Let me summarize first: Company A writes code X that they want to keep proprietary. The code is stolen and incorporated into some open source software Y that is distributed under the GPL. Company A receives a copy of that software Y, adds code Z, and redistributes the changed software under the GPL, not knowing that their own code X in stolen form is part of the software. Then they detect that their own code X had been included in the software illegally.
First, that code X has never been distributed under the GPL so far. The whole package (X + Y + Z) always consisted of two parts Y and Z that are distributed under the GPL, and one part X that is the property of A and has been distributed illegally (although most distributors were not aware of this). A has no obligation to distribute the source for X at all.
Now the situation is obviously quite unfortunate. If A finds out who stole the software, they can sue them for damages. In the SCO vs IBM case, if SCO could demonstrate that IBM stole the code, they would receive damages from IBM, and IBM is rich enough to pay - if some bankrupt programmer stole the code, A will not be able to get damages. Whether anyone distributing so far innocently can be sued for damages, I don't know. But what is the next step A should do?
There are several choices: A can stop distributing the code, and tell all other distributors, like Redhat, SuSe to stop distributing it as well. They'd have to stop immediately, otherwise damages have to be paid. This is obviously very painful for everyone involved, distributors and users alike. Second, A can give up and distribute the software under GPL, which is painful for A only and good for everyone else - however, A would still be entitled to full damages. Third, A could negotiate with say Redhat and SuSe. A could say: I stop distributing my version right now, which costs me money. I'll give you three days to think about this and I won't sue anyone who distributes within the next three days, but after that I want a cash offer for distributing X under GPL, otherwise I'll force everyone to stop distributing it, as I am entitled to. (Note that anyone who would distribute X + Y + Z in the next three days would be save from being sued by A, but would be in violation of the GPL. This is not A's fault. )
What A can not do: Continue to distribute (X + Y + Z) without licensing X under the GPL.
'' A problem with obviousness is that because inventions rely on the laws of nature, in hindsight they are always logic. The reward should be for recognizing something that not the ordinary person skilled in the art would recognize. Now, try to make that objective! So, a gray area is bound to remain. ''
Given that your average patent nowadays is written in such a way to make it impossible to glean any useful information off it, one could just measure the amount of effort that it took a so-called "infringer" to come up with the "invention" itself. Many patents that have been granted are on a level that I would expect to ask someone in a job interview "how would you do this" and expect some reasonable answer. Software patents have been issued for things that I would be expected to develop in the normal course of my job.
Some court has come up with a ridiculous measure for obviousness: If it is obvious, then somebody should have found and published it, right? Of course not. If it is obviousness, then nobody should ever have bothered publishing it.
'' I had a calculus professor who thought calculators were a Commie plot to inhibit the minds of Americans. ''
More likely you had a calculus professor who claimed that calculators were a Commie plot to inhibit the minds of Americans, and you were just incapable of detecting humor and/or sarcasm.
Apparently filming kids secretly in a classroom is "underhanded" but not illegal; teachers seem to be unhappy about film material demonstrating that they lost control completely. I can't find anything that says whether open filming would be legal or illegal.
'' Kids who deliberately provoke a teacher to film the results don't need to be yelled at so much as slapped around a little. And that's why I'd be a terrible teacher. ''
You would lose your job as a teacher rather quickly. Doesn't mean you would be a bad teacher.
My suggestion: Install video cameras in all classrooms. The teachers can film everything, edit to their hearts content, and publish the best bits on YouTube. Must be great fun seeing thousands of creepy little morons being shown to be creepy little morons for everyone to see. Like the bloody idiot who wrote the headline for this submission and couldn't even get the spelling of "relevant" right.
'' Last time I checked on the iTunes Store, songs were 0.99$, not 2.80$. Where is that 2.80$ price coming from? ''
Just a suggestion: First, the $0.99 is irrelevant. The damage to the recording industry is not $0.99. It is estimated that Apple pays $0.70 per song to the recording industry, $0.29 is for Apple to run the store, supply the music, make a bit of profit. Since you didn't buy from iTMS, Apple can't really complain (they could complain if you found a way of downloading from the iTMS without paying), so the damage is $0.70. Multiply by four as a reasonable estimate that a song that you offer for download will be download four times on the average.
