Free, Open and Net BSD have essentially nothing in common at a kernel level with OSX.
Well, I wouldn't go that far. There are a number of subsystems and in-kernel programming interfaces that differ, in varying degrees, from those in other *BSDs - but, then again, that's also true of ${X}BSD and ${Y}BSD for all values of X and Y. However, a lot of kernel code can be moved between members of the set {XNU, FreeBSD kernel, OpenBSD kernel, NetBSD kernel}, and such code has been moved.
However, that is a bit of an obstacle to Apple "giving back" code, especially kernel code, as it would involve Apple doing work to port that code to the BSD in question.
Frankly the kernel that has the most Mach/XNU code in it other than OSX is NT which has some common code from a common (grand)parent, the Accent kernel.
Does NT share code with Accent or Mach, or does it just share concepts with it?
It shares no "XNU" code, in the sense of "stuff Apple's done"; for example, the interface into which file systems plug in XNU is somewhat similar to the interfaces into which they plug in the *BSDs, and the programming interfaces for kernel services they use vary from "similar to the ones in the *BSDs" to "significantly different form the ones in the *BSDs", but the interface into which file systems plug in NT is not very much like that of XNU or any of the *BSDs, and the programming interfaces for the kernel services aren't very much like those, either.
This is false. You have every right to do whatever you want with your code. Your problem stems from the fact that the GPL'd parts? Aren't your code. You don't want the freedom to do whatever you want with your code, you want the freedom to do whatever you want with someone else's code and you're annoyed that they won't give it to you for free.
If the GPL'd code is such a minor portion it should be easy enough to not use it, or to go to the owner and negotiate a non-GPL license to it (most authors of GPL'd software aren't morally opposed to giving you a conventional commercial license if you're willing to pay for it).
And, of course, you have the freedom not to use the GPLed code in the first place, which is what you should do if you don't want to put your code under the GPL. It might be inconvenient or expensive to avoid using the GPLed code, perhaps prohibitively so, but that's not the problem of the GPLed code's author.
But what the GPP actually said was: "The GPL... restricts you from taking private the hard work of the original authors." My point was that it does no such thing - the original authors (or current copywrite holders) are always allowed to keep their work public if they so desire. The GPL restricts you from taking private your work, in that if you wish to redistribute it at all you have to include your full source. A BSD license allows you to keep your modified source private, but gives you no power to make the unmodified code private.
I guess it's a question of what the person who said "The GPL... restricts you from taking private the hard work of the original authors."
If they meant that literally, it's true, but it's also true of most if not all other free-software licenses - and, yes, that's a point of confusion in anti-non-GPL-style license arguments. In that case, either they were confused about what non-copyleft licenses do or were, indeed, spreading FUD (by falsely stating that the BSD license allows somebody to take the work upon which they based their work private, not just to keep their derived work private).
If they meant it more loosely, as in "The GPL restricts you from taking private work that builds on the hard work of the original authors", that's true of copyleft licenses, but not other licenses. In that case, either they didn't take enough care when stating it or they were, again, spreading FUD (by overstating the case).
You are way too worked up, and are not really considering what I am saying
I am considering it. I'm just considering it to be incorrect; please do not get trapped by the fallacy that, once people understand what you have to say, they'll necessarily agree with what you say.
You simply are got grasping that my primary goal is the same as the GPL has
I think you are not grasping what the primary goals of the GPL are (if you want to know what the goals are, try reading, for example, A Quick Guide to GPLv3, in particular the "The Foundations of the GPL" section; if giving users the freedoms listed there aren't your goals, your goals aren't the same as the goals of the GPL, whether you think they are or not).
I seriously think that the end goals of the GPL if still desirable must be looked at carefully as to which approaches actually have the effect the GPL seeks (something you insist on off and on but no-one seems to have done any research on either way).
Sorry but I'm not understanding why you guys separate programmers from everyone else. If you write a GPLed word processor, why don't you require everyone to GPL every document they create with it? If you write a photo editing program, why don't you require every photo edited with it to be GPLed?
Person A uses your work to create thing B and you're fine with them not GPLing it.
Person C uses your work to creating thing D and you're not fine with it.
Why do you single out programmers but not anyone else?
The analogy to a GPLed word processor or photo editing program would be a GPLed compiler or a GPLed code editor or a tool such as that. The result of compiling code with GCC is not forced to be under the GPL - not even if the assembler-language output of GCC is turned into object code by the GNU assembler and linked into an executable or shared library by the GNU linker. Code that you've edited with GNU Emacs is not forced to be under the GPL. Heck, even code generated by Bison is not, by default, forced to be under the GPL, even though it includes the GPLed parser skeleton, "so long as that work isn't itself a parser generator using the skeleton or a modified version thereof as a parser skeleton", as of Bison 2.2.
The analogy to the GPL for documents would be something such as the GNU Free Documentation License, which requires that if you "copy and distribute a Modified Version of the Document" it must be released under that license. I.e., it's not the tools you use to produce the item in question that affect the license for the item, it's the licensed content you incorporate into the item that affects the license of the item.
So the GPL "singles out programmers" only to the extent that it covers the stuff programmers produce, i.e. code, just as the GFDL "singles out writers" in that it covers the stuff writers produce, i.e. text. Somebody could create a "GNU Free Images License" that covers images, so that if you took an image under such a "GFIL" and edited it, you couldn't prevent others from taking the image and editing it and putting it in a book or on a Web site or... (and they, in turn, couldn't do so with their edited version of the image).
