coalition, and so forth are more than happy to tell you. The stuff I get in the mail from some of these people is constantly telling me who I should contact since they assume I'm too lazy to find out for myself. That's what these groups are for - to make it easy for their members to be heard.
You get the frickin security advisory email, you run up2date. RedHat, unlike MS, is not afraid to email customers running servers and say, "You got a hole. Go fix it!" I've never had one Linux patch regress another.
Re:Fixating on the OS is the first mistake.
on
Linux on the Desktop
·
· Score: 2
Cloning, by definition, is doomed to fail because it is a game of catch-up. A better approach is to think "What exactly do people _need_ to do?"
I disagree. Cloning with modifications is the core of this business. IE was a clone of Netscape, Word was a clone of WordPerfect, Windows was a clone of MacOS, Athlons are Pentium clones, and every Unix is a clone of the original. There are two types of clones - the ones that are a subset of their predecessors and those that are a superset. Microsoft excels at making the first type - Word used to include all the WordPerfect shortcuts while adding its own. Some Linux apps clone well, others don't. Mozilla has some neat features IE doesn't (better cookie management, pop-up control, etc.) while trying to replicate the core behavior. StarOffice is the opposite - it's a pure clone, afraid to strike out on its own.
This is the really sad part of the windows monopoly - Unix is far easier to program for than windows if you know what you're doing - and good programmers are drawn to the power of unix. The development tools are simply better thanks to Unix's long tradition of being the OS of the developers, by the developers, for the developers. But it's the problem that all platforms suffer from - "We'll build apps when people use it, we'll use it when there's software for it...."
I'm always amazed when I see kiosk-type systems (run only one specialized app) that are running Windows. I always wonder who in God's name approved paying for win2k licenses, win2k capable hardware, and windows development costs for a computer that only runs an app developed in-house.
Compared to the other server-grade OS (win2k) this is more than reasonable - 2k server runs $800 off the shelf, I think. (never bought it that way) You have additional licensing fees on top of that. And you get some pretty good support from RH if you pay.
You can get the distro images for free from their ftp site if you have the bandwith. So the price isn't that bad.
BNTW - if you want to go after a real monopoly, consider the US Postal Service or the Social Security system.
Huh, neither one of them has doubled their price for the services they provide in the last decade without losing market share. Probably cause the evil government regulators won't let them.
Business is pretty cutthroat, but in all cases the competition is directed at who can make the consumer happiest. MS has suceeded here better than anyone else. That's why they have a dominant market position.
Cutthroat is one thing - breaking contracts (JVM), not paying royalties(Spyglass), using secret contracts to deny your competiton market entry (OEM licenses), losing money to drive your competitiors out of business (Netscape) - the list goes on. These aren't cutthroat, they're crooked. {sarcasm} If the mob gives me good protection, they've made me a happy customer, it doesn't matter that they had to break the law to do it.{/sarcasm} I'm so sick of this nonsense.
As Time points out, the govt gave something and got nothing. Normally, if two parties can't reach an agreement, one party doesn't just say "Oh, we'll moderate our position without a similiar concession from the other side". The Bush adminstration is signalling that these crooks won't be punished - plain and simple. They'll get off real easy, another "consent decree" to ignore.
I think using Linux is an asset for a company in terms of personnel costs. People who love computing, love solving problems, and love working hard are attracted to the open source philosophy by nature. Working with Linux has what economists call a "psychic reward" - people do it partly because it's a reward in and of itself and will therefore do it for somewhat less money than they would otherwise. This is one of the reasons teachers are paid relatively poorly - they do it because they want to. A friend of mine took a small pay cut to work in a Unix environment because "it's more fun and what's five grand when you still make 75K?" (Exact quote, it was one of those memorable statements).
You can do this with anything, it's one of the oldest novel business models. People don't realize how many companies sell their most valuable product at a loss: The Gillette Mach3 razor cost many millions of dollars to develop, is sold well below cost, and the replacement blades make up the difference. Video game consoles, ditto, profit is in game licenses. On and on....
And support is a far more lucrative business than general-purpose software. Much of the cost of support is marginal. If you provide h engineer-hours of support, and charge r dollars/hr, pay the support team c dollar/hr, your profit is h(r-c) dollars, which is always positive. (This assumes no overhead costs, which is why small many Linux companies fail, because the overhead costs overwhelm them.)
