With no real threat of serious (ie costly) legal action for violating the GPL, what's to stop this happening again and again? How many other companies have stolen GPL code and are distributing it without our knowing about it?
Then again, if someone did sue for copyright infringement, what kind of damages could you claim?
Why is this rocket science? If someone is "violating the GPL", they are violating copyright law. The damages, which every slashdot reader should know well since they rail against the RIAA everytime they attempt to collect them, is $150K per incident. These are "statutory damages", meaning that I don't have to prove that $150K of damage was done to me for each incident.
Read this paragraph until it makes sense. LinkSys has no inherent right to distribute any of that GPL'd software because someone else owns the copyright. The copyright owners have said "you may distribute my copyrighted code under these very strict conditions (which you and I know as the GPL), but I reserve all other rights." If LinkSys decides to not abide by those conditions, then they have no right to distribute the software. This is why the whole "but the GPL hasn't been tested in court!" argument shows remarkable cluelessness in those who use it. It's not relevant, standard copyright law applies.
In looking over these tables, one can't help but wonder why SCO's UnixWare and OpenServer are even mentioned. They offer nothing over GNU/Linux, *BSD, BSD/OS, and Solaris, yet UnixWare is astonishingly more expensive than its competitors.
In every single instance that I've seen SCO installed, it's been running a vertical market application running on unibase. The single biggest factor driving SCO sales has been a varitable legion of programmers and resellers who are making money from programs that were written 10 years ago when SCO made some amount of sense.
Given that the programs are unique to Unibase, and given that Unibase runs just fine under Linux and has for some years, SCO's market (which is small businesses that are just large enough to spend a few thousand on a computer system up to ~$50M/year businesses that aren't large enough to buy a real Unix system) is running to Linux. I've seen a few VAR's holding out on SCO, but very few and dwindling.
I have one client still using SCO, and they're doing all they can to leave it. I've been out in the real world as a consultant for 9 years now and in that time I have never (not even one time) heard of or observed a new SCO installation, nor have I found anybody who has even considered it.
SCO was basically dead a long time ago, I guess nobody bothered to tell them.
I can't imagine what their licensing fee will be, what's the licensing fee for an equivalent machine running openserver? There must be some out there since all these enterprise capabilities came straight from the heart of SCO.
While the overall system may be as efficient for an electric car, there's a gross inefficiency in that you have to carry around 1000 lbs. of batteries just to get 80 miles (and note that's in a tiny tiny car, it's more on the order of 60 miles for a car that would be considered "compact" and seat 4).
With a gasoline/diesel engine, I have to settle for a drive train that weighs in at quite a few hundred pounds as overhead, but then I can take all the fuel that I can carry, and I get (talking about a normal sized car) 30 miles for every 8 or so pounds of fuel. A 12 gallon tank adds another 100 or so pounds, but I can go 360 miles.
I also have much lower maintenance (just *one* battery to change every 5 years). The engine also has two fringe benefits that I can take advantage of. First, there's a lot of waste heat (remember that inefficiency), so I can heat the interior of the car if need be. It's not practical to waste the batteries in an electric car for anything other than minimal heat. And there's AC. Again, are you going to knock 10 or 20% of your miles off to be comfortable? Not a big deal if you have hundreds of miles to start with, big deal if you can only go a few.
Given the drawbacks, I don't think battery powered cars will catch on unless we can have a monumental gain in battery effiency (in terms of the amount of power per pound of battery). Hybrids look good short term (and they are far more efficient than either pure battery or pure fossil fuel). Hopefully we'll get fuel cells to the point that they are a possibility in the near future.
There is no consideration, sorry. The GPL doesn't state that you have to give/sell anything back to the original author.
The GPL is not a contract, it is a grant of rights that you wouldn't normally have (i.e. the right to copy a copyrighted work).
Ever notice how most copyright notices have "all rights reserved" beside them? What they're saying is "I own the copyright (i.e. right to copy) this work and I am giving you none of those rights. I am reserving them for myself."
