Yeah right. If you're trying to make a file manager suitable for the masses you had better make it run tolerably on the masses' hardware. They all have IDE and could care less about what you think.
My dictionary has 17 definitions of "free", one of which corresponds to "free as in beer" and one corresponds to "free as in speech". Most of the other fifteen don't correspond to either (free end of a rope, free electron, free verse, etc.).
Two group them into broad categories, there are three common meanings: free from obligation (free beer); free from external authority (free speech); and unrestricted or unhindered (free use).
"Free as in speech" means "allowed to do whatever you want with it".
Someone should arrest and jail your High School civics teacher for fraud.
"Free Speech" means speech that is unrestricted. In the context of western civilization is usually refers to the legal right to speak without legal hinderances, particularly with regards to political opinion, though it also covers non-political expressions of a creative, commercial or mundane nature.
"Free Software" may indeed mean "allowed to do (almost) whatever you want with it". But that is NOT what "Free Speech" means.
In regards to software, there are two kinds of "speech" available. The first is the "speech" of actually creating the software. The second is the "speech" of modifying and/or redistributing the first kind. The first kind is a legal and unalienable right. Unfortunately, the second is not an unalienable right, but a granted priviledge. Despite the existance of the first ammendment to the US Constitution, you are not allowed to take the political writings of Richard Stallman and modify them for redistribution. You are not allowed to take the New York Times and republish it as the Yonkers Yodeler.
The right to your own speech is unalienable and protected by law. The right to someone else's speech is not. By equating Free Software with Free Speech, you are asserting that the ability to modify and redistribute someone else's creative works is an unalienable right that should be enforced by law. But that is not how rights work. A right is something that you POSSESS. It is not something that you TAKE from someone else.
I wonder how it's possible to write a "Linux Emulation Layer" without using the Linux source in a way that violates the GPL.
The GPL only restricts the distribution of the original and derivative works (albeit with a very liberal interpretation of "derivative"). There is ZERO restriction on looking at the code and seeing how it works. There is ZERO restriction on 'reverse engineering' it. Writing an Linux emulation layer without violating the GPL would be trivial (but tedious) for anyone familiar with kernel programming.
It has to have some stuff behave exactly the same, or just wouldn't work.
Linux is under copyright but it is not under patent. That means that you can legally duplicate how the stuff is done.
...operating systems that are not Free as in speech.
It's statements like that which make the rest of the world suspect we are a brainwashed cult just waiting for the Koolaid to arrive.
The right to free speech is a right that belongs to the speaker, and not to the listener. Linux is a manifestation of Linus Torvald's free speech. SkyOS is a manifestation of its authors' free speech. Any OS is a product of its authors' free speech. You have your own right of free speech but it is not predicated upon the existance of any operating system.
It would make sense to say that you only want operating systems that are "Free Software", or operating systems that are "Open Source", or operating systems that grant you specific permissions. But to say that you only want operating systems that are "free as in speech" is a non sequitur. It only makes sense if you have undergone the GNU Indoctrination Protocols, as it is not a sensible English phrase.
(considering that most Slashdot readers are openly hostile toward non-Linux Free Software operating systems, like OpenBSD, I think the original supposition was a correct evaluation of the Slashdot attitudes)
There is no need for this. Voice recognition already works. And it works well. And it already works with REAL words. No need to grunt, squeal of burp into your microphone.
I first used voice recognition software with OS/2 4.0 on a P100 with 16MB. I was amazed at how well it worked. Of course, 16Mb was inadequate for dictation, but even with that puny system I had it trained in half and hour.
There's a reason that voice recognition hasn't caught on. It's not because it doesn't work. It's because people don't want to talk to their computers. It's embarassing. It's not convenient. It's awkward to say those commands that computers need, like "arrem minus arref slash star".
For the sake on folks thinking of choosing FreeBSD and OpenBSD to learn *nix, PLEASE REPLY with your thoughts on the future of FreeBSD.
