"Well sure, but I expect that of any modern operating system."
As far as expectations go, many users might think that Windows compatibility is something they expect from a modern operating system (however impractical that might be).
Creating a stable OS on a 386 or later Intel processor isn't that much of a challenge and MS could have done it easily. Doing it while maintaining backward-compatibility with applications that were designed to run on a 8088-based Windows version is the real challenge.
"For the most part the useful changes in Windows are in the user interface."
I disagree. The most significant change between Windows 9x/ME and Windows NT/2000/XP is the stability gained by dropping real-mode code. Buggy applications that caused a BSOD on Windows 9x will only crash themselves on current versions.
The fact is that people who are still running win98 and win95 are not really a factor in the development of new applications and are unlikely to buy them.
Eventually their machines will die and it will be too expensive to repair them and they'll have to buy new ones. If this happens 2 or 3 years from now, they'll most likely end up with longhorn.
"There was absolutely high demand for this type of application in 1995, and even more so now."
Most individuals and businesses were just learning about what the Internet was in 1995, there was no significant demand for browser apps in those days.
One big difference between a product like this and a real book is its attractiveness to thieves. If you leave a book behind and return an hour later to pick it up, chances are it'll still be there.
A device like this is more likely to be stolen even if the thief doesn't know what it is. So these devices end up being just another burden to worry about.
You don't quote anything that says that copyright is automatic, only that the ommision of a notice doesn't result in the loss of copyright protection for works published on or after March 1, 1989. This doesn't appear to exclude the possibility of public domain documents that aren't explicitly declared as such. So I think intent is still key.
"Kind of, you must have intent but since we signed the Berne convention you show that intent by not explicitly placing the work in the public domain."
You have me at a disadvantage since I don't know the details of the Berne convention, but I note that at least in the US, nobody counts on it to protect their IP so perhaps there's some question about its content or jurisdiction. You'd be hard-pressed to find any published book without a copyright notice unless it's a public domain work.
As a license, the GPL can be reviewed by a court. The court is not obligated to accept or reject it in its entirety but has the option to redline it if there is a legal justification for doing so.
RMS's writings may be interesting and informative but they aren't relevant in a legal sense. In court the judge will decide what the GPL means based on the parties arguments and his knowledge of the law.
The author must have the intent to copyright their work in order for copyright right law to be in force. If you copy this post and send it to a friend saying that you disagree with my argument, you're not violating copyright law even though I haven't given you a license to do so.
So the work starts out free and then if the author wants to protect it under the GPL he copyrights it and licenses it under the GPL. If he wants to protect it using some other license, he copyrights it and licenses it under that license.
The terms of each license are different, but they all restrict what the licensee is allowed to do with the work. The GPL is no exception.
So developing an application and trying to get people to pay attention to it (no other reason for putting on the web) and then offering no help with it is perfectly reasonable, but someone who indulges you by trying out and then complains about something is an "antisocial prick". Interesting.
Since RMS *is* a person who wrote an opinion, he should expect some to disagree with it. Gosling didn't make a personal attack on RMS, he just challenged some of his arguments and assumptions.
"The motivating reason for "Free" software is that software entails no per-unit reproduction cost, while hardware does."
I thought "free as in freedom" was supposed to be the motivating reason for "free" software. What does freedom have to do with per-unit cost?
In any case, many hardware devices contain software, firmware or microcode that doesn't add to any per-unit reproduction cost.
By your logic, Sun could distribute their JVM only on CD and then RMS would be satisfied because Sun could use the per-unit reproduction cost loophole.
"You realize, of course, that all software RMS writes is written in platform-independant and CPU-independant languages, right?"
First I would have to believe that there was a platform-independent, CPU-independent language before I would agree that RMS writes with one. I certainly know of CPUs that have no "C" compiler available. Even for those with a "C" compiler, there's no guarantee that RMS's code will run the same on all of them.
"Not to mention, I have no idea what you mean by 'non-free'. It's prefectly legal to design your own version of most processors out there, witness AMD, and the specifications are open."