The GPL says "you have the right to copy this work if and only if you pass this right (and a few others) on to anyone who obtains a copy." It is a grant of rights that you don't have otherwise.
That's why the whole "but the GPL hasn't been tested in court!" argument is bullshit. It's moronic. If you don't abide by the GPL, then you have no right to redistribute the code in question, basically standard copyright law which is "tested" in the courts on a daily basis.
That's the position SCO is in. The bill that they've racked up to Linus Torvalds would basically turn McBride's great-great grandchildren into debt-slaves of Linus Torvald's great-great grandchildren. But only if Linus gets his head out of his ass and does something about it now.
It's not a breach of contract, as the GPL isn't a contract. The GPL is simply a listing of your limited rights to redistribute a copyrighted work. If you don't wish to abide by the conditions, then you have no right (the position that SCO's in), and hence it's a simple copyright violation.
If Linus cared to pull his head out of his ass, he could own SCO simply because the only way that they'd have to pay the statutory damages ($150K/violation) would be to give him the company.
I have to ask, and it's too bad that I'm too late to be modded up, but at what point is Linus Torvalds going to pull his head out of his ass and sue SCO for violating HIS copyrights? They are still distributing his copyrighted material, and they have decided to not abide by the terms of the GPL. $150K/incident (download) in statutory damages are waiting for him. He can own their company (literally, they can't begin to pay what they owe) and make their "IP" freely available.
Please, Linus, take care of this festering wound. It's not going away.
I've been predicting for a couple of years now that the pure CGI movies will be going this direction, i.e. rendered real-time in the theatre or your house. We're fewer than 5 years away from being able to render a Monsters Inc. style movie in real time at a high resolution using a $100 video card. We're almost at the level of the original Toy Story right now, with the major piece left being hair (which is a complete bitch to render, and not present in the original Toy Story).
The current state of CGI is such that we can make a mostly believable CGI character, and very believable CGI backgrounds and such. It's very likely that one of the major players could turn out a pure CGI movie that would be nearly indistinguishable from live action. Within 8 or 9 years, we'll be rendering those in real time on a $100 video card.
Should be interesting, since Hollywoood's going to be crazy stupid over preventing anyone from getting to their character and scene definitions and making incredibly bitching fan films. How many people out there could write a better Star Wars flick than George Lucas? Let's make that simpler to count: how many couldn't?
It's a lot less effort to sic the lawyers on people than actually PATCH the vulnerability.
Yeah, but you might want to look at what they did. It's easier yet to just tell the Justice Dept. to arrest and prosecute someone than to sic your lawyers on them. Think about it, it's no skin off their backs.
Well, they're betting the company anyway, so if they lose then Stallman can get his $150,000 after all the other creditors have picked over the carcass. In other words, heads they win, tails don't matter anyway.
It's not $150K, it's $150K per violation. Each time somebody downloads it, it is a separate violation.
You're right that copyright won't protect everybody else, but hopefully Linus will get his head out of the sand at some point and have his lawyer tell SCO they're going to get a bill that'll make their IBM lawsuit look like chump change if they don't give it up.
$1B is fewer than 7000 downloads. I could probably pull the kernel source down that many times tonight with a script.
Finally, he makes the old, tired argument that the GPL saves us. That's all well and good except that the GPL, being a bit of a non-standard contract, has never been tested in court.
I cringe every time I see this point being made. The GPL isn't relevant, really. If the GPL doesn't "stand up in court", then no problem, standard copyright law takes over. SCO has no other right to distribute the Linux kernel under copyright law beyond those given by the copyright holder.
What this means practically speaking is that if SCO violated the GPL, and the GPL is declared invalid, then SCO has instead violated standard copyright law. The statutory damages are $150,000 per incident. Everytime someone downloads the Linux kernel source from their website, ka-ching. Every sale that they've made, ka-ching. Pissing off Richard Stallman suddenly sounds like the better way off, eh?
The bottom line is that SCO continues to distribute the Linux source code. As such, they're either agreeing to distribute it under the terms of the GNU GPL (which don't allow extra burdens to be placed on the code) or they are directly violating copyrights. They are far better off if they get off with a simple GPL violation; $150K/download adds up quickly.