It's fine if you want to use the GPL to force PROGRAMMERS to pay you back with their changes. Of course it's unclear to me why you don't also expect USERS to pay you back was well and only single out PROGRAMMERS.
Because it's not about the Benjamins. Many developers put stuff under the GPL not because they want to get paid (yes, you can use the GPL for that, by dual-licensing - pay me for a license that lets you link with my code without having to put your code under a GPL-compatible license) but because they don't want others who use their code to restrict availability of derivative works in order to ensure they get paid for those derivative works.
BSD is for people to like to give things away. I like to give my friends and family presents at Christmas. I don't require that they pay me back. I like to give donations to my favorite charities and organizations whether it's the EFF, PBS, NPR, ACLU, Red Cross, etc. I don't expect them to pay me back. I like to throw parties and invite my friends. I buy all the food and spend all day cooking. I don't require my anyone to pay me back.
It's perfectly reasonable of me to BSD my code in the same spirit of giving. Giving breeds giving. Sure a few people night not give back. So what. I don't care. Even if no one gave back I'd still do it.
I write free software code with no expectation of getting paid for it. For the GPLed code (GPLed because it's part of a GPLed project), I'm very happy that people can't modify it to make their own derivative works without making the modifications available to the user base as a whole. For the BSD-licensed code (BSD-licensed because it's part of a BSD-licensed project), I'm willing to put up with derivative works of that sort being made, given that 1) the original developers of that code gave it a BSD license and 2) that makes it more likely that some OSes will pick it up, even though a QNX port, for example, disappeared without a trace and later people asked about it.
Fortunately people do give back without being forced. Apache, Python, PHP, Lua, zlib,libjpeg, libpng, and thousands of other projects are doing just fine BSDed.
Sometimes they do, sometimes they don't. Some people did make changes to the GPLed project without making the changes available, and had to be yelled at to get them to give them back.
Because you want all improvements to your software to be available in the same way that your software is available, regardless of whether that makes it difficult or impossible for somebody else to make money from it? I.e., because you released the software to make it widely available, in source form, and you want improvements to be equally widely available?
so you just simply want code that some other guy worked hard on, even if he doesn't want to give it to you.
No. The "other guy" simply wants to use some GPLed code its developer worked hard on, and not release his changes to that code under a GPL-compatible license, even if the code's original developer wants him to. The original developer is, probably, quite happy for the "other guy" not to work on the code in the first place, and thus not to have any code of his own, if the "other guy" isn't willing to make his work available under the GPL.
Correct. They're not letting go because they don't want it to be "let go"
In other words, they do not really want it to be free. Any restrictions imply a lowering of freedom.
They want it and its derivatives to be free. That means that those who make derivatives of it aren't free to make non-free derivatives; there's a tradeoff of freedoms here.
No, it's like a parent insisting that, if they give their child a watch, the child take care of the watch in the same way that the parent did.
And then checking every month to make sure the watch is untouched.
No, then yelling at the kid if they hear that they haven't treated the watch properly. There's no GPL License Audits where somebody shows up at your company and checks all your products.
...which still doesn't get the first someone the source to the second someone's derivative work.
AHA! This is where you are wrong. Because in real life it WILL give them that source, were it to be had at all... because that source will either be given back to you or it would never have been available to be had by anyone else anyway (either because the GPL scared them off from using the framework to start with or because they extend and do not give back ignoring the license)
I.e., if you actually try to ensure that they give stuff back, the stuff won't exist in the first place. There are some for whom that's an acceptable tradeoff. (It's certainly an acceptable tradeoff for me as a Wireshark developer.)
"Healthier" in what sense?
Because changes given back to you are not forced.
For whom is that "healthier"? For people who want to make a derived work without allowing you to incorporate the improvements in your free-software version, yes, it's healthier. As, say, a Wireshark developer, why should I care about those people and their user base rather than the user base of the core Wireshark distribution?
Because users of your software are more free and need not worry that something they may do might accidentally trigger a restriction you are imposing on them.
End users who just download a binary distribution of Wireshark have nothing to worry about. (The ones who are confused about the license mostly think they need to worry about whether they need to purchase a site license or something such as that - i.e., they're confused in that they think Wireshark's under a commercial license.)
I did not use my opening quote lightly. If you REALLY love something, set it free. REALLY free. No conditions. Just let it be.
In case you hadn't noticed, software is an "it". It doesn't give a flying fuck whether it's "free" to do what it wants. That saying might apply to living beings with brains (although if you set a 3-year-old child "free", he or she might just run in front of a car and get killed), but it's just bullshit when applied to a pile of characters or a sequence of opcodes and operands. It's not as if software "really wants" to be "free" in the sense that anybody can get the source or "free" in the sense that anybody can modify it and not distribute the source; software doesn't even "want" not to be anthromorphized.
With non-GPL-style licenses, the freedom to change the terms of distribution, including restricting the freedom to redistribute, flows down the chain until somebody changes the term.
But there's the thing. The chain is an illusion. In real life there's almost never actually a chain - or at least no more so than with GPL projects, just look at the Emacs forks!!
There is not a chain, just a link...