What systems are going to be popular five, ten, twenty years from now? Nobody knows. So don't learn a specific system. Learn the general concepts and skills that WILL be useful in five, ten or twenty years.
My opinion is that as long as there are Unix and unix-like operating systems, there will be a BSD.
Am I wasting my time to learn it if I want a job or to run a home server or to develop software for some companies server "back end"?
If I had to hire one systems administrator and there were two applicants, one who knew only Linux and the other who know both Linux and FreeBSD, I would choose the second one every time. The last kind of person I would want to hire is someone who only knew one way of doing things.
So learn both systems. They're free. In the process you will discover which one you prefer. Make that one your primary system. But keep the other around on another partition. Having immediate access to more than one operating system is invaluable when it comes to software development. As for myself, I'm using both Slackware and FreeBSD, and after two years I still can't decide between the two.
I'm a programmer getting into *nix -- but not sure which one.
So don't get into "one" *nix, get into *nix itself. Generalists always have an advantage over specialists in times of change. And as history teaches, change is constant:-)
Those "better programmers" you talk about are all in the past. Ignore them. They are as irrelevant to your daily needs as are Babbage, Turing, Hopper, Shannon and Ritchie. Don't trust anyone over thirty.
"Want to know why adding programmers to a late software project makes it later?" I'm a programmer not a project manager and don't want to be a project manager, ever.
Just wait until they make you be the late programmer on someone else's project.
"Are there some problems which are not solvable by computer?" I never encounter those problems, thank Heaven.
And when you finally run across one you'll spend months banging your head against the wall.
"How does entropy and communications work?" Don't need to know in order to copy a file.
Now your ignorance is really showing.
"Check out the history of AI research." It's mostly a failure.
AI is a failure. The stuff that came out of AI research is not.
"Does Babbage's engine really work?" Not as well as my PC.
Oh puh'leaze! Somewhere there is someone who is a better programmer than you. So it's time to toss yourself into the dustbin of history.
RMS started GNU to demonstrate the practicality of his beliefs, and the ``Linux'' operating system rides on that demonstration.
By that logic *any* free operating system that makes use of that "demonstration" should be called "GNU". Let's see now, GNU/FreeBSD, GNU/OpenBSD, GNU/AtheOS, etc. All use software from GNU to one extent or another.
Why, then, do so many people not want to let RMS point that out?
No one, and I repeat, no one is denying that RMS and his projects have contributed greatly to the OS formerly known as "LiGNuX" (his previous monicker for it). Every distro credits RMS, FSF and GNU. The problem isn't attribution, because RMS is getting all the attribution he deserves, and more. The problem is that the wants to name the whole of which he is only a part.
And, for a better question, why would Sun want to pay TT for a licence for QT?
Why would anyone want to pay Sun for a license to use Solaris? Tit-for-tat.
Why would any company want to pay for a widget set to develop (closed-source, mind you) for Linux?
Nobody wants to pay for anything. I don't want to pay for lunch. I don't want to pay my rent. I don't want to pay for gas in my car. But I seldom get my wishes fulfilled.
Trolltech requires payment for Qt *only* if you in turn require payment for the software that uses it.
There is only one GNU System, and so far, only Debian GNU/Hurd qualifies. No Linux based systems that I am aware of qualify as The GNU System. The GNU components in the prototypical Linux distribution do not count as an operating system, but as the major portion of the operating environment for that operating system.
The description of The GNU System in the GNU Manifesto makes it clear that it is an operating system AND environment. LinuxOS only used the environment portion of it, and not all of it, and not in isolation.
Re:A great example of an RMS witch-hunt
on
Five Years of KDE
·
· Score: 2
The code used for KGhostview was under the copyright of Aladdin, and not the FSF. The only other GPLd code incorporated into KDE was in KFloppy, which was copyrighted by Linus Torvalds.
have the GCC people thought of any ways to make binary compatibility easier to maintain in C++ programs without requiring vtable hacks and other ugliness?