I agree with your sentiment but I need to correct a statement of fact. Mosaic was free but I don't think it was open source. Remember how they licensed it to a bunch of commercial companies like Spyglass and Microsoft? Mosaic was free to redistribute but the code was not open.
I'm serious. It's ironic that the "end of innovation" coincides with his leaving Netscape as well as Netscape's doomed 4.x series piece of shit^H^H^H^H^H^H^H^H^H^H^H^H^Hbrowser. Before that, "innovation" was Netscape ignoring the W3C and making up new "standards" every other week.
Andreessen should be a pariah in the open source world. He abandoned an open source project (Mosaic & NCSA httpd) in order to compete with it in the commercial world. "Competition" in the Microsoft sense of the word: Gain the upper hand in the market then "innovate" so much that nobody can keep up. And, of course, give away the browser free of charge in order to sell the server. When Microsoft finally woke up to the web, Netscape was playing on their ballfield and obviously lost.
Anyway, I'm tired of hearing him and Jim Barksdale whine about the browser market. Get over it already.
In 1879 the Indiana House of Representatives unanimously passed a bill that redefined the area of a circle and the value of Pi. Luckily the bill died in the State Senate, or y'all might have real problems with things like highway interchanges.
This is part urban legend, part true. The "History of Pi" book by Petr Beckman actually shows the bill and gives more information. However, the pi==3 aspect is false. And, the bill never got anywhere.
Unfortunately in this case the overflow is in the hardware in the wiring. To fix it you need to rip the machine down to bare metal and replace every component with a cricuit board with a new one.
Right, but the argument really comes back around to the fact that cell phones and other electronics don't cause harm to planes. If they did, planes would be dropping like flies. If you take time to read this article, you'll find that if you hold your cell phone next to a particular circuit board (which is likely buried deeply enough within a panel as to be impossible in real life, anyway) then you might cause a navigation sensor to be wrong.
It's obvious that the FTC knows that cell phones don't cause problems, otherwise there would be a required upgrade for all airplanes, and incoming passengers and baggage would be scanned to see if they were giving off EMR.
My bad attitude toward this is caused by the fact that various elements are continually putting out scare material to make people think that their cell phone is going to crash a plane. It should be blatantly obvious to people by now that it's not going to happen, but we need the FTC to decide one way or the other and make it public.
And if the exploit actually does exist, it needs to be fixed. But I say that knowing that there is no exploit.
Why isn't "Please turn off your cellphones" a fix? It would seem to deal with the problem--crudely, yes, but effectively. So what's the disqualifier? It's inconvenient? If so, your point would be...?
It doesn't work because it's voluntary. Again, why do we fix computer exploits? Why not just convince crackers and virus writers to play nice?
People forget about their cellphone, might put it in the cargo hold, who knows? We can't plan for all of those cases, we can't trust everybody (is that not obvious?), so the only thing left is to fix the problem.
Fine, so cell phones really do disrupt airplanes. I still don't believe it, but, if it's true then we've identified an exploit that needs to be fixed. "Please turn off your cellphone" is not a fix.
"Software company X has identified a buffer overflow in our popular Y software, which can lead to a remote root exploit. Rather than fixing it, we're asking that you please don't connect to port yz and send a string that is 5000 characters long and ends with the binary sequence..."
Moronic. Fix the bug and quit boring us with the details.
Slashdot Mirror
With no real threat of serious (ie costly) legal action for violating the GPL, what's to stop this happening again and again? How many other companies have stolen GPL code and are distributing it without our knowing about it?
Then again, if someone did sue for copyright infringement, what kind of damages could you claim?
Why is this rocket science? If someone is "violating the GPL", they are violating copyright law. The damages, which every slashdot reader should know well since they rail against the RIAA everytime they attempt to collect them, is $150K per incident. These are "statutory damages", meaning that I don't have to prove that $150K of damage was done to me for each incident.