If MS published all their Windows API's and someone cloned Windows, would that make MS Windows open source? It's not enough to publish the external specs and interface of a software or hardware object to make it "free", you have to make the internal design available.
The point is that if Stallman decides he wants to draw the line at hardware, he should let others draw the line where they want to without him complaining.
"If your program is free software, it is basically ethical" does not logically lead to "If your program is non-free software, it is basically unethical".
So the only people who should feel "hurt" by this argument are those open-source developers who were developing free software out of a desire to be unethical.
But it's not just about having open drivers. If he wants total purity he has to insist on having all the source for the software inside the printer as well.
While its true that there is a tangible per-unit cost for hardware, it's still the IP costs that often dominate in the sales price.
"As this thread progresses, I'm certain we'll find that a lot of people whine about and take cheap shots at RMS. Coincidentally, these are typically people who haven't accomplished anything useful in their entire lives except post witty one-liners and flames of others here on Slashdot."
It's a pity that you started a long comment with a ad hominem argument. The fact is that the validity of RMS's arguments or those of his critics have nothing to do with what they have or have not accomplished. Let the arguments stand on their own.
"A modelling tool is quite different to a programming language.."
Yes, but not really in this context. If IBM is going to buy something, they expect some return from their investment regardless if it's a programming language or a modeling tool.
So let's carry RMS's ideas on the purity of "free" software to their logical conclusion. If you can't have free software running on non-free virtual machines or OS's then you can't have truly free software running on non-free processors like those made by Intel, AMD, Sun, IBM, Motorola etc.
Perhaps RMS should take a few years off from his Guru work to learn how to design microprocessors so he can finally write some truly free software.
Slashdot Mirror
"Well sure, but I expect that of any modern operating system."
As far as expectations go, many users might think that Windows compatibility is something they expect from a modern operating system (however impractical that might be).
Creating a stable OS on a 386 or later Intel processor isn't that much of a challenge and MS could have done it easily. Doing it while maintaining backward-compatibility with applications that were designed to run on a 8088-based Windows version is the real challenge.
"For the most part the useful changes in Windows are in the user interface."
I disagree. The most significant change between Windows 9x/ME and Windows NT/2000/XP is the stability gained by dropping real-mode code. Buggy applications that caused a BSOD on Windows 9x will only crash themselves on current versions.
"The easiest thing for them to use is the Internet."
Yes. Typing URLs and clicking on the "Back" and "Forward" buttons is pretty easy, but doing real work is always going to be harder.
The fact is that people who are still running win98 and win95 are not really a factor in the development of new applications and are unlikely to buy them.
Eventually their machines will die and it will be too expensive to repair them and they'll have to buy new ones. If this happens 2 or 3 years from now, they'll most likely end up with longhorn.
"There was absolutely high demand for this type of application in 1995, and even more so now."
Most individuals and businesses were just learning about what the Internet was in 1995, there was no significant demand for browser apps in those days.
One big difference between a product like this and a real book is its attractiveness to thieves. If you leave a book behind and return an hour later to pick it up, chances are it'll still be there.
A device like this is more likely to be stolen even if the thief doesn't know what it is. So these devices end up being just another burden to worry about.
I might add that many 2600 developers were working from reverse-engineered specs so that made it even more difficult.
You don't quote anything that says that copyright is automatic, only that the ommision of a notice doesn't result in the loss of copyright protection for works published on or after March 1, 1989. This doesn't appear to exclude the possibility of public domain documents that aren't explicitly declared as such. So I think intent is still key.
"All works are copyright by default, unless explicity placed into the public domain."
I would need to see a specific quote within copyright law before I'd believe that.
"Kind of, you must have intent but since we signed the Berne convention you show that intent by not explicitly placing the work in the public domain."
You have me at a disadvantage since I don't know the details of the Berne convention, but I note that at least in the US, nobody counts on it to protect their IP so perhaps there's some question about its content or jurisdiction. You'd be hard-pressed to find any published book without a copyright notice unless it's a public domain work.