It's more than just the C++ issue. You change the C++ interface to a class and you lose backward compatibility. But the very same thing happens if you change the interface to a C library. Just because it's easier to mangle [sic] the interfaces with C++, doesn't mean that C is immune to it.
Re:A great example of an RMS witch-hunt
on
Five Years of KDE
·
· Score: 2
The restrictions in place in the GPL do not apply to the copyright holder. This means that the KDE team, as owners of KDE, can do any damn thing they want with KDE. Including granting permission to distribute KDE binaries.
Their webpage said I could distribute KDE. Their FAQ said I could distribute KDE. They listed mirrors and distros they recommended to get KDE binaries. I asked a KDE member if it was okay to distribute KDE (reply: "duh!"). Even their licenses said I could distribute KDE.
The terms of distribution may not have been clear to you and others, but they were obvious to the rest of us.
RMS simply said that that old KDE binary distributions were OK anyways at least for the parts of KDE that used/linked/included GPL code property of FSF
No code that was the property of the FSF and was ever used in KDE, and no code that was the property of the FSF and licensed under an exception-less GPL was ever linked to by KDE. RMS knew this. Thus is grant of forgiveness to the KDE team was insulting.
C++ is deeply ingrained in the system; I don't believe that's where the future of application programming is going.
If you want applications compiled natively for your processor, then C and C++ are your only realistic options. Every other common programming language is interpreted. Now if there were *standard* native compilers for Python and Java, I would switch in a heartbeat. But I don't see them coming, so I'll stick with C/C++ for now. Semantically, Java is clearly superior. But the real world demands more than semantics.
A lot of KDE just duplicates existing functionality, but using the Qt toolkit and KDE libraries, all in the name of KDE integration.
It's that integration that makes it worthwhile. Konqueror is the best web browser out there, IMO, but it would never exist if the KDE developers decided not to duplicate existing functionality.
Take away KDE and you *STILL* have duplicated functionality. Take away GNOME and Xfce and you *STILL* have duplicated functionality. Take way Windowmaker, Blackbox, IceWM and Enlightenment and you still have two score window managers. Take way joe, jed, vim, and emacs and you *STILL* have two dozen text editors.
What KDE has offered up is a suite of tools, utilities and applications that are integrated with a common look and feel. That's a valuable asset right there.
KDE consumes huge amounts of resources and starts up lots of processes.
The resources aren't that bad at all. If you have a P100 with 16M RAM, then by all means stick with twm and rxvt. Actually, I find that I'm using LESS resources with KDE-2.2.1 than I did with KDE-1.1.2. Why? Because it has Konqueror now so I don't have to load up that behemoth known as Netscape.
The KDE/Qt licenses (GPL/commercial) restrict my ability to create open source software (say, under BSD or LGPL licenses).
Hmmm, I write BSD licensed software that uses Qt. Should I be on the lookout for the police? Is my freedom in jeopardy?
Qt is not licensed GPL/Commercial. It's licensed GPL/QPL/Commercial. That means you can write software with ANY open source license. You only have to pay if you write proprietary software.
KDE is replicating an old paradigm--the Windows desktop
Hah! If anything, KDE is closer to the OS/2 Workplace Shell than the Microsoft Start Button desktop.
I'm not sure it can compete with Windows
Who cares? KDE isn't for the DOS or NT operating systems. It's for Unix operating systems with X11. Different domains. Wanting KDE to compete with Windows is like wanting your Dodge Ram Truck to compete with a Ford Escort.
Yes, which basically means that I have to use the GPL for any open source software I write based on it--that's too restrictive. I want to let people use my open source software under BSD or LGPL licenses.
Qt is dual licensed under the GPL -and- the QPL. You aren't allowed to write any proprietary software using Qt (like you are with the GNU products), but you are allowed to write software licensed under any Open Source license.
Actually, Microsoft and other systems are increasingly going over to browser-like interfaces, and that's likely where the future lies.