The $700 minimal damages are completely unreasonable, because that would require 1000 downloads. Now if you made an illegal copy of the latest U2 album before it is released, and started selling records based on that copy before U2's record company starts selling, then I wouldn't say that the upper range of $150,000 for unproven damages couldn't be appropriate (and the record company would be free to find evidence that actual damages are even higher), but $700 for uploading a song that is available for $0.99 everywhere is unreasonable.
'' Things here in the UK always seem to be thought of as failing or disaster before they're completed. I'm sure we hate success as a nation. We also have a huge obsession with celebrity and magazines that publish how fat celebrities are, or how their lives are in a mess always do very well.
I say wait until the project's finished before kicking it to the ground. ''
Ok, lets start then by kicking the CSA (Child Support Agency) to the ground first, followed by whoever is responsible for the passport debacle.
If we have to wait until a project is finished, we will likely _never_ have the right to complain about the NHS projects.
Intentional or error?
on
The Zune Cometh
·
· Score: 4, Interesting
The submitter calls the Zune an "iPod wannabe-killer".
That would make it something that kills iPod wannabe's, like Creative or Rio or Sandisk players. I wonder whether that is what the submitter meant, or did he mean "wannabe iPod" or "wannabe iPod-killer"? And I wonder what Microsoft's goal is?
'' This machine possibly made 1 error out of a total of 36 votes. Since we don't know what actually caused the error, it is possible that this machine would make the same error for every 1/36 votes. ''
If the guy is saying the truth, then all we know is that _at least_ one vote was not counted correctly. Since 18 votes were counted for candidates A and B, you would need either 19 people who state that they voted for A, or 19 people who stated they voted for B, to _prove_ that another vote was counted wrong, but we don't have those 19 people.
So the correct version is: For one of the 36 votes cast we know whether it was right or wrong, and that vote was counted wrong. For the other 35 votes, we have no information. So _all_ votes we can verify were counted wrong. Those that we cannot verify, we don't know anything about, but we sure would want to know.
'' If you live in the UK, what happens if you submit a story linking to a UK website that you do not own, the server can't handle the resulting traffic, and the owner doesn't appreciate the attention? ''
Just try to think about this logically. Lets say I have a web page; I want people to read it (that can be assumed safely, because otherwise I wouldn't have a web page), and my ISP is fine with 1000 page hits per day, but not more. Instead of 100 page hits I get 1000 hits because of a Slashdot article, and then the server crashes. Obviously, my web page has been seen ten times more often than it would have been seen without the Slashdot article, so I would have a very hard time of convincing anyone that any harm has been done.
'' Oh and those Singapore immigration landing cards are a hoot, with large red friendly letters "possession of drugs is punished by DEATH'. ''
I think it is in Malta where they have a sign that importing drugs is illegal, and if you pass _this_ sign then it is considered importing, and they have a bin right in front of the sign where you can drop everything that you don't want to import. Very sensible approach.
'' The user agreement for virtually all ISPs does not allow their users to share their internet connection wirelessly, no matter how generous your neighbors feel. ''
I checked the terms and conditions for Orange, former Wanadoo. Quite interesting. I am only allowed to use my connection for residential use. I think that allowing my neighbour free access is clearly residential use. I am explicitely not allowed to resell the connection. So allowing my neighbour access for money is disallowed. Commercial use is not allowed. This is a change in their terms; it used to be that the service was "not intended for commercial use", which I assume means that if my business goes bankrupt because my broadband connection fails, that's my problem and not theirs. Then there is "fair usage"; sharing the connection obviously makes it easier to exceed whatever "fair usage" means.
Most importantly: I am sure that Orange terms allow me to make broadband available to others for free as long as they don't use too much bandwidth.
'' Is this really a surprise? This is the same country where you can be caned for simple acts of vandalism. ''
Many people believe that this kind of punishment is very appropriate. I have much less sympathy for a brainless idiot who thinks it is fun to scratch a new car than for someone robbing a bank.