Err, umm, there was most definitely "a chain" when, for example, the ClearSight people decide
My understanding is that Samsung uses a complex circular ownership structure with Samsung Everland at the top. This allow Lee GunHee and family to completely control Samsung even while owning only a small piece of the pie. This form of ownership is not valid for a public company in the United States, but Korea has different rules. Other companies achieve this same effect in other ways. Newscorp, for example, uses a multiclass stock structure whereby the family owns a small minority of the stock, but retain near majority voting rights.
Yeah, I was wondering whether there was, for example, some other form of shareholding that didn't show up in the pie charts, perhaps along the lines of the "A and B shares" type of stuff you mentioned for News Corporation.
Don't feel bad - perhaps I'm misjudging the other big Anglo-Saxon financial center, but I suspect replacing "Wall Street" with "the City" would only require also replacing "fcking nuts" and "crazy" with "barking mad".:-)
Wait until Apple makes an LCD TV... it will be prettier, more expensive, and have an Apple logo on it which won't glow except to let you know that it's off. It will also have a single sheet of laser cut something or other somewhere on it, and probably laser pin holes so you can't see the LEDs unless they're on.
No, the Apple logo will only glow when it's on. However, there will be something right below the screen that will pulse when it's off and be off when it's on (unlike our Vizio, which glows white when it's on and orange when it's off).
And at the bottom of the screen it'll show an image of the knobs on an old DuMont TV, to make people old enough to remember those desktop calendar pads and bookmark ribbons in address books comfortable.
That combination presumably translates to at least "I'm reducing the growth rate of my energy usage" or perhaps "I'm keeping my energy use steady by wasting less but using more in a non-wasteful fashion".
I have no plans to give up my electrical appliances
"Reducing my energy (usage)" does not ipso facto mean "giving up my electrical appliances". It might, however, mean, for example, "unplugging the flat-screen TeeVee if you're not actually watching something" (which I do primarily because the alternative is to have extension cords across the floor; some day I'll have to get or rent one of those plug-it-into-the-current-path electricity meters and see how much of a difference that would actually make).
So what you're saying that the American 99% will get a bit richer, at a price of Korean (or Chinese, or whatever) 1% getting a lot richer instead of the local 1%.
In case of Samsung in particular, its stock only trades on the Korean stock market.
Not exactly - it also trades on the London and Luxembourg stock exchanges, but what's sold there might be their Global Depository Receipts. Apparently, "as someone residing outside Korea, [I] may invest in Samsung stock (005930, 005935) through a qualified institutional broker in Seoul", and "For information regarding investment in Samsung GDR, [I should] contact [my] broker."
Apparently a slight majority of their stock, and a significant majority of their preferred stock, is foreign-held, although I don't know whether there's anything those pie charts are leaving out. The "List of a Major Shareholder & Related Parties" includes several individuals with Korean names and affiliates with names that include the string "Samsung" (today's lesson is brought to you by the letters "c" "h" "a" "e" "b" "o", and "l"), and the "List of Shareholders with the Ownership of 5% and above" includes Good Old American First National City Bank, err, umm, Citibank, along with Samsung Life Insurance and National Pension Service (probably meaning the Korean national pension service)
...at least not in the sense that they used to make those chips in their own fabs and are now having somebody else fab them. They've always outsourced the production of all their chip designs, as they've never owned any fabs. (Well, not as far as I know, at least.)
(And, unless you consider either the California Republic or the Republic of Texas to still exist, they didn't offshore it, either - not even if you include doing stuff across one or more land borders "offshoring".:-))
And, in real life, some may want to give a gift away and ensure that all subsequent recipients of derivative works of that gift have the same rights
They they are not really letting go.
Correct. They're not letting go because they don't want it to be "let go" in the sense that people can use it however they choose, including making improvements that aren't available to the entire user base of the software under the same terms as the original software.
It's like a parent insisting that when a daughter has a child they be allowed to name it.
No, it's like a parent insisting that, if they give their child a watch, the child take care of the watch in the same way that the parent did.
Someone can always come back to YOU for the gift if someones derivation will not be given freely.
...which still doesn't get the first someone the source to the second someone's derivative work.
But in practice what really happens is that few derivations occur, because almost no-one wants the administrative overhead of caring for the gift to begin with - so changes naturally mostly migrate back to yourself, or simply get lost in some one-off mutation that is seen and used by very few people.
Just out of curiosity, are there any actual statistics on the amount of closed-source and open-source changes to non-GPLed projects, so that there are actual numbers to indicate whether that is what happens in practice?
That organic growth is I think a lot healthier than the micromanagement of what people can do with your gift after they receive it.
"Healthier" in what sense?
That's why I'd really just rather produce code with a license that lets people do whatever without restriction. That is true freedom, imposing restrictions on freedom and then in siting it is "Double-Plus Freedom" is just whitewashing the yoke, so to speak.
No. It's just a question of how different forms of freedom flow down the distribution chain. With non-GPL-style licenses, the freedom to change the terms of distribution, including restricting the freedom to redistribute, flows down the chain until somebody changes the term. With GPL-style licenses, the freedom to change the terms of distribution, including restricting the freedom to redistribute, is pretty much absent. Different people value different freedoms, and I'm perfectly willing to accept different people's valuation of those freedoms; I refuse, however, to deem certain freedoms "worthy of preservation" and others "not worthy of preservation".
I don't see why anyone would not want to use the GPL if they want their software to be free and open. Why create something, give it out for free, and then allow businesses to take your work, profit from it, and give nothing back?