Do I have to go to the future, or can I stay here? I've used browser-like interfaces and they suck rocks. It may indeed be the future, but not because human beings wants it, but because the subhuman creatures in marketing will think it's cool.
Re:A great example of an RMS witch-hunt
on
Five Years of KDE
·
· Score: 2
Precisely where in the GPL does it mandate a grant of forgiveness before someone can distribute their own code?
I'm not a member of the KDE "tribe", but I do prefer KDE. Then Redhat came out with a white paper saying that distributing KDE was illegal. Redhat called me a criminal! I was only doing what RMS told me to do. You see, RMS said that if a friend asked for a copy of software, it would be morally wrong to refuse. A friend asked me for a copy of KDE and I gave him one. I distributed KDE. Evil me.
According to Redhat, I was a criminal. Debian and GNU agreed. I may not be a member of any "tribe", but it's clear to me which tribes only talk about freedom, and which ones actually practice it.
Re:A great example of open-source at work.
on
Five Years of KDE
·
· Score: 2
And after 5 years I'll bet most KDE users are using pretty much the latest version. Nearly 7 years after its release it is not odd to find Windows 95 still in use
Compare the costs/benefits of upgrading KDE-1.0 to KDE-2.2.1. Now compare the costs/benefits of upgrading Windows95 to WindowsME.
Frankly, for zero cost I get a wealth of new features and stability with KDE-2.2.1, but I have to spend $100 to get the miserly offerings of WindowsME. No contest. I know several people who still use Windows95. Their stated reason is simply that they see no compelling reason to upgrade.
What percentage of Windows users actually chose to upgrade from Window95 to 98 to ME? A very low percentage. Most Win98 users got theirs with their computer. Most WinME users got theirs with their computer.
(and of course, the demographics are completely different from Unix to Win/DOS users, and also for Win/NT, which is why you find a more NT users upgrading to 2K than you do W98 users to WME)
Slashdot Mirror
Is anyone else having problems posting to slashdot, or am I just losing my mind? And why do I need a sourceforge account to log a slashdot bug?
Yeah right. If you're trying to make a file manager suitable for the masses you had better make it run tolerably on the masses' hardware. They all have IDE and could care less about what you think.
Free as in Legos
which, of course, means free to do whatever we want with the building blocks...
I love it! It gets precisely the right point across.
My dictionary has 17 definitions of "free", one of which corresponds to "free as in beer" and one corresponds to "free as in speech". Most of the other fifteen don't correspond to either (free end of a rope, free electron, free verse, etc.).
Two group them into broad categories, there are three common meanings: free from obligation (free beer); free from external authority (free speech); and unrestricted or unhindered (free use).
"Free as in speech" means "allowed to do whatever you want with it".
Someone should arrest and jail your High School civics teacher for fraud.
"Free Speech" means speech that is unrestricted. In the context of western civilization is usually refers to the legal right to speak without legal hinderances, particularly with regards to political opinion, though it also covers non-political expressions of a creative, commercial or mundane nature.
"Free Software" may indeed mean "allowed to do (almost) whatever you want with it". But that is NOT what "Free Speech" means.
In regards to software, there are two kinds of "speech" available. The first is the "speech" of actually creating the software. The second is the "speech" of modifying and/or redistributing the first kind. The first kind is a legal and unalienable right. Unfortunately, the second is not an unalienable right, but a granted priviledge. Despite the existance of the first ammendment to the US Constitution, you are not allowed to take the political writings of Richard Stallman and modify them for redistribution. You are not allowed to take the New York Times and republish it as the Yonkers Yodeler.
The right to your own speech is unalienable and protected by law. The right to someone else's speech is not. By equating Free Software with Free Speech, you are asserting that the ability to modify and redistribute someone else's creative works is an unalienable right that should be enforced by law. But that is not how rights work. A right is something that you POSSESS. It is not something that you TAKE from someone else.