Read this paragraph until it makes sense. LinkSys has no inherent right to distribute any of that GPL'd software because someone else owns the copyright. The copyright owners have said "you may distribute my copyrighted code under these very strict conditions (which you and I know as the GPL), but I reserve all other rights." If LinkSys decides to not abide by those conditions, then they have no right to distribute the software. This is why the whole "but the GPL hasn't been tested in court!" argument shows remarkable cluelessness in those who use it. It's not relevant, standard copyright law applies.
Michael
In looking over these tables, one can't help but wonder why SCO's UnixWare and OpenServer are even mentioned. They offer nothing over GNU/Linux, *BSD, BSD/OS, and Solaris, yet UnixWare is astonishingly more expensive than its competitors.
In every single instance that I've seen SCO installed, it's been running a vertical market application running on unibase. The single biggest factor driving SCO sales has been a varitable legion of programmers and resellers who are making money from programs that were written 10 years ago when SCO made some amount of sense.
Given that the programs are unique to Unibase, and given that Unibase runs just fine under Linux and has for some years, SCO's market (which is small businesses that are just large enough to spend a few thousand on a computer system up to ~$50M/year businesses that aren't large enough to buy a real Unix system) is running to Linux. I've seen a few VAR's holding out on SCO, but very few and dwindling.
I have one client still using SCO, and they're doing all they can to leave it. I've been out in the real world as a consultant for 9 years now and in that time I have never (not even one time) heard of or observed a new SCO installation, nor have I found anybody who has even considered it.
SCO was basically dead a long time ago, I guess nobody bothered to tell them.
I can't imagine what their licensing fee will be, what's the licensing fee for an equivalent machine running openserver? There must be some out there since all these enterprise capabilities came straight from the heart of SCO.
While the overall system may be as efficient for an electric car, there's a gross inefficiency in that you have to carry around 1000 lbs. of batteries just to get 80 miles (and note that's in a tiny tiny car, it's more on the order of 60 miles for a car that would be considered "compact" and seat 4).
With a gasoline/diesel engine, I have to settle for a drive train that weighs in at quite a few hundred pounds as overhead, but then I can take all the fuel that I can carry, and I get (talking about a normal sized car) 30 miles for every 8 or so pounds of fuel. A 12 gallon tank adds another 100 or so pounds, but I can go 360 miles.
I also have much lower maintenance (just *one* battery to change every 5 years). The engine also has two fringe benefits that I can take advantage of. First, there's a lot of waste heat (remember that inefficiency), so I can heat the interior of the car if need be. It's not practical to waste the batteries in an electric car for anything other than minimal heat. And there's AC. Again, are you going to knock 10 or 20% of your miles off to be comfortable? Not a big deal if you have hundreds of miles to start with, big deal if you can only go a few.
Given the drawbacks, I don't think battery powered cars will catch on unless we can have a monumental gain in battery effiency (in terms of the amount of power per pound of battery). Hybrids look good short term (and they are far more efficient than either pure battery or pure fossil fuel). Hopefully we'll get fuel cells to the point that they are a possibility in the near future.
That Bill Gates really is Santa Claus. At what point will their incredible generosity meet its limit?
There is no consideration, sorry. The GPL doesn't state that you have to give/sell anything back to the original author.
The GPL is not a contract, it is a grant of rights that you wouldn't normally have (i.e. the right to copy a copyrighted work).
Ever notice how most copyright notices have "all rights reserved" beside them? What they're saying is "I own the copyright (i.e. right to copy) this work and I am giving you none of those rights. I am reserving them for myself."
The GPL says "you have the right to copy this work if and only if you pass this right (and a few others) on to anyone who obtains a copy." It is a grant of rights that you don't have otherwise.
That's why the whole "but the GPL hasn't been tested in court!" argument is bullshit. It's moronic. If you don't abide by the GPL, then you have no right to redistribute the code in question, basically standard copyright law which is "tested" in the courts on a daily basis.
That's the position SCO is in. The bill that they've racked up to Linus Torvalds would basically turn McBride's great-great grandchildren into debt-slaves of Linus Torvald's great-great grandchildren. But only if Linus gets his head out of his ass and does something about it now.