Selling software at a flat rate is a tremendously risky business. You pay $d to develop it, and sell n copies at cost c. Your profit is d - cn dollars, which can be negative. One dud product and you're out of business. This is why there is only one large company on the face of the planet that generates a substantial amount of revenue from software - Microsoft. For them, the risk is mitigated to an absurd extent because n will be very high no matter what because people are forced to buy their products. And this is why MS places such a low priority on quality - they don't get more money for high quality software. Making Windows more stable increases costs but doesn't affect revenue substantially. This is why in 2001 my Windows box still locks up all the time - because implementing full memory protection and a host of other features common to every other OS wouldn't earn them another cent. It helps a lot that I can't return their software when it fails to meet expectations. (This is why each successive version of Windows is flashier, but not necessarily better from a technical standpoint.)
It makes far more sense to "sell" free software, have little general-purpose development, and provide services (including specialized development), which offer better and more predictable profits.
What happens if you are a customer who was sold a system with one of those "other" operating systems. You'd scream bloody murder if the OS running your systems was going to be trashed in favor of something else.
If that happens you can be sure that Microsoft is going to try and sweep in to pick through the carnage...
I don't think the BizWeek guy is suggesting they drop support for their Unices tommorrow. (BizWeek cheering on Linux??? What is this world coming to?)
They're all Unix flavors, all conforming to POSIX to some degree or another. If your inhouse developers followed the POSIX standard closely, (which they should have, as that gives you the flexibility to switch unix flavors), tranistioning to Linux is an order of magnitude easier than switching to NT/2K. And training an HPUX staff to use Linux is substantially easier than teaching them the MS way. Running to Microsoft would be even more difficult.
They're startups, they're supposed to lose money initially. Startups succeed when you go from losing money to breaking even to turning a profit. They fail when they lose more and more money. Redhat is breaking even, which means their revenue is rising faster than their costs and we can assume they'll proceed on to profitability. Microsoft is making money, but making less every year. Their costs are rising faster than their revenue. Factor in their impending breakup, dozens of lawsuits getting thrown at them, and a few other things, and their long-term outlook isn't very good.
Why not create a govt agency or a private company that certifies proprietary code as GPL-free? You give them the source code, they compare it to the source code of similar GPL products, most likely using automation and a good string matching algorithm, and if the algorithm finds no nontrival matches in the code, then the program is good.
Of course, the code-checking engine would have to be open-source so companies could be sure that it didn't steal code:)
You can make a tremendous amount of money even if you sell a highly visible product at a loss - sell a profit-making complement too.
Gllette spent millions developing the Mach 3 razor and sells each one at less than 20% of the manufacturing price. They make money selling replacement blades above cost. It works, its worked for decades.
RedHat sells RH Linux below cost ($0). But compared to Windows, it's cheap to develop - you take other people's innovations, add a few, and package it in a useable format. Then you sell services above cost - no OS is so easy to use that there are no support issues. RH Linux is a loss-leader to get people to establish a relationship with the company. It would make sense that if I need high-level support for RH Linux, I would turn to RedHat, since they know the product better than anyone esle. IBM has the same business model now - they want to sell you a complete Linux solution - they've largely given up on PCs to be a services company.
This sounds stupid, but service/support is a better business than proprietary software. All of your costs are marginal - if you sell n hours of support for $r/hr and you pay $e/hr to the support guy, you make n(e-r) dollars, which is always a positive number if e>r. Software costs d dollars to develop, where d is a very fucking large number. If n people buy it a cost c, you make (n*c - d). This can be negative if not enough people buy your product. This is a gross oversimplification, there's overhead in support, and some neglible marginal costs in selling software, but this is why there are so many little support shops - you don't need to invest a tremendous amount of cash upfront.
Microsoft is one of the few large companies on the face of the earth whose only product is proprietary consumer software. Oracle, IBM, Apple, you name it, all sell something else besides software - consumer software is a ridiculously risky business. They've managed to succeed by becoming a monopoly and essentially forcing people to buy their products. They used illegal pricing games to drive the competition out of business - offering "competitive upgrades" on competitetors products, giving away IE, breaking cross-platform compatibility and so forth. They can't use the same strategy against Linux because you can't undercut it on price, so they've resorted to their other famous tactic, FUD.