Why require something back only from those who improve your software with bug fixes and additions, and not from all other users and distributors? Sounds more like an unfair disincentive for creating better software to me.
It's only "unfair" to people who think they should be able to use source code in the way they want, not in the way the developer of the source code wants, no matter how inconvenient the way the developer of the source code wants the software to be used is inconvenient to the person who wants to use it.
It's a disincentive to creating better software and not licensing the improvements in a fashion compatible with the GPL. That's entirely the intent of people releasing GPLed software; they are, for better or worse, willing to forego proprietary improvements being made to their software.
Why create something, give it out for free, and then allow businesses to take your work, profit from it, and give nothing back?
If you love something, let it go free.
If it does not return to you, it was never yours...
If I am truly being charitable, why NOT let someone profit from something I have made? I would like them to return something if possible but why would I wish to place that burden as a legal demand instead of a request that they can choose to honor?
A lot of times you create something for others to use simply for the joy of having something used and helping people. Letting them use it without necessarily giving back or even crediting you does not diminish that.
I like the GPL a lot, I appreciate what it is trying to do... but in real life I love to gift a gift more freely, so that others may be encouraged to give back the same gift in return.
And, in real life, some may want to give a gift away and ensure that all subsequent recipients of derivative works of that gift have the same rights to the derivative work as the original recipient had to the original gift. I.e., they may be thinking of more people than just the initial recipient of the gift - or maybe they're thinking of "the world" or "the community" or something such as that as the recipient, not just some members of the community.
So, no, a newagey appeal probably won't convince people to drop the GPL. Srsly.
What the GPL requires is, simply put, a royalty payment made for the rights to publish or produce derivative works yourself of the protected and licensed work.
Presumably by "royalty payment" you mean something other than money. Yes, some software is "dual licensed" so that you can either use it for free-as-in-beer under the terms of the GPL or can pay a licensing fee for the right to make a derivative work without the derivative work being under the GPL, but I don't think all GPLed software is dual-licensed, and some people and organizations that might want to make a derivative work from some GPLed software might not want to be obliged to make the source code available to everybody who gets the binary form of the work or to allow anybody who receives the source or binary code for the work to redistribute it freely under the terms of the GPL.
Instead of asking/b*tching/moaning and groaning, why not try it?
Presumably "it" doesn't mean "trying to build Mac OS X, in its entirety, from the stuff on opensource.apple.com", because he already knows for certain that it won't work. Apple don't open-source WindowServer or Cocoa, for example, so no Mac OS X GUI.
Yes, but that's because he hyperbolized with "all their achievements will be lost". Not only will WebKit not instantly disappear, but, for example, none of the LLVM project work (which is not based on GPLed code, so it's not as if the license required Apple to make it open source) will disappear.
However, Mac OS X would be lost (unless they suddenly open-sourced it or sold it to somebody else).
BTW - if you download every single package from gnu.org, you STILL don't have a complete working build of an OS, so what are you complaining about, really?
The fact that some people insist on calling it GNU/Linux but don't insist on calling it GNU/{various other contributors}/Linux?:-)
Nice. So if I download all that and build it properly, do I get a complete working MacOS X build? If the answer is "no", I rest my case.
Good, because, if you rest your case, that means we get to further attack your case and you've admitted you have no response that you can offer to further attacks.:-)
"Should Apple fall one day or just discontinue its BSD-based products, all their achievements will be lost." is, like it or not, inequivalent to "I can't build Mac OS X entirely from free software".
"Should Apple fall one day or just continue its BSD-based products, Mac OS X {and iOS} will be lost" is equivalent to "I can't build Mac OS X {or iOS} entirely from free software", but that's not what you said. If that's what you meant, you should have said it, rather than hyperbolizing with "all their achievements will be lost".
And not contributing anything back to the community. Should Apple fall one day or just discontinue its BSD-based products, all their achievements will be lost.
...with the possible exception of WebKit and clang and LLDB and their contributions to LLVM and their contributions to GCC to handle Objective-C (assuming they've been picked up by GCC) and their contributions to GDB (assuming they've been picked up by GDB) and libdispatch and launchd and other Mac OS Forge projects and any stuff people have picked up from opensource.apple.com.
There is nothing more annoying then say a company that makes an OS which uses an MIT licensed graphics library or a BSD licensed network stack but at the same time fights aggressively against free and open source software.
Apple?
Well, maybe, except that
if the "MIT licensed graphics library" refers to one of the X11 libraries, it's not used by the native GUI of the OS, it (and the other X11 libraries, along with the X server) are provided to make it a bit easier to build X11 applications and run them on Mac OS X without having to port them to a toolkit that runs directly on the OS X native window system (whether it's one of Apple's toolkits such as the GUI parts of Carbon or Cocoa or a third-party toolkit such as Qt);
Apple may compete aggressively against FOSS, and may protect its non-FOSS software, but it also develops and contributes to a number of FOSS projects.
Slashdot Mirror
Free, Open and Net BSD have essentially nothing in common at a kernel level with OSX.
Well, I wouldn't go that far. There are a number of subsystems and in-kernel programming interfaces that differ, in varying degrees, from those in other *BSDs - but, then again, that's also true of ${X}BSD and ${Y}BSD for all values of X and Y. However, a lot of kernel code can be moved between members of the set {XNU, FreeBSD kernel, OpenBSD kernel, NetBSD kernel}, and such code has been moved.