Right now, spam goes past spam filters by including a large amount of random nonsense text that resembles English language reasonably well. So we will get spam filters that detect large amounts of random nonsense text. So spam will include text that makes actual sense. Give it twenty years, and your average spam email will consist of 300 pages of text that is better than anything Shakespeare has ever written, followed by two lines begging you to buy viagra. Thirty years, spam will be two hour Quicktime movies better than anything you can watch in the cinema today, with the hero using viagra bought from the spammer in the right places.
''Are the binaries encrypted, or just signed? '' Encrypted.
"Does the hardware have a public key hardwired into it, and if so, can someone just extract that key from a particular mac, for everybody else to use?" Some part of the decryption is hardwired in Macintosh hardware. I don't know if it can be extracted, but by doing so you move from the area of civil law (copyright) to criminal law (DMCA).
''Can Apple's mechanism be used to forbid people from running software that doesn't come from a vendor that's registered itself with Apple? '' Such things can be done, but the use of a decryption key that is available on Macs and nowhere else doesn't help.
"Are the components we're talking about open-source, or not?" Not.
AES-128 is one of the more expensive codes to decrypt, but with a bit of effort a single core on a Core Duo chip can do it at a speed that is quite comparable to harddisk read speed. And since I haven't read any complaints from Mac users how slow these machines are booting, I think there is no reason to worry about that.
'' The moment I ask "how much?" and they tell me, they have entered into a contract to supply the goods at that price or a lower price if they can't make change. ''
Of course "they" haven't. Where do you get that nonsense from?
Wrong analogy. What you mean is: If BMW was to make it a condition of purchase of one of their spare parts that you can only use them in a BMW, that would be anti-competitive. Which of course it wouldn't. BMW is absolutely free to refuse to make spare parts for Mercedes, Honda, or whatever other cars.
Please, it is _not_ "virii". There's nothing worse then someone trying to be clever and failing miserably. "virus" is not a latin word. And if it was a latin word, the plural would be "viri" and not "virii". "virii" would be the plural of "virius", and there is no such word.
'' I've heard that Apple is claiming that the new Core2 Duo is 7 times faster than the old "top of the line" 1.67Ghz PowerBook. But comparing the XBench output that was posted earlier to an XBench run that I just ran (see below my signature) shows only a 2x increase in almost every single category (there was one or two that were about 2.5 times higher). ''
I thought people would know by now that XBench is about the most useless "benchmark" around.
There is the fact that they measure the speed of the user interface by changing the title of a single button as fast as possible - unfortunately, the OS very sensibly limits this to 60 frames per second or whatever the display rate is, so this code will _never_ get faster, no matter how fast the hardware is.
Then the benchmark is single threaded. Guess what effect two or four processors have on this benchmark: None. In the real world, they double or quadruple the speed of speed critical code. The floating-point code uses a single precision FFT implementation that was written to highlight the capabilities of Altivec but never used by anyone in the real world. Let's say it wasn't high on the priority list when rewriting it for x86.
Current MacOS X versions that you can buy in a retail box are only for PowerPC. Apple doesn't sell an Intel version of MacOS X 10.4 at the moment, and probably never will.
Now on the box or somewhere in a box it states that you are only allowed to install the software on an Apple labelled computer. Whether this is legally binding or not is debatable. The fact however is, that 99.9999% of all retail copies of MacOS X _will_ in fact be installed on an Apple-labelled computer. Maybe there are three hackers in the world who built their own Macintosh compatible hardware from scratch, and a few old PowerComputing or Motorola Mac clones might still be alive, but Apple doesn't worry about those very much.
When MacOS X 10.5 is released, the situation will be different. There will be a few hundred million computers out there on which MacOS X 10.5 could technically be installed, possibly with a bit of hacking, so the writing on the package and the license agreement will be different.
'' You've got it backwards. Your argument only holds if the filesystem allocates space at the start of the drive and works its way to the end without leaving gaps when it fact it is optimized to do precisely the opposite. Files are dispersed across the whole drive in order to give them room to grow without fragmenting them. ''
'' The one-button mouse is crap. The "hack" of emulating a second button by putting two fingers on the trackpad and using a third to click the button is not only hugely uncomfortable, but it also illustrates the desire of the mac-using community to have a second button! ''
Context-click = tap on the trackpad with two fingers.
Slashdot Mirror
Your argument is not entirely correct.