No, Felten makes very clear those provisions would only apply if they took the cash prize, which they politely declined. And they're not "attacking content", they're explaining watermarking schemes a second-year math undergrad would understand.
The paper contains pseudocode-like descriptions of how to retain your rights. It's a "device" under the DMCA
That's the whole problem, if source code, a very precise and computer-centric format for describing a process is a device, translating the "device" into a less precise and more human-centric format means it's still a "device".
It's easy. If you work on a project that does x during the day, any freetime work should be reasonably unrelated to x. It's safer for both parties that way, there's no chance the company will end up with GPLed code in its product, and there's no way the GPL project will end up with code that belongs to somebody else. Furthermore, it gives you a broader experience and a chance to do something different.
I think the submittor is talking about the annoying tendency of comapnies to claim off-hours project as their IP. If I write a program on my own time, I should be able to claim it as my intellectual property, and slap the GPL on it. It's not like I'm taking a second job, I'm giving away my work.
But the land-grab mentality about intellectual property causes problems. If I use a company laptop, there might be grounds for them to claim they own my project. Same if I use a company email account on a project mailing list, or if post a quick note about a bug from work.
Something like this makes sense:
"XYZ corp disclaims all rights to Joe Coder's project Foo. Use of corporate assets such as laptops, workstations, or other systems does not grant XYZ corp any rights to Joe Coder's project."
It's pretty reasonable, and assuming Joe is releasing under the GPL there's no chance he'll ever make money off the product himself, so he doesn't have a financial incentive to neglect his real job for the Foo project.
You're right, but this illustrates the problem with the USPTO today - they can't keep up. By the time a patent is granted, 100 other people have idependently created the same technology, and now they're screwed because the stealth patent beats them up.
On the other hand, I think it's pretty clear there is a tremendous amount of "prior art" ie pre-1996 here, so there seems to be no ground for a patent either way.
....wherein the step of limiting is a step of determining which avatars are closest in the graphical space to the local avatar.
The patent uses the word "avatar" over 30 times to describe a person's representation in a 3-D environment. Funny, Stephenson called them that in SnowCrash. Wait, he basically described the thing being patented. Oops.
I think the/. crowd is most pissed off by the fact that Intel's technical decisions are made partly with a view to marketing. The "Internet architecture" BS a couple years back Intel's own have engineers bitched there is no way to optimize a CPU for Internet - net bandwith will always be the limiting factor, not the CPU. But we still saw all those idiotic "Intel gives you power for the Internet" commercials. Now we've got a chip that appears to be grossly overpriced for its performance, released now only because of "MHz marketing". I think the P4 has an good architecture which scales like crazy, but why was it released before it reached 2 GHz and that scaleablity was an asset and not a liability? The UID for CPUs and didn't help Intel's image either - that pissed off the general public too.
That said, assuming the P4 continues to play catch-up well, I'll buy one if it offers more value for my money.
Slashdot Mirror
You're right, but the point is that there are limits on rights with regard to property. You own the land your housse is on, but that doesn't permit you to dump tons of toxic chemicals on your lawn because they seep down and screw everybody. The extent to which you own the sky above your house is limited by your national laws and ceases completely at ~60 miles above sea level. Even what you can use your house for is limited by zoning restrictions.
It sounds like I'm making a case for the ever-expanding IP rights, but I'm not. The extent of intellectual property used to be limited by law (both statue and precedent). The publisher of a book did not have the right to limit resale or many reproductions - "fair use". When I sell my house, I can't add a clause stating the purchaser is entitled to dump toxic chemicals, because laws limiting my control over the product supersede the contract. For the record, I can't insert a clause forbidding resale of the house. Similar restrictions apply to IP.
This comment makes an important point: The only way we can learn about security is by studying security problems.
In a adversarial environment like computer security, you can't be any good if you only understand one side of the game. Even if you are a "good guy" you must understand how to be a "bad guy" to be worth anything. It's impossible to write antivirus software or truly understand viruses without looking at the code for them. It's impossible to develop a good cryptosystem if you don't have a detailed understanding why previous systems are bad.
Many people don't quite get how a buffer overflow works (or why they should check buffer limits in their code) until someone describes how the attack works in painstaking detail. This person will now check their buffer limits, but they also know how to write a buffer overflow attack if they are maliciously inclined - a net gain in my book.