Mitigation of damages is a two-way street...
Michael
It's not a breach of contract, as the GPL isn't a contract. The GPL is simply a listing of your limited rights to redistribute a copyrighted work. If you don't wish to abide by the conditions, then you have no right (the position that SCO's in), and hence it's a simple copyright violation.
If Linus cared to pull his head out of his ass, he could own SCO simply because the only way that they'd have to pay the statutory damages ($150K/violation) would be to give him the company.
I have to ask, and it's too bad that I'm too late to be modded up, but at what point is Linus Torvalds going to pull his head out of his ass and sue SCO for violating HIS copyrights? They are still distributing his copyrighted material, and they have decided to not abide by the terms of the GPL. $150K/incident (download) in statutory damages are waiting for him. He can own their company (literally, they can't begin to pay what they owe) and make their "IP" freely available.
Please, Linus, take care of this festering wound. It's not going away.
Michael
Using a series of magnets to accelerate a metal slug - it doesn't seem like it would be that hard to do. Right?
If the metal slug is made out of something that isn't attracted by magnets, it'd be a complete bitch...
How many of us have used fake email addresses to identify spammers?
I've been predicting for a couple of years now that the pure CGI movies will be going this direction, i.e. rendered real-time in the theatre or your house. We're fewer than 5 years away from being able to render a Monsters Inc. style movie in real time at a high resolution using a $100 video card. We're almost at the level of the original Toy Story right now, with the major piece left being hair (which is a complete bitch to render, and not present in the original Toy Story).
The current state of CGI is such that we can make a mostly believable CGI character, and very believable CGI backgrounds and such. It's very likely that one of the major players could turn out a pure CGI movie that would be nearly indistinguishable from live action. Within 8 or 9 years, we'll be rendering those in real time on a $100 video card.
Should be interesting, since Hollywoood's going to be crazy stupid over preventing anyone from getting to their character and scene definitions and making incredibly bitching fan films. How many people out there could write a better Star Wars flick than George Lucas? Let's make that simpler to count: how many couldn't?
Michael
My head was starting to get itchy inside the tinfoil hat!
It's a lot less effort to sic the lawyers on people than actually PATCH the vulnerability.
Yeah, but you might want to look at what they did. It's easier yet to just tell the Justice Dept. to arrest and prosecute someone than to sic your lawyers on them. Think about it, it's no skin off their backs.
Well, they're betting the company anyway, so if they lose then Stallman can get his $150,000 after all the other creditors have picked over the carcass. In other words, heads they win, tails don't matter anyway.
It's not $150K, it's $150K per violation. Each time somebody downloads it, it is a separate violation.
You're right that copyright won't protect everybody else, but hopefully Linus will get his head out of the sand at some point and have his lawyer tell SCO they're going to get a bill that'll make their IBM lawsuit look like chump change if they don't give it up.
$1B is fewer than 7000 downloads. I could probably pull the kernel source down that many times tonight with a script.
Michael
Finally, he makes the old, tired argument that the GPL saves us. That's all well and good except that the GPL, being a bit of a non-standard contract, has never been tested in court.
I cringe every time I see this point being made. The GPL isn't relevant, really. If the GPL doesn't "stand up in court", then no problem, standard copyright law takes over. SCO has no other right to distribute the Linux kernel under copyright law beyond those given by the copyright holder.
What this means practically speaking is that if SCO violated the GPL, and the GPL is declared invalid, then SCO has instead violated standard copyright law. The statutory damages are $150,000 per incident. Everytime someone downloads the Linux kernel source from their website, ka-ching. Every sale that they've made, ka-ching. Pissing off Richard Stallman suddenly sounds like the better way off, eh?
The bottom line is that SCO continues to distribute the Linux source code. As such, they're either agreeing to distribute it under the terms of the GNU GPL (which don't allow extra burdens to be placed on the code) or they are directly violating copyrights. They are far better off if they get off with a simple GPL violation; $150K/download adds up quickly.