However, that is a bit of an obstacle to Apple "giving back" code, especially kernel code, as it would involve Apple doing work to port that code to the BSD in question.
Frankly the kernel that has the most Mach/XNU code in it other than OSX is NT which has some common code from a common (grand)parent, the Accent kernel.
Does NT share code with Accent or Mach, or does it just share concepts with it?
It shares no "XNU" code, in the sense of "stuff Apple's done"; for example, the interface into which file systems plug in XNU is somewhat similar to the interfaces into which they plug in the *BSDs, and the programming interfaces for kernel services they use vary from "similar to the ones in the *BSDs" to "significantly different form the ones in the *BSDs", but the interface into which file systems plug in NT is not very much like that of XNU or any of the *BSDs, and the programming interfaces for the kernel services aren't very much like those, either.
This is false. You have every right to do whatever you want with your code. Your problem stems from the fact that the GPL'd parts? Aren't your code. You don't want the freedom to do whatever you want with your code, you want the freedom to do whatever you want with someone else's code and you're annoyed that they won't give it to you for free.
If the GPL'd code is such a minor portion it should be easy enough to not use it, or to go to the owner and negotiate a non-GPL license to it (most authors of GPL'd software aren't morally opposed to giving you a conventional commercial license if you're willing to pay for it).
And, of course, you have the freedom not to use the GPLed code in the first place, which is what you should do if you don't want to put your code under the GPL. It might be inconvenient or expensive to avoid using the GPLed code, perhaps prohibitively so, but that's not the problem of the GPLed code's author.
But what the GPP actually said was: "The GPL... restricts you from taking private the hard work of the original authors." My point was that it does no such thing - the original authors (or current copywrite holders) are always allowed to keep their work public if they so desire. The GPL restricts you from taking private your work, in that if you wish to redistribute it at all you have to include your full source. A BSD license allows you to keep your modified source private, but gives you no power to make the unmodified code private.
I guess it's a question of what the person who said "The GPL... restricts you from taking private the hard work of the original authors."
If they meant that literally, it's true, but it's also true of most if not all other free-software licenses - and, yes, that's a point of confusion in anti-non-GPL-style license arguments. In that case, either they were confused about what non-copyleft licenses do or were, indeed, spreading FUD (by falsely stating that the BSD license allows somebody to take the work upon which they based their work private, not just to keep their derived work private).
If they meant it more loosely, as in "The GPL restricts you from taking private work that builds on the hard work of the original authors", that's true of copyleft licenses, but not other licenses. In that case, either they didn't take enough care when stating it or they were, again, spreading FUD (by overstating the case).
You are way too worked up, and are not really considering what I am saying
I am considering it. I'm just considering it to be incorrect; please do not get trapped by the fallacy that, once people understand what you have to say, they'll necessarily agree with what you say.
You simply are got grasping that my primary goal is the same as the GPL has
I think you are not grasping what the primary goals of the GPL are (if you want to know what the goals are, try reading, for example, A Quick Guide to GPLv3, in particular the "The Foundations of the GPL" section; if giving users the freedoms listed there aren't your goals, your goals aren't the same as the goals of the GPL, whether you think they are or not).
I seriously think that the end goals of the GPL if still desirable must be looked at carefully as to which approaches actually have the effect the GPL seeks (something you insist on off and on but no-one seems to have done any research on either way).
"The effect the GPL seeks" is keeping the software it covers and all derived works of it as free software; it's not just "maximize giveback of changes". Here's Richard Stallman's explanation of why the GPL is the way it is.
On a side note I find it odd you bring up the Linux kernel in defending the GPL as they will not even shift to GPL3!
News flash: not all defenders of the general goals of the GPL agree with all of the means the FSF have taken in GPLv3. Hell, the GPLed free software project on which I'm a core developer is under the GPLv2, not v3.
Sorry but I'm not understanding why you guys separate programmers from everyone else. If you write a GPLed word processor, why don't you require everyone to GPL every document they create with it? If you write a photo editing program, why don't you require every photo edited with it to be GPLed?
Person A uses your work to create thing B and you're fine with them not GPLing it. Person C uses your work to creating thing D and you're not fine with it.
Why do you single out programmers but not anyone else?
The analogy to a GPLed word processor or photo editing program would be a GPLed compiler or a GPLed code editor or a tool such as that. The result of compiling code with GCC is not forced to be under the GPL - not even if the assembler-language output of GCC is turned into object code by the GNU assembler and linked into an executable or shared library by the GNU linker. Code that you've edited with GNU Emacs is not forced to be under the GPL. Heck, even code generated by Bison is not, by default, forced to be under the GPL, even though it includes the GPLed parser skeleton, "so long as that work isn't itself a parser generator using the skeleton or a modified version thereof as a parser skeleton", as of Bison 2.2.
The analogy to the GPL for documents would be something such as the GNU Free Documentation License, which requires that if you "copy and distribute a Modified Version of the Document" it must be released under that license. I.e., it's not the tools you use to produce the item in question that affect the license for the item, it's the licensed content you incorporate into the item that affects the license of the item.
So the GPL "singles out programmers" only to the extent that it covers the stuff programmers produce, i.e. code, just as the GFDL "singles out writers" in that it covers the stuff writers produce, i.e. text. Somebody could create a "GNU Free Images License" that covers images, so that if you took an image under such a "GFIL" and edited it, you couldn't prevent others from taking the image and editing it and putting it in a book or on a Web site or... (and they, in turn, couldn't do so with their edited version of the image).