In more general terms, the Army trains people who will never do anything except defend their position in how to attack. Law schools don't break criminal law into classes on prosecution and defense, and police study methods used by criminals. But hey, Microsoft says software is too complex for this traditional process of learning how to defend.
i.e.: I would like to use the linux RPM system for software deployment, but need to make a rational decision to go with the standard Solaris Package system since my 20 system admins refuse to learn anything new. Again, these are real world realities beyond the academic/hobbyist/startup environment.
Then you're not doing a good job explaining the benefits of RPM to them, or you have people who will fossilize when their preffered solution for a particular task becomes obsolete. I'm always amazed at people who let these considerations completely outweigh technical issues. At the point where solely internal politics forces you to choice something you believe is inferior, you have a problem.
Amazing what you can find in textbooks:
"Like previous versions of NT, Win2K comes in several product levels, this time: Professional, Server, Advanced server, and Datacenter Server. The difference between all these versions are minor however, with the same excutable binary being used for all versions. When the system is installed, the product type is recorded in the registry..., the default parameters are tuned differently on Professonial to favor interactive work over batch work, althogh these can be changed if desired."
-Andrew Tannebaum, Modern Operating Systems, 2000. Probably one of the smartest and most objective people in the world when it comes to OSes, and he's not too fond of Linux.
Funny, couldn't agree more. I'm a Unix guy who donates my time to adminster a win98/2k network for a nonprofit. I've found that I can do anything for them I need to, given enough time and books (which is my nominal compensation, a stack of MCSE books) - it's not a question of knowing exactly "what", but of understanding how to figure out "what".
I wonder how many can actually name their congressman.
Most people can't. That's why the ACLU, NRA, Christian Coalition, and so forth are more than happy to tell you. The stuff I get in the mail from some of these people is constantly telling me who I should contact since they assume I'm too lazy to find out for myself. That's what these groups are for - to make it easy for their members to be heard.
You get the frickin security advisory email, you run up2date. RedHat, unlike MS, is not afraid to email customers running servers and say, "You got a hole. Go fix it!" I've never had one Linux patch regress another.
Cloning, by definition, is doomed to fail because it is a game of catch-up. A better approach is to think "What exactly do people _need_ to do?"
I disagree. Cloning with modifications is the core of this business. IE was a clone of Netscape, Word was a clone of WordPerfect, Windows was a clone of MacOS, Athlons are Pentium clones, and every Unix is a clone of the original. There are two types of clones - the ones that are a subset of their predecessors and those that are a superset. Microsoft excels at making the first type - Word used to include all the WordPerfect shortcuts while adding its own. Some Linux apps clone well, others don't. Mozilla has some neat features IE doesn't (better cookie management, pop-up control, etc.) while trying to replicate the core behavior. StarOffice is the opposite - it's a pure clone, afraid to strike out on its own.
Can always use WINE instead.
This is the really sad part of the windows monopoly - Unix is far easier to program for than windows if you know what you're doing - and good programmers are drawn to the power of unix. The development tools are simply better thanks to Unix's long tradition of being the OS of the developers, by the developers, for the developers. But it's the problem that all platforms suffer from - "We'll build apps when people use it, we'll use it when there's software for it...."
I'm always amazed when I see kiosk-type systems (run only one specialized app) that are running Windows. I always wonder who in God's name approved paying for win2k licenses, win2k capable hardware, and windows development costs for a computer that only runs an app developed in-house.
Compared to the other server-grade OS (win2k) this is more than reasonable - 2k server runs $800 off the shelf, I think. (never bought it that way) You have additional licensing fees on top of that. And you get some pretty good support from RH if you pay.
You can get the distro images for free from their ftp site if you have the bandwith. So the price isn't that bad.
BNTW - if you want to go after a real monopoly, consider the US Postal Service or the Social Security system.
Huh, neither one of them has doubled their price for the services they provide in the last decade without losing market share. Probably cause the evil government regulators won't let them.
Business is pretty cutthroat, but in all cases the competition is directed at who can make the consumer happiest. MS has suceeded here better than anyone else. That's why they have a dominant market position.
Cutthroat is one thing - breaking contracts (JVM), not paying royalties(Spyglass), using secret contracts to deny your competiton market entry (OEM licenses), losing money to drive your competitiors out of business (Netscape) - the list goes on. These aren't cutthroat, they're crooked. {sarcasm} If the mob gives me good protection, they've made me a happy customer, it doesn't matter that they had to break the law to do it.{/sarcasm} I'm so sick of this nonsense.