Let me summarize first: Company A writes code X that they want to keep proprietary. The code is stolen and incorporated into some open source software Y that is distributed under the GPL. Company A receives a copy of that software Y, adds code Z, and redistributes the changed software under the GPL, not knowing that their own code X in stolen form is part of the software. Then they detect that their own code X had been included in the software illegally.
First, that code X has never been distributed under the GPL so far. The whole package (X + Y + Z) always consisted of two parts Y and Z that are distributed under the GPL, and one part X that is the property of A and has been distributed illegally (although most distributors were not aware of this). A has no obligation to distribute the source for X at all.
Now the situation is obviously quite unfortunate. If A finds out who stole the software, they can sue them for damages. In the SCO vs IBM case, if SCO could demonstrate that IBM stole the code, they would receive damages from IBM, and IBM is rich enough to pay - if some bankrupt programmer stole the code, A will not be able to get damages. Whether anyone distributing so far innocently can be sued for damages, I don't know. But what is the next step A should do?
There are several choices: A can stop distributing the code, and tell all other distributors, like Redhat, SuSe to stop distributing it as well. They'd have to stop immediately, otherwise damages have to be paid. This is obviously very painful for everyone involved, distributors and users alike. Second, A can give up and distribute the software under GPL, which is painful for A only and good for everyone else - however, A would still be entitled to full damages. Third, A could negotiate with say Redhat and SuSe. A could say: I stop distributing my version right now, which costs me money. I'll give you three days to think about this and I won't sue anyone who distributes within the next three days, but after that I want a cash offer for distributing X under GPL, otherwise I'll force everyone to stop distributing it, as I am entitled to. (Note that anyone who would distribute X + Y + Z in the next three days would be save from being sued by A, but would be in violation of the GPL. This is not A's fault. )
What A can not do: Continue to distribute (X + Y + Z) without licensing X under the GPL.
'' A problem with obviousness is that because inventions rely on the laws of nature, in hindsight they are always logic. The reward should be for recognizing something that not the ordinary person skilled in the art would recognize. Now, try to make that objective! So, a gray area is bound to remain. ''
Given that your average patent nowadays is written in such a way to make it impossible to glean any useful information off it, one could just measure the amount of effort that it took a so-called "infringer" to come up with the "invention" itself. Many patents that have been granted are on a level that I would expect to ask someone in a job interview "how would you do this" and expect some reasonable answer. Software patents have been issued for things that I would be expected to develop in the normal course of my job.
Some court has come up with a ridiculous measure for obviousness: If it is obvious, then somebody should have found and published it, right? Of course not. If it is obviousness, then nobody should ever have bothered publishing it.
'' I had a calculus professor who thought calculators were a Commie plot to inhibit the minds of Americans. ''
More likely you had a calculus professor who claimed that calculators were a Commie plot to inhibit the minds of Americans, and you were just incapable of detecting humor and/or sarcasm.
finds this gem:
n ewsid_4476100/4476105.stm
http://news.bbc.co.uk/cbbcnews/hi/newsid_4470000/
Apparently filming kids secretly in a classroom is "underhanded" but not illegal; teachers seem to be unhappy about film material demonstrating that they lost control completely. I can't find anything that says whether open filming would be legal or illegal.
'' Kids who deliberately provoke a teacher to film the results don't need to be yelled at so much as slapped around a little. And that's why I'd be a terrible teacher. ''
You would lose your job as a teacher rather quickly. Doesn't mean you would be a bad teacher.
My suggestion: Install video cameras in all classrooms. The teachers can film everything, edit to their hearts content, and publish the best bits on YouTube. Must be great fun seeing thousands of creepy little morons being shown to be creepy little morons for everyone to see. Like the bloody idiot who wrote the headline for this submission and couldn't even get the spelling of "relevant" right.
'' God Bless America, and thank God I live here. ''
Ad 1: America certainly needs it.
Ad 2: Thank God you are living there.
'' Last time I checked on the iTunes Store, songs were 0.99$, not 2.80$. Where is that 2.80$ price coming from? ''
Just a suggestion: First, the $0.99 is irrelevant. The damage to the recording industry is not $0.99. It is estimated that Apple pays $0.70 per song to the recording industry, $0.29 is for Apple to run the store, supply the music, make a bit of profit. Since you didn't buy from iTMS, Apple can't really complain (they could complain if you found a way of downloading from the iTMS without paying), so the damage is $0.70. Multiply by four as a reasonable estimate that a song that you offer for download will be download four times on the average.