As a license, the GPL can be reviewed by a court. The court is not obligated to accept or reject it in its entirety but has the option to redline it if there is a legal justification for doing so.
RMS's writings may be interesting and informative but they aren't relevant in a legal sense. In court the judge will decide what the GPL means based on the parties arguments and his knowledge of the law.
That sentence should be: There's a huge difference between "free" and public domain.
That's because there's a huge difference between "free" and free.
The author must have the intent to copyright their work in order for copyright right law to be in force. If you copy this post and send it to a friend saying that you disagree with my argument, you're not violating copyright law even though I haven't given you a license to do so.
So the work starts out free and then if the author wants to protect it under the GPL he copyrights it and licenses it under the GPL. If he wants to protect it using some other license, he copyrights it and licenses it under that license.
The terms of each license are different, but they all restrict what the licensee is allowed to do with the work. The GPL is no exception.
"direct email access to the developer"
Let me guess, the email address is MakeMeAnAntisocialPrick@IgnoreThisEmail.com.
So developing an application and trying to get people to pay attention to it (no other reason for putting on the web) and then offering no help with it is perfectly reasonable, but someone who indulges you by trying out and then complains about something is an "antisocial prick". Interesting.
Bill picked the first letter of the name.
Since RMS *is* a person who wrote an opinion, he should expect some to disagree with it. Gosling didn't make a personal attack on RMS, he just challenged some of his arguments and assumptions.
"The motivating reason for "Free" software is that software entails no per-unit reproduction cost, while hardware does."
I thought "free as in freedom" was supposed to be the motivating reason for "free" software. What does freedom have to do with per-unit cost?
In any case, many hardware devices contain software, firmware or microcode that doesn't add to any per-unit reproduction cost.
By your logic, Sun could distribute their JVM only on CD and then RMS would be satisfied because Sun could use the per-unit reproduction cost loophole.
"You realize, of course, that all software RMS writes is written in platform-independant and CPU-independant languages, right?"
First I would have to believe that there was a platform-independent, CPU-independent language before I would agree that RMS writes with one. I certainly know of CPUs that have no "C" compiler available. Even for those with a "C" compiler, there's no guarantee that RMS's code will run the same on all of them.
"Not to mention, I have no idea what you mean by 'non-free'. It's prefectly legal to design your own version of most processors out there, witness AMD, and the specifications are open."
If MS published all their Windows API's and someone cloned Windows, would that make MS Windows open source? It's not enough to publish the external specs and interface of a software or hardware object to make it "free", you have to make the internal design available.
The point is that if Stallman decides he wants to draw the line at hardware, he should let others draw the line where they want to without him complaining.
"If your program is free software, it is basically ethical" does not logically lead to "If your program is non-free software, it is basically unethical".
So the only people who should feel "hurt" by this argument are those open-source developers who were developing free software out of a desire to be unethical.
But it's not just about having open drivers. If he wants total purity he has to insist on having all the source for the software inside the printer as well.
While its true that there is a tangible per-unit cost for hardware, it's still the IP costs that often dominate in the sales price.
"As this thread progresses, I'm certain we'll find that a lot of people whine about and take cheap shots at RMS. Coincidentally, these are typically people who haven't accomplished anything useful in their entire lives except post witty one-liners and flames of others here on Slashdot."
It's a pity that you started a long comment with a ad hominem argument. The fact is that the validity of RMS's arguments or those of his critics have nothing to do with what they have or have not accomplished. Let the arguments stand on their own.
"A modelling tool is quite different to a programming language.."
Yes, but not really in this context. If IBM is going to buy something, they expect some return from their investment regardless if it's a programming language or a modeling tool.
So let's carry RMS's ideas on the purity of "free" software to their logical conclusion. If you can't have free software running on non-free virtual machines or OS's then you can't have truly free software running on non-free processors like those made by Intel, AMD, Sun, IBM, Motorola etc.
Perhaps RMS should take a few years off from his Guru work to learn how to design microprocessors so he can finally write some truly free software.