It's fine if you want to use the GPL to force PROGRAMMERS to pay you back with their changes. Of course it's unclear to me why you don't also expect USERS to pay you back was well and only single out PROGRAMMERS.
Because it's not about the Benjamins. Many developers put stuff under the GPL not because they want to get paid (yes, you can use the GPL for that, by dual-licensing - pay me for a license that lets you link with my code without having to put your code under a GPL-compatible license) but because they don't want others who use their code to restrict availability of derivative works in order to ensure they get paid for those derivative works.
BSD is for people to like to give things away. I like to give my friends and family presents at Christmas. I don't require that they pay me back. I like to give donations to my favorite charities and organizations whether it's the EFF, PBS, NPR, ACLU, Red Cross, etc. I don't expect them to pay me back. I like to throw parties and invite my friends. I buy all the food and spend all day cooking. I don't require my anyone to pay me back.
It's perfectly reasonable of me to BSD my code in the same spirit of giving. Giving breeds giving. Sure a few people night not give back. So what. I don't care. Even if no one gave back I'd still do it.
I write free software code with no expectation of getting paid for it. For the GPLed code (GPLed because it's part of a GPLed project), I'm very happy that people can't modify it to make their own derivative works without making the modifications available to the user base as a whole. For the BSD-licensed code (BSD-licensed because it's part of a BSD-licensed project), I'm willing to put up with derivative works of that sort being made, given that 1) the original developers of that code gave it a BSD license and 2) that makes it more likely that some OSes will pick it up, even though a QNX port, for example, disappeared without a trace and later people asked about it.
Fortunately people do give back without being forced. Apache, Python, PHP, Lua, zlib,libjpeg, libpng, and thousands of other projects are doing just fine BSDed.
Sometimes they do, sometimes they don't. Some people did make changes to the GPLed project without making the changes available, and had to be yelled at to get them to give them back.
Because you want all improvements to your software to be available in the same way that your software is available, regardless of whether that makes it difficult or impossible for somebody else to make money from it? I.e., because you released the software to make it widely available, in source form, and you want improvements to be equally widely available?
so you just simply want code that some other guy worked hard on, even if he doesn't want to give it to you.
No. The "other guy" simply wants to use some GPLed code its developer worked hard on, and not release his changes to that code under a GPL-compatible license, even if the code's original developer wants him to. The original developer is, probably, quite happy for the "other guy" not to work on the code in the first place, and thus not to have any code of his own, if the "other guy" isn't willing to make his work available under the GPL.
Correct. They're not letting go because they don't want it to be "let go"
In other words, they do not really want it to be free. Any restrictions imply a lowering of freedom.
They want it and its derivatives to be free. That means that those who make derivatives of it aren't free to make non-free derivatives; there's a tradeoff of freedoms here.
No, it's like a parent insisting that, if they give their child a watch, the child take care of the watch in the same way that the parent did.
And then checking every month to make sure the watch is untouched.
No, then yelling at the kid if they hear that they haven't treated the watch properly. There's no GPL License Audits where somebody shows up at your company and checks all your products.
...which still doesn't get the first someone the source to the second someone's derivative work.
AHA! This is where you are wrong. Because in real life it WILL give them that source, were it to be had at all... because that source will either be given back to you or it would never have been available to be had by anyone else anyway (either because the GPL scared them off from using the framework to start with or because they extend and do not give back ignoring the license)
I.e., if you actually try to ensure that they give stuff back, the stuff won't exist in the first place. There are some for whom that's an acceptable tradeoff. (It's certainly an acceptable tradeoff for me as a Wireshark developer.)
"Healthier" in what sense?
Because changes given back to you are not forced.
For whom is that "healthier"? For people who want to make a derived work without allowing you to incorporate the improvements in your free-software version, yes, it's healthier. As, say, a Wireshark developer, why should I care about those people and their user base rather than the user base of the core Wireshark distribution?
Because users of your software are more free and need not worry that something they may do might accidentally trigger a restriction you are imposing on them.
End users who just download a binary distribution of Wireshark have nothing to worry about. (The ones who are confused about the license mostly think they need to worry about whether they need to purchase a site license or something such as that - i.e., they're confused in that they think Wireshark's under a commercial license.)
I did not use my opening quote lightly. If you REALLY love something, set it free. REALLY free. No conditions. Just let it be.
In case you hadn't noticed, software is an "it". It doesn't give a flying fuck whether it's "free" to do what it wants. That saying might apply to living beings with brains (although if you set a 3-year-old child "free", he or she might just run in front of a car and get killed), but it's just bullshit when applied to a pile of characters or a sequence of opcodes and operands. It's not as if software "really wants" to be "free" in the sense that anybody can get the source or "free" in the sense that anybody can modify it and not distribute the source; software doesn't even "want" not to be anthromorphized.
With non-GPL-style licenses, the freedom to change the terms of distribution, including restricting the freedom to redistribute, flows down the chain until somebody changes the term.
But there's the thing. The chain is an illusion. In real life there's almost never actually a chain - or at least no more so than with GPL projects, just look at the Emacs forks!!
There is not a chain, just a link...