The $700 minimal damages are completely unreasonable, because that would require 1000 downloads. Now if you made an illegal copy of the latest U2 album before it is released, and started selling records based on that copy before U2's record company starts selling, then I wouldn't say that the upper range of $150,000 for unproven damages couldn't be appropriate (and the record company would be free to find evidence that actual damages are even higher), but $700 for uploading a song that is available for $0.99 everywhere is unreasonable.
'' Things here in the UK always seem to be thought of as failing or disaster before they're completed. I'm sure we hate success as a nation. We also have a huge obsession with celebrity and magazines that publish how fat celebrities are, or how their lives are in a mess always do very well.
I say wait until the project's finished before kicking it to the ground. ''
Ok, lets start then by kicking the CSA (Child Support Agency) to the ground first, followed by whoever is responsible for the passport debacle.
If we have to wait until a project is finished, we will likely _never_ have the right to complain about the NHS projects.
The submitter calls the Zune an "iPod wannabe-killer".
That would make it something that kills iPod wannabe's, like Creative or Rio or Sandisk players. I wonder whether that is what the submitter meant, or did he mean "wannabe iPod" or "wannabe iPod-killer"? And I wonder what Microsoft's goal is?
'' This machine possibly made 1 error out of a total of 36 votes. Since we don't know what actually caused the error, it is possible that this machine would make the same error for every 1/36 votes. ''
If the guy is saying the truth, then all we know is that _at least_ one vote was not counted correctly. Since 18 votes were counted for candidates A and B, you would need either 19 people who state that they voted for A, or 19 people who stated they voted for B, to _prove_ that another vote was counted wrong, but we don't have those 19 people.
So the correct version is: For one of the 36 votes cast we know whether it was right or wrong, and that vote was counted wrong. For the other 35 votes, we have no information. So _all_ votes we can verify were counted wrong. Those that we cannot verify, we don't know anything about, but we sure would want to know.
'' If you live in the UK, what happens if you submit a story linking to a UK website that you do not own, the server can't handle the resulting traffic, and the owner doesn't appreciate the attention? ''
Just try to think about this logically. Lets say I have a web page; I want people to read it (that can be assumed safely, because otherwise I wouldn't have a web page), and my ISP is fine with 1000 page hits per day, but not more. Instead of 100 page hits I get 1000 hits because of a Slashdot article, and then the server crashes. Obviously, my web page has been seen ten times more often than it would have been seen without the Slashdot article, so I would have a very hard time of convincing anyone that any harm has been done.
'' Oh and those Singapore immigration landing cards are a hoot, with large red friendly letters "possession of drugs is punished by DEATH'. ''
I think it is in Malta where they have a sign that importing drugs is illegal, and if you pass _this_ sign then it is considered importing, and they have a bin right in front of the sign where you can drop everything that you don't want to import. Very sensible approach.
'' The user agreement for virtually all ISPs does not allow their users to share their internet connection wirelessly, no matter how generous your neighbors feel. ''
I checked the terms and conditions for Orange, former Wanadoo. Quite interesting. I am only allowed to use my connection for residential use. I think that allowing my neighbour free access is clearly residential use. I am explicitely not allowed to resell the connection. So allowing my neighbour access for money is disallowed. Commercial use is not allowed. This is a change in their terms; it used to be that the service was "not intended for commercial use", which I assume means that if my business goes bankrupt because my broadband connection fails, that's my problem and not theirs. Then there is "fair usage"; sharing the connection obviously makes it easier to exceed whatever "fair usage" means.
Most importantly: I am sure that Orange terms allow me to make broadband available to others for free as long as they don't use too much bandwidth.
'' Is this really a surprise? This is the same country where you can be caned for simple acts of vandalism. ''
Many people believe that this kind of punishment is very appropriate. I have much less sympathy for a brainless idiot who thinks it is fun to scratch a new car than for someone robbing a bank.