Or: If you make something idiot-proof, they'll make a better idiot. Makes me think of Crab's record player (obscure reference).
I agree with your sentiment but I need to correct a statement of fact. Mosaic was free but I don't think it was open source. Remember how they licensed it to a bunch of commercial companies like Spyglass and Microsoft? Mosaic was free to redistribute but the code was not open.
More info:L icense/LicenseInfo.html
http://archive.ncsa.uiuc.edu/SDG/Software/Mosaic/
It was free but not Free, and source was available for Un*x under a BSD style license, I think.
Michael
I'm serious. It's ironic that the "end of innovation" coincides with his leaving Netscape as well as Netscape's doomed 4.x series piece of shit^H^H^H^H^H^H^H^H^H^H^H^H^Hbrowser. Before that, "innovation" was Netscape ignoring the W3C and making up new "standards" every other week.
Andreessen should be a pariah in the open source world. He abandoned an open source project (Mosaic & NCSA httpd) in order to compete with it in the commercial world. "Competition" in the Microsoft sense of the word: Gain the upper hand in the market then "innovate" so much that nobody can keep up. And, of course, give away the browser free of charge in order to sell the server. When Microsoft finally woke up to the web, Netscape was playing on their ballfield and obviously lost.
Anyway, I'm tired of hearing him and Jim Barksdale whine about the browser market. Get over it already.
In 1879 the Indiana House of Representatives unanimously passed a bill that redefined the area of a circle and the value of Pi. Luckily the bill died in the State Senate, or y'all might have real problems with things like highway interchanges.
This is part urban legend, part true. The "History of Pi" book by Petr Beckman actually shows the bill and gives more information. However, the pi==3 aspect is false. And, the bill never got anywhere.
Michael
To this end, I'm taking suggestions as to innovative and torturous ways to take a SCO Unixware box down.
Carefully zero out any parts of the running kernel which are covered by the GPL since being copied from Linux...
Michael
Nice book, whatever. What's SCO doing today?
Unfortunately in this case the overflow is in the hardware in the wiring. To fix it you need to rip the machine down to bare metal and replace every component with a cricuit board with a new one.
Right, but the argument really comes back around to the fact that cell phones and other electronics don't cause harm to planes. If they did, planes would be dropping like flies. If you take time to read this article, you'll find that if you hold your cell phone next to a particular circuit board (which is likely buried deeply enough within a panel as to be impossible in real life, anyway) then you might cause a navigation sensor to be wrong.
It's obvious that the FTC knows that cell phones don't cause problems, otherwise there would be a required upgrade for all airplanes, and incoming passengers and baggage would be scanned to see if they were giving off EMR.My bad attitude toward this is caused by the fact that various elements are continually putting out scare material to make people think that their cell phone is going to crash a plane. It should be blatantly obvious to people by now that it's not going to happen, but we need the FTC to decide one way or the other and make it public.
And if the exploit actually does exist, it needs to be fixed. But I say that knowing that there is no exploit.
Michael
Why isn't "Please turn off your cellphones" a fix? It would seem to deal with the problem--crudely, yes, but effectively. So what's the disqualifier? It's inconvenient? If so, your point would be...?
It doesn't work because it's voluntary. Again, why do we fix computer exploits? Why not just convince crackers and virus writers to play nice?
People forget about their cellphone, might put it in the cargo hold, who knows? We can't plan for all of those cases, we can't trust everybody (is that not obvious?), so the only thing left is to fix the problem.
Michael
Fine, so cell phones really do disrupt airplanes. I still don't believe it, but, if it's true then we've identified an exploit that needs to be fixed. "Please turn off your cellphone" is not a fix.
"Software company X has identified a buffer overflow in our popular Y software, which can lead to a remote root exploit. Rather than fixing it, we're asking that you please don't connect to port yz and send a string that is 5000 characters long and ends with the binary sequence..."
Moronic. Fix the bug and quit boring us with the details.
Michael
Think about that next time you buy a CD and give these greedy pigs another $18.
That'll be around the next time I buy a new SCO machine...
Michael