I wonder how it's possible to write a "Linux Emulation Layer" without using the Linux source in a way that violates the GPL.
The GPL only restricts the distribution of the original and derivative works (albeit with a very liberal interpretation of "derivative"). There is ZERO restriction on looking at the code and seeing how it works. There is ZERO restriction on 'reverse engineering' it. Writing an Linux emulation layer without violating the GPL would be trivial (but tedious) for anyone familiar with kernel programming.
It has to have some stuff behave exactly the same, or just wouldn't work.
Linux is under copyright but it is not under patent. That means that you can legally duplicate how the stuff is done.
...operating systems that are not Free as in speech.
It's statements like that which make the rest of the world suspect we are a brainwashed cult just waiting for the Koolaid to arrive.
The right to free speech is a right that belongs to the speaker, and not to the listener. Linux is a manifestation of Linus Torvald's free speech. SkyOS is a manifestation of its authors' free speech. Any OS is a product of its authors' free speech. You have your own right of free speech but it is not predicated upon the existance of any operating system.
It would make sense to say that you only want operating systems that are "Free Software", or operating systems that are "Open Source", or operating systems that grant you specific permissions. But to say that you only want operating systems that are "free as in speech" is a non sequitur. It only makes sense if you have undergone the GNU Indoctrination Protocols, as it is not a sensible English phrase.
(considering that most Slashdot readers are openly hostile toward non-Linux Free Software operating systems, like OpenBSD, I think the original supposition was a correct evaluation of the Slashdot attitudes)
There is no need for this. Voice recognition already works. And it works well. And it already works with REAL words. No need to grunt, squeal of burp into your microphone.
I first used voice recognition software with OS/2 4.0 on a P100 with 16MB. I was amazed at how well it worked. Of course, 16Mb was inadequate for dictation, but even with that puny system I had it trained in half and hour.
There's a reason that voice recognition hasn't caught on. It's not because it doesn't work. It's because people don't want to talk to their computers. It's embarassing. It's not convenient. It's awkward to say those commands that computers need, like "arrem minus arref slash star".
For the sake on folks thinking of choosing FreeBSD and OpenBSD to learn *nix, PLEASE REPLY with your thoughts on the future of FreeBSD.
:-)
What systems are going to be popular five, ten, twenty years from now? Nobody knows. So don't learn a specific system. Learn the general concepts and skills that WILL be useful in five, ten or twenty years.
My opinion is that as long as there are Unix and unix-like operating systems, there will be a BSD.
Am I wasting my time to learn it if I want a job or to run a home server or to develop software for some companies server "back end"?
If I had to hire one systems administrator and there were two applicants, one who knew only Linux and the other who know both Linux and FreeBSD, I would choose the second one every time. The last kind of person I would want to hire is someone who only knew one way of doing things.
So learn both systems. They're free. In the process you will discover which one you prefer. Make that one your primary system. But keep the other around on another partition. Having immediate access to more than one operating system is invaluable when it comes to software development. As for myself, I'm using both Slackware and FreeBSD, and after two years I still can't decide between the two.
I'm a programmer getting into *nix -- but not sure which one.
So don't get into "one" *nix, get into *nix itself. Generalists always have an advantage over specialists in times of change. And as history teaches, change is constant
Those "better programmers" you talk about are all in the past. Ignore them. They are as irrelevant to your daily needs as are Babbage, Turing, Hopper, Shannon and Ritchie. Don't trust anyone over thirty.
"Want to know why adding programmers to a late software project makes it later?" I'm a programmer not a project manager and don't want to be a project manager, ever.
Just wait until they make you be the late programmer on someone else's project.
"Are there some problems which are not solvable by computer?" I never encounter those problems, thank Heaven.
And when you finally run across one you'll spend months banging your head against the wall.
"How does entropy and communications work?" Don't need to know in order to copy a file.
Now your ignorance is really showing.
"Check out the history of AI research." It's mostly a failure.