Right now, spam goes past spam filters by including a large amount of random nonsense text that resembles English language reasonably well. So we will get spam filters that detect large amounts of random nonsense text. So spam will include text that makes actual sense. Give it twenty years, and your average spam email will consist of 300 pages of text that is better than anything Shakespeare has ever written, followed by two lines begging you to buy viagra. Thirty years, spam will be two hour Quicktime movies better than anything you can watch in the cinema today, with the hero using viagra bought from the spammer in the right places.
''Are the binaries encrypted, or just signed? '' Encrypted.
"Does the hardware have a public key hardwired into it, and if so, can someone just extract that key from a particular mac, for everybody else to use?" Some part of the decryption is hardwired in Macintosh hardware. I don't know if it can be extracted, but by doing so you move from the area of civil law (copyright) to criminal law (DMCA).
''Can Apple's mechanism be used to forbid people from running software that doesn't come from a vendor that's registered itself with Apple? '' Such things can be done, but the use of a decryption key that is available on Macs and nowhere else doesn't help.
"Are the components we're talking about open-source, or not?" Not.
AES-128 is one of the more expensive codes to decrypt, but with a bit of effort a single core on a Core Duo chip can do it at a speed that is quite comparable to harddisk read speed. And since I haven't read any complaints from Mac users how slow these machines are booting, I think there is no reason to worry about that.
'' The moment I ask "how much?" and they tell me, they have entered into a contract to supply the goods at that price or a lower price if they can't make change. '' Of course "they" haven't. Where do you get that nonsense from?
Wrong analogy. What you mean is: If BMW was to make it a condition of purchase of one of their spare parts that you can only use them in a BMW, that would be anti-competitive. Which of course it wouldn't. BMW is absolutely free to refuse to make spare parts for Mercedes, Honda, or whatever other cars.
Please, it is _not_ "virii". There's nothing worse then someone trying to be clever and failing miserably. "virus" is not a latin word. And if it was a latin word, the plural would be "viri" and not "virii". "virii" would be the plural of "virius", and there is no such word.
'' I've heard that Apple is claiming that the new Core2 Duo is 7 times faster than the old "top of the line" 1.67Ghz PowerBook. But comparing the XBench output that was posted earlier to an XBench run that I just ran (see below my signature) shows only a 2x increase in almost every single category (there was one or two that were about 2.5 times higher). ''
I thought people would know by now that XBench is about the most useless "benchmark" around.
There is the fact that they measure the speed of the user interface by changing the title of a single button as fast as possible - unfortunately, the OS very sensibly limits this to 60 frames per second or whatever the display rate is, so this code will _never_ get faster, no matter how fast the hardware is.
Then the benchmark is single threaded. Guess what effect two or four processors have on this benchmark: None. In the real world, they double or quadruple the speed of speed critical code. The floating-point code uses a single precision FFT implementation that was written to highlight the capabilities of Altivec but never used by anyone in the real world. Let's say it wasn't high on the priority list when rewriting it for x86.
Current MacOS X versions that you can buy in a retail box are only for PowerPC. Apple doesn't sell an Intel version of MacOS X 10.4 at the moment, and probably never will.
Now on the box or somewhere in a box it states that you are only allowed to install the software on an Apple labelled computer. Whether this is legally binding or not is debatable. The fact however is, that 99.9999% of all retail copies of MacOS X _will_ in fact be installed on an Apple-labelled computer. Maybe there are three hackers in the world who built their own Macintosh compatible hardware from scratch, and a few old PowerComputing or Motorola Mac clones might still be alive, but Apple doesn't worry about those very much.
When MacOS X 10.5 is released, the situation will be different. There will be a few hundred million computers out there on which MacOS X 10.5 could technically be installed, possibly with a bit of hacking, so the writing on the package and the license agreement will be different.
'' You've got it backwards. Your argument only holds if the filesystem allocates space at the start of the drive and works its way to the end without leaving gaps when it fact it is optimized to do precisely the opposite. Files are dispersed across the whole drive in order to give them room to grow without fragmenting them. ''
Show me an operating system that does that.
'' The one-button mouse is crap. The "hack" of emulating a second button by putting two fingers on the trackpad and using a third to click the button is not only hugely uncomfortable, but it also illustrates the desire of the mac-using community to have a second button! ''
Context-click = tap on the trackpad with two fingers.