Err, umm, there was most definitely "a chain" when, for example, the ClearSight people decide
My understanding is that Samsung uses a complex circular ownership structure with Samsung Everland at the top. This allow Lee GunHee and family to completely control Samsung even while owning only a small piece of the pie. This form of ownership is not valid for a public company in the United States, but Korea has different rules. Other companies achieve this same effect in other ways. Newscorp, for example, uses a multiclass stock structure whereby the family owns a small minority of the stock, but retain near majority voting rights.
Yeah, I was wondering whether there was, for example, some other form of shareholding that didn't show up in the pie charts, perhaps along the lines of the "A and B shares" type of stuff you mentioned for News Corporation.
Don't feel bad - perhaps I'm misjudging the other big Anglo-Saxon financial center, but I suspect replacing "Wall Street" with "the City" would only require also replacing "fcking nuts" and "crazy" with "barking mad". :-)
Wait until Apple makes an LCD TV... it will be prettier, more expensive, and have an Apple logo on it which won't glow except to let you know that it's off. It will also have a single sheet of laser cut something or other somewhere on it, and probably laser pin holes so you can't see the LEDs unless they're on.
No, the Apple logo will only glow when it's on. However, there will be something right below the screen that will pulse when it's off and be off when it's on (unlike our Vizio, which glows white when it's on and orange when it's off).
And at the bottom of the screen it'll show an image of the knobs on an old DuMont TV, to make people old enough to remember those desktop calendar pads and bookmark ribbons in address books comfortable.
Perhaps this is why Samsung stock is only traded on the Korean stock exchange, and not in U.S.
So what happens if you replace "Wall Street" with "The City"? :-) (Samsung GDRs are traded on the London Stock Exchange.)
I'm not reducing my energy. I'm using it smarter
That combination presumably translates to at least "I'm reducing the growth rate of my energy usage" or perhaps "I'm keeping my energy use steady by wasting less but using more in a non-wasteful fashion".
I have no plans to give up my electrical appliances
"Reducing my energy (usage)" does not ipso facto mean "giving up my electrical appliances". It might, however, mean, for example, "unplugging the flat-screen TeeVee if you're not actually watching something" (which I do primarily because the alternative is to have extension cords across the floor; some day I'll have to get or rent one of those plug-it-into-the-current-path electricity meters and see how much of a difference that would actually make).
So what you're saying that the American 99% will get a bit richer, at a price of Korean (or Chinese, or whatever) 1% getting a lot richer instead of the local 1%.
Well, I wouldn't go quite that far; see my other reply - an Indian-American member of the local 1%, along with an all-American member of the local 1%, and other members of the local 1% will probably get richer as well (for "local" defined as "the U.S." - and, at least for those two, you could probably define it as "Manhattan Island").
In case of Samsung in particular, its stock only trades on the Korean stock market.
Not exactly - it also trades on the London and Luxembourg stock exchanges, but what's sold there might be their Global Depository Receipts. Apparently, "as someone residing outside Korea, [I] may invest in Samsung stock (005930, 005935) through a qualified institutional broker in Seoul", and "For information regarding investment in Samsung GDR, [I should] contact [my] broker."
Apparently a slight majority of their stock, and a significant majority of their preferred stock, is foreign-held, although I don't know whether there's anything those pie charts are leaving out. The "List of a Major Shareholder & Related Parties" includes several individuals with Korean names and affiliates with names that include the string "Samsung" (today's lesson is brought to you by the letters "c" "h" "a" "e" "b" "o", and "l"), and the "List of Shareholders with the Ownership of 5% and above" includes Good Old American First National City Bank, err, umm, Citibank, along with Samsung Life Insurance and National Pension Service (probably meaning the Korean national pension service)
...at least not in the sense that they used to make those chips in their own fabs and are now having somebody else fab them. They've always outsourced the production of all their chip designs, as they've never owned any fabs. (Well, not as far as I know, at least.)
(And, unless you consider either the California Republic or the Republic of Texas to still exist, they didn't offshore it, either - not even if you include doing stuff across one or more land borders "offshoring". :-))
And, in real life, some may want to give a gift away and ensure that all subsequent recipients of derivative works of that gift have the same rights
They they are not really letting go.
Correct. They're not letting go because they don't want it to be "let go" in the sense that people can use it however they choose, including making improvements that aren't available to the entire user base of the software under the same terms as the original software.
It's like a parent insisting that when a daughter has a child they be allowed to name it.
No, it's like a parent insisting that, if they give their child a watch, the child take care of the watch in the same way that the parent did.
Someone can always come back to YOU for the gift if someones derivation will not be given freely.
...which still doesn't get the first someone the source to the second someone's derivative work.
But in practice what really happens is that few derivations occur, because almost no-one wants the administrative overhead of caring for the gift to begin with - so changes naturally mostly migrate back to yourself, or simply get lost in some one-off mutation that is seen and used by very few people.
Just out of curiosity, are there any actual statistics on the amount of closed-source and open-source changes to non-GPLed projects, so that there are actual numbers to indicate whether that is what happens in practice?
That organic growth is I think a lot healthier than the micromanagement of what people can do with your gift after they receive it.
"Healthier" in what sense?
That's why I'd really just rather produce code with a license that lets people do whatever without restriction. That is true freedom, imposing restrictions on freedom and then in siting it is "Double-Plus Freedom" is just whitewashing the yoke, so to speak.