AI is a failure. The stuff that came out of AI research is not.
"Does Babbage's engine really work?" Not as well as my PC.
Oh puh'leaze! Somewhere there is someone who is a better programmer than you. So it's time to toss yourself into the dustbin of history.
RMS started GNU to demonstrate the practicality of his beliefs, and the ``Linux'' operating system rides on that demonstration.
By that logic *any* free operating system that makes use of that "demonstration" should be called "GNU". Let's see now, GNU/FreeBSD, GNU/OpenBSD, GNU/AtheOS, etc. All use software from GNU to one extent or another.
Why, then, do so many people not want to let RMS point that out?
No one, and I repeat, no one is denying that RMS and his projects have contributed greatly to the OS formerly known as "LiGNuX" (his previous monicker for it). Every distro credits RMS, FSF and GNU. The problem isn't attribution, because RMS is getting all the attribution he deserves, and more. The problem is that the wants to name the whole of which he is only a part.
And, for a better question, why would Sun want to pay TT for a licence for QT?
Why would anyone want to pay Sun for a license to use Solaris? Tit-for-tat.
Why would any company want to pay for a widget set to develop (closed-source, mind you) for Linux?
Nobody wants to pay for anything. I don't want to pay for lunch. I don't want to pay my rent. I don't want to pay for gas in my car. But I seldom get my wishes fulfilled.
Trolltech requires payment for Qt *only* if you in turn require payment for the software that uses it.
There is only one GNU System, and so far, only Debian GNU/Hurd qualifies. No Linux based systems that I am aware of qualify as The GNU System. The GNU components in the prototypical Linux distribution do not count as an operating system, but as the major portion of the operating environment for that operating system.
The description of The GNU System in the GNU Manifesto makes it clear that it is an operating system AND environment. LinuxOS only used the environment portion of it, and not all of it, and not in isolation.
The code used for KGhostview was under the copyright of Aladdin, and not the FSF. The only other GPLd code incorporated into KDE was in KFloppy, which was copyrighted by Linus Torvalds.
have the GCC people thought of any ways to make binary compatibility easier to maintain in C++ programs without requiring vtable hacks and other ugliness?
It's more than just the C++ issue. You change the C++ interface to a class and you lose backward compatibility. But the very same thing happens if you change the interface to a C library. Just because it's easier to mangle [sic] the interfaces with C++, doesn't mean that C is immune to it.
The official name is "The GNU System".
The restrictions in place in the GPL do not apply to the copyright holder. This means that the KDE team, as owners of KDE, can do any damn thing they want with KDE. Including granting permission to distribute KDE binaries.
Their webpage said I could distribute KDE. Their FAQ said I could distribute KDE. They listed mirrors and distros they recommended to get KDE binaries. I asked a KDE member if it was okay to distribute KDE (reply: "duh!"). Even their licenses said I could distribute KDE.
The terms of distribution may not have been clear to you and others, but they were obvious to the rest of us.
RMS simply said that that old KDE binary distributions were OK anyways at least for the parts of KDE that used/linked/included GPL code property of FSF
No code that was the property of the FSF and was ever used in KDE, and no code that was the property of the FSF and licensed under an exception-less GPL was ever linked to by KDE. RMS knew this. Thus is grant of forgiveness to the KDE team was insulting.
C++ is deeply ingrained in the system; I don't believe that's where the future of application programming is going.
If you want applications compiled natively for your processor, then C and C++ are your only realistic options. Every other common programming language is interpreted. Now if there were *standard* native compilers for Python and Java, I would switch in a heartbeat. But I don't see them coming, so I'll stick with C/C++ for now. Semantically, Java is clearly superior. But the real world demands more than semantics.
A lot of KDE just duplicates existing functionality, but using the Qt toolkit and KDE libraries, all in the name of KDE integration.
It's that integration that makes it worthwhile. Konqueror is the best web browser out there, IMO, but it would never exist if the KDE developers decided not to duplicate existing functionality.