As Time points out, the govt gave something and got nothing. Normally, if two parties can't reach an agreement, one party doesn't just say "Oh, we'll moderate our position without a similiar concession from the other side". The Bush adminstration is signalling that these crooks won't be punished - plain and simple. They'll get off real easy, another "consent decree" to ignore.
I think using Linux is an asset for a company in terms of personnel costs. People who love computing, love solving problems, and love working hard are attracted to the open source philosophy by nature. Working with Linux has what economists call a "psychic reward" - people do it partly because it's a reward in and of itself and will therefore do it for somewhat less money than they would otherwise. This is one of the reasons teachers are paid relatively poorly - they do it because they want to. A friend of mine took a small pay cut to work in a Unix environment because "it's more fun and what's five grand when you still make 75K?" (Exact quote, it was one of those memorable statements).
You can do this with anything, it's one of the oldest novel business models. People don't realize how many companies sell their most valuable product at a loss: The Gillette Mach3 razor cost many millions of dollars to develop, is sold well below cost, and the replacement blades make up the difference. Video game consoles, ditto, profit is in game licenses. On and on....
And support is a far more lucrative business than general-purpose software. Much of the cost of support is marginal. If you provide h engineer-hours of support, and charge r dollars/hr, pay the support team c dollar/hr, your profit is h(r-c) dollars, which is always positive. (This assumes no overhead costs, which is why small many Linux companies fail, because the overhead costs overwhelm them.)
Selling software at a flat rate is a tremendously risky business. You pay $d to develop it, and sell n copies at cost c. Your profit is d - cn dollars, which can be negative. One dud product and you're out of business. This is why there is only one large company on the face of the planet that generates a substantial amount of revenue from software - Microsoft. For them, the risk is mitigated to an absurd extent because n will be very high no matter what because people are forced to buy their products. And this is why MS places such a low priority on quality - they don't get more money for high quality software. Making Windows more stable increases costs but doesn't affect revenue substantially. This is why in 2001 my Windows box still locks up all the time - because implementing full memory protection and a host of other features common to every other OS wouldn't earn them another cent. It helps a lot that I can't return their software when it fails to meet expectations. (This is why each successive version of Windows is flashier, but not necessarily better from a technical standpoint.)
It makes far more sense to "sell" free software, have little general-purpose development, and provide services (including specialized development), which offer better and more predictable profits.
What happens if you are a customer who was sold a system with one of those "other" operating systems. You'd scream bloody murder if the OS running your systems was going to be trashed in favor of something else.
If that happens you can be sure that Microsoft is going to try and sweep in to pick through the carnage...
I don't think the BizWeek guy is suggesting they drop support for their Unices tommorrow. (BizWeek cheering on Linux??? What is this world coming to?)
They're all Unix flavors, all conforming to POSIX to some degree or another. If your inhouse developers followed the POSIX standard closely, (which they should have, as that gives you the flexibility to switch unix flavors), tranistioning to Linux is an order of magnitude easier than switching to NT/2K. And training an HPUX staff to use Linux is substantially easier than teaching them the MS way. Running to Microsoft would be even more difficult.
They're startups, they're supposed to lose money initially. Startups succeed when you go from losing money to breaking even to turning a profit. They fail when they lose more and more money. Redhat is breaking even, which means their revenue is rising faster than their costs and we can assume they'll proceed on to profitability. Microsoft is making money, but making less every year. Their costs are rising faster than their revenue. Factor in their impending breakup, dozens of lawsuits getting thrown at them, and a few other things, and their long-term outlook isn't very good.
Why not create a govt agency or a private company that certifies proprietary code as GPL-free? You give them the source code, they compare it to the source code of similar GPL products, most likely using automation and a good string matching algorithm, and if the algorithm finds no nontrival matches in the code, then the program is good.
:)
Of course, the code-checking engine would have to be open-source so companies could be sure that it didn't steal code
You can make a tremendous amount of money even if you sell a highly visible product at a loss - sell a profit-making complement too.
Gllette spent millions developing the Mach 3 razor and sells each one at less than 20% of the manufacturing price. They make money selling replacement blades above cost. It works, its worked for decades.