No. It's just a question of how different forms of freedom flow down the distribution chain. With non-GPL-style licenses, the freedom to change the terms of distribution, including restricting the freedom to redistribute, flows down the chain until somebody changes the term. With GPL-style licenses, the freedom to change the terms of distribution, including restricting the freedom to redistribute, is pretty much absent. Different people value different freedoms, and I'm perfectly willing to accept different people's valuation of those freedoms; I refuse, however, to deem certain freedoms "worthy of preservation" and others "not worthy of preservation".
I don't see why anyone would not want to use the GPL if they want their software to be free and open. Why create something, give it out for free, and then allow businesses to take your work, profit from it, and give nothing back?
Why require something back only from those who improve your software with bug fixes and additions, and not from all other users and distributors? Sounds more like an unfair disincentive for creating better software to me.
It's only "unfair" to people who think they should be able to use source code in the way they want, not in the way the developer of the source code wants, no matter how inconvenient the way the developer of the source code wants the software to be used is inconvenient to the person who wants to use it.
It's a disincentive to creating better software and not licensing the improvements in a fashion compatible with the GPL. That's entirely the intent of people releasing GPLed software; they are, for better or worse, willing to forego proprietary improvements being made to their software.
Why create something, give it out for free, and then allow businesses to take your work, profit from it, and give nothing back?
If you love something, let it go free.
If it does not return to you, it was never yours...
If I am truly being charitable, why NOT let someone profit from something I have made? I would like them to return something if possible but why would I wish to place that burden as a legal demand instead of a request that they can choose to honor?
A lot of times you create something for others to use simply for the joy of having something used and helping people. Letting them use it without necessarily giving back or even crediting you does not diminish that.
I like the GPL a lot, I appreciate what it is trying to do... but in real life I love to gift a gift more freely, so that others may be encouraged to give back the same gift in return.
And, in real life, some may want to give a gift away and ensure that all subsequent recipients of derivative works of that gift have the same rights to the derivative work as the original recipient had to the original gift. I.e., they may be thinking of more people than just the initial recipient of the gift - or maybe they're thinking of "the world" or "the community" or something such as that as the recipient, not just some members of the community.
So, no, a newagey appeal probably won't convince people to drop the GPL. Srsly.
What the GPL requires is, simply put, a royalty payment made for the rights to publish or produce derivative works yourself of the protected and licensed work.
Presumably by "royalty payment" you mean something other than money. Yes, some software is "dual licensed" so that you can either use it for free-as-in-beer under the terms of the GPL or can pay a licensing fee for the right to make a derivative work without the derivative work being under the GPL, but I don't think all GPLed software is dual-licensed, and some people and organizations that might want to make a derivative work from some GPLed software might not want to be obliged to make the source code available to everybody who gets the binary form of the work or to allow anybody who receives the source or binary code for the work to redistribute it freely under the terms of the GPL.
Short of the kernel ... which they stopped releasing the source too since it kept being used to easily work around the OSX license restriction
So what's this - chopped liver? They haven't released the source to Don't Steal Mac OS X, but that's another matter. :-)
They don't release the source to all kernel extensions (kernel loadable modules, for the Linux weenies in the audience).
Instead of asking/b*tching/moaning and groaning, why not try it?
Presumably "it" doesn't mean "trying to build Mac OS X, in its entirety, from the stuff on opensource.apple.com", because he already knows for certain that it won't work. Apple don't open-source WindowServer or Cocoa, for example, so no Mac OS X GUI.
Besides, the link already proved your original knee-jerk claim was totally wrong.
Yes, but that's because he hyperbolized with "all their achievements will be lost". Not only will WebKit not instantly disappear, but, for example, none of the LLVM project work (which is not based on GPLed code, so it's not as if the license required Apple to make it open source) will disappear.
However, Mac OS X would be lost (unless they suddenly open-sourced it or sold it to somebody else).
BTW - if you download every single package from gnu.org, you STILL don't have a complete working build of an OS, so what are you complaining about, really?
The fact that some people insist on calling it GNU/Linux but don't insist on calling it GNU/{various other contributors}/Linux? :-)
Nice. So if I download all that and build it properly, do I get a complete working MacOS X build? If the answer is "no", I rest my case.
Good, because, if you rest your case, that means we get to further attack your case and you've admitted you have no response that you can offer to further attacks. :-)
"Should Apple fall one day or just discontinue its BSD-based products, all their achievements will be lost." is, like it or not, inequivalent to "I can't build Mac OS X entirely from free software".
"Should Apple fall one day or just continue its BSD-based products, Mac OS X {and iOS} will be lost" is equivalent to "I can't build Mac OS X {or iOS} entirely from free software", but that's not what you said. If that's what you meant, you should have said it, rather than hyperbolizing with "all their achievements will be lost".
And not contributing anything back to the community. Should Apple fall one day or just discontinue its BSD-based products, all their achievements will be lost.
...with the possible exception of WebKit and clang and LLDB and their contributions to LLVM and their contributions to GCC to handle Objective-C (assuming they've been picked up by GCC) and their contributions to GDB (assuming they've been picked up by GDB) and libdispatch and launchd and other Mac OS Forge projects and any stuff people have picked up from opensource.apple.com.
There is nothing more annoying then say a company that makes an OS which uses an MIT licensed graphics library or a BSD licensed network stack but at the same time fights aggressively against free and open source software.
Apple?
Well, maybe, except that