Take away KDE and you *STILL* have duplicated functionality. Take away GNOME and Xfce and you *STILL* have duplicated functionality. Take way Windowmaker, Blackbox, IceWM and Enlightenment and you still have two score window managers. Take way joe, jed, vim, and emacs and you *STILL* have two dozen text editors.
What KDE has offered up is a suite of tools, utilities and applications that are integrated with a common look and feel. That's a valuable asset right there.
KDE consumes huge amounts of resources and starts up lots of processes.
The resources aren't that bad at all. If you have a P100 with 16M RAM, then by all means stick with twm and rxvt. Actually, I find that I'm using LESS resources with KDE-2.2.1 than I did with KDE-1.1.2. Why? Because it has Konqueror now so I don't have to load up that behemoth known as Netscape.
The KDE/Qt licenses (GPL/commercial) restrict my ability to create open source software (say, under BSD or LGPL licenses).
Hmmm, I write BSD licensed software that uses Qt. Should I be on the lookout for the police? Is my freedom in jeopardy?
Qt is not licensed GPL/Commercial. It's licensed GPL/QPL/Commercial. That means you can write software with ANY open source license. You only have to pay if you write proprietary software.
KDE is replicating an old paradigm--the Windows desktop
Hah! If anything, KDE is closer to the OS/2 Workplace Shell than the Microsoft Start Button desktop.
I'm not sure it can compete with Windows
Who cares? KDE isn't for the DOS or NT operating systems. It's for Unix operating systems with X11. Different domains. Wanting KDE to compete with Windows is like wanting your Dodge Ram Truck to compete with a Ford Escort.
Yes, which basically means that I have to use the GPL for any open source software I write based on it--that's too restrictive. I want to let people use my open source software under BSD or LGPL licenses.
Qt is dual licensed under the GPL -and- the QPL. You aren't allowed to write any proprietary software using Qt (like you are with the GNU products), but you are allowed to write software licensed under any Open Source license.
Actually, Microsoft and other systems are increasingly going over to browser-like interfaces, and that's likely where the future lies.
Do I have to go to the future, or can I stay here? I've used browser-like interfaces and they suck rocks. It may indeed be the future, but not because human beings wants it, but because the subhuman creatures in marketing will think it's cool.
Precisely where in the GPL does it mandate a grant of forgiveness before someone can distribute their own code?
I'm not a member of the KDE "tribe", but I do prefer KDE. Then Redhat came out with a white paper saying that distributing KDE was illegal. Redhat called me a criminal! I was only doing what RMS told me to do. You see, RMS said that if a friend asked for a copy of software, it would be morally wrong to refuse. A friend asked me for a copy of KDE and I gave him one. I distributed KDE. Evil me.
According to Redhat, I was a criminal. Debian and GNU agreed. I may not be a member of any "tribe", but it's clear to me which tribes only talk about freedom, and which ones actually practice it.
And after 5 years I'll bet most KDE users are using pretty much the latest version. Nearly 7 years after its release it is not odd to find Windows 95 still in use
Compare the costs/benefits of upgrading KDE-1.0 to KDE-2.2.1. Now compare the costs/benefits of upgrading Windows95 to WindowsME.
Frankly, for zero cost I get a wealth of new features and stability with KDE-2.2.1, but I have to spend $100 to get the miserly offerings of WindowsME. No contest. I know several people who still use Windows95. Their stated reason is simply that they see no compelling reason to upgrade.
What percentage of Windows users actually chose to upgrade from Window95 to 98 to ME? A very low percentage. Most Win98 users got theirs with their computer. Most WinME users got theirs with their computer.
(and of course, the demographics are completely different from Unix to Win/DOS users, and also for Win/NT, which is why you find a more NT users upgrading to 2K than you do W98 users to WME)
Because their license says "free to use"?
So I can dynamically link my own 100% original application to a GPL (not LGPL) library with no restriction?