RedHat sells RH Linux below cost ($0). But compared to Windows, it's cheap to develop - you take other people's innovations, add a few, and package it in a useable format. Then you sell services above cost - no OS is so easy to use that there are no support issues. RH Linux is a loss-leader to get people to establish a relationship with the company. It would make sense that if I need high-level support for RH Linux, I would turn to RedHat, since they know the product better than anyone esle. IBM has the same business model now - they want to sell you a complete Linux solution - they've largely given up on PCs to be a services company.
This sounds stupid, but service/support is a better business than proprietary software. All of your costs are marginal - if you sell n hours of support for $r/hr and you pay $e/hr to the support guy, you make n(e-r) dollars, which is always a positive number if e>r. Software costs d dollars to develop, where d is a very fucking large number. If n people buy it a cost c, you make (n*c - d). This can be negative if not enough people buy your product. This is a gross oversimplification, there's overhead in support, and some neglible marginal costs in selling software, but this is why there are so many little support shops - you don't need to invest a tremendous amount of cash upfront.
Microsoft is one of the few large companies on the face of the earth whose only product is proprietary consumer software. Oracle, IBM, Apple, you name it, all sell something else besides software - consumer software is a ridiculously risky business. They've managed to succeed by becoming a monopoly and essentially forcing people to buy their products. They used illegal pricing games to drive the competition out of business - offering "competitive upgrades" on competitetors products, giving away IE, breaking cross-platform compatibility and so forth. They can't use the same strategy against Linux because you can't undercut it on price, so they've resorted to their other famous tactic, FUD.
No, Felten makes very clear those provisions would only apply if they took the cash prize, which they politely declined. And they're not "attacking content", they're explaining watermarking schemes a second-year math undergrad would understand.
The paper contains pseudocode-like descriptions of how to retain your rights. It's a "device" under the DMCA
That's the whole problem, if source code, a very precise and computer-centric format for describing a process is a device, translating the "device" into a less precise and more human-centric format means it's still a "device".
It's easy. If you work on a project that does x during the day, any freetime work should be reasonably unrelated to x. It's safer for both parties that way, there's no chance the company will end up with GPLed code in its product, and there's no way the GPL project will end up with code that belongs to somebody else. Furthermore, it gives you a broader experience and a chance to do something different.
I think the submittor is talking about the annoying tendency of comapnies to claim off-hours project as their IP. If I write a program on my own time, I should be able to claim it as my intellectual property, and slap the GPL on it. It's not like I'm taking a second job, I'm giving away my work.
But the land-grab mentality about intellectual property causes problems. If I use a company laptop, there might be grounds for them to claim they own my project. Same if I use a company email account on a project mailing list, or if post a quick note about a bug from work.
Something like this makes sense:
"XYZ corp disclaims all rights to Joe Coder's project Foo. Use of corporate assets such as laptops, workstations, or other systems does not grant XYZ corp any rights to Joe Coder's project."
It's pretty reasonable, and assuming Joe is releasing under the GPL there's no chance he'll ever make money off the product himself, so he doesn't have a financial incentive to neglect his real job for the Foo project.
You're right, but this illustrates the problem with the USPTO today - they can't keep up. By the time a patent is granted, 100 other people have idependently created the same technology, and now they're screwed because the stealth patent beats them up.
On the other hand, I think it's pretty clear there is a tremendous amount of "prior art" ie pre-1996 here, so there seems to be no ground for a patent either way.
....wherein the step of limiting is a step of determining which avatars are closest in the graphical space to the local avatar.
The patent uses the word "avatar" over 30 times to describe a person's representation in a 3-D environment. Funny, Stephenson called them that in SnowCrash. Wait, he basically described the thing being patented. Oops.
I think the /. crowd is most pissed off by the fact that Intel's technical decisions are made partly with a view to marketing. The "Internet architecture" BS a couple years back Intel's own have engineers bitched there is no way to optimize a CPU for Internet - net bandwith will always be the limiting factor, not the CPU. But we still saw all those idiotic "Intel gives you power for the Internet" commercials. Now we've got a chip that appears to be grossly overpriced for its performance, released now only because of "MHz marketing". I think the P4 has an good architecture which scales like crazy, but why was it released before it reached 2 GHz and that scaleablity was an asset and not a liability? The UID for CPUs and didn't help Intel's image either - that pissed off the general public too.
That said, assuming the P4 continues to play catch-up well, I'll buy one if it offers more value for my money.