Let me fix that for you: BSD is about the freedom to make choices for everyone in the distribution chain. GPL is about imposing restrictions ON THE USE OF THE CODE for everyone in the distribution chain.
BSD allows any stage in the distribution chain to impose whatever restrictions on distribution they want to upon all subsequent stages in the distribution chain.
GPL requires all stages in the distribution chain to impose the same restrictions on distribution upon all stages in the distribution chain.
This is the flip side of "BSD is about freedom of the distributor; GPL is about freedom of the code". Attaching "freedom" to one and only one of the licenses - regardless of whether it's a non-GPL-style license or a GPL-style license - comes across as an attempt to bias the discussion in favor of the freedoms offered by that license. Different people in the distribution chain may want different freedoms and restrictions at various stages in the chain.
I think the purpose of the GPL is to ensure that those that profit from your work also give back.
Nonsense. In fact, potentially completely the opposite. Someone who profits from distributing unmodified copies of your code doesn't have to give you anything back. But someone who tries to combine your code with code which they have previously written, and give away the combined result for free (or maybe even at a loss including their server hosting fees), is still required to send you a whole bunch of code unrelated to your work.
A commercial distribution license based on a percentage of sales would be more in line with your stated purpose. Or maybe a commercial license with a 100% discount for people who send you rights to use a number of lines of their own code equal to yours.
If "give back" means "give back directly to the "you" of "your work"", yes. If it means "give back to "the community"", not necessarily - you may not directly care about the code "unrelated to your work", but you might like your work to be used only in projects where the users can get the source, modify it, and redistribute the results under the terms of the GPL, so that somebody else could get code that's related to their work.
you would have to provide any improvement to the library you are using, but not code for other parts of the project.
If the library is GPLed rather than LGPLed, you would most definitely have to provide code for parts of the project linked with it, unless it falls under the "system library exception".
What is "profit from" also? If I borrow some internationalization code as part of a huge project, this saves the company money from having to buy a propriety product perhaps. More likely though it means we don't implement it ourselves and have less bugs down the road than if we rolled our own. Now does that mean we profited and the original author now gets access to every single line of our code as the GPL would imply?
No, it means that (whether you profited or not), all recipients of your huge project get to give that project away to anybody they want (under the terms of the GPL, so they can't further restrict its redistribution) and get to ask for and get the source to your huge project, and that anybody they give it away to gets the same rights.
Even if the code is proprietary and our competitors are anxious to get a peek at it?
Yes, in which case you would be advised not to use GPLed internationalization code. Sorry.
Even if various government agencies disallow giving away the code or allowing end user customization of the machines?
Yes, in which case you would be advised not to use GPLed internationalization code. Sorry.
With GPL the return payment is that you must also be GPL in absolutely everything you do. With BSD the return payment is that you give recognition to the author and keep the copyright notices intact. The first type of payment is too high for most companies unless they've got a software model that fits (ie, dynamic libraries, separately loaded programs, kernel modules, multiple cpus). The second payment is much easier but many companies don't know of it and they associate all free or open source software as GPL tainted.
Then the right answer isn't "complain about the GPL", it's "promote non-GPL-style licenses and code that uses them, so that more companies know about it and use non-GPLed software in those cases where it works better for them".
So the result is many smaller companies reinvent small pieces of code or libraries all the time, not the result desired if the author wanted to share code.
Perhaps the desire of authors of at least some GPLed code isn't simply "to share code"; perhaps it's "to share code in a fashion that keeps it, and any improvements to it, under the GPL", in which case that might well be the result desired - they'd rather have people reinvent the wheel than to use their wheel design in a fashion they don't like. People who want their code used even in cases where the GPL gets in the way, including people whose sole goal is "to share code", would presumably use a BSD license or an MIT license or some other non-GPL-style license. Perhaps smaller companies might not realize that they won't get trapped by the GPL if they use that software; if so, that is, as noted, an evangelism problem.
So if I invent a robot with my own ai code but it uses GPL apis
Presumably you mean "GPLed libraries"; I don't know of any interfaces that are GPLed in any sense, e.g. somebody could reimplement a GPLed library, duplicating all its APIs, and release it under the BSD license without violating the GPL. I guess you could see legal arguments about, say, header files for the library, but those arguments may already have come up and been resolved.
What if I onvested 5,000,000 making the code? Whoops competitors now use my code and undercut me because they didnt have to invest the 5,000,000. I go out of business.
Then whoops maybe you should have invested a little more to implement that functionality yourself, or paying somebody else to implement it, or even just to try to find a free-as-in-beer-at-least implementation with a non-GPL license, rather than using the work of somebody else who doesn't want you to use it in a fashion where you don't have to make source code available under the terms of the GPL.
Google gets a free ride because they are not redistributing. For everyone else who makes smart appliances you are screwed.
If a smart appliance maker is going to invest a lot of money in the software, they're not "screwed", they're just required to invest a little more money making or buying or finding non-GPL software. If their "secret sauce" is the hardware, and giving away the source to the software doesn't make it significantly easier for others to make that hardware - or if they have a patent on the hardware - they may not be "screwed" at all.
Small business owners are too. You cant sell your company as that too counts as redistribution.
A small business using GPLed software most definitely can sell their company. It's not as if using GPLed software GPLs your entire company, including the office desks and the software you bought to run on your computers and your customer lists and the software you've developed without incorporating GPLed code into it.
Let's not, unless the developers of the GPLed software you're complaining about do so as well; otherwise, you're just constructing a strawman.
Do you use Visual Studio at work? Hell, did you use any include statement for any program you wrote while you were in school?
Therefore MS owns all your code right?
Wrong. If, however, you include any of the "Distributable Code" in your code when you distribute it, then you must do a number of things, including "for any Distributable Code having a filename extension of.lib, distribut[ing] only the results of running such Distributable Code through a linker with your program", "distribut[ing] Distributable Code included in a setup program only as part of that setup program without modification", and not "modify[ing] or distribut[ing] the source code of any Distributable Code so that any part of it becomes subject to an Excluded License" (where "Excluded License" includes, and is probably targeted at, licenses such as the GPL).
Similarly, if you use GCC at work, the FSF doesn't ipso facto own your code. If, however, you use GPLed "distributable code" as a part of your software, they impose constraints on its redistribution.
Microsoft's terms might be less problematic for some developers than the GPL's terms, but nobody who develops software is under any moral or legal obligation to avoid the GPL's terms - don't like those terms, don't use their code. Now, you might have to pay somebody if you're buying non-GPLed code from them to use in your product (e.g., getting a "developer license" from Essential Objects for their.NET widgets, so that you can redistribute those widgets in your application), or you might have to pay somebody to write non-GPLed code, or you might have to spend time and energy writing it yourself, or you might have to spend time and energy searching for non-GPLed free-as-in-beer-at-least code, but, well, that's life.
Under your own definition a simple link means that entity or person owns the code.
No. Under the GPL, a simple link means that the entity or person who owns the code with which you're linking gets to impose certain requirements on the redistribution of your code.
If I write something awesome for a specialized task with R&D then I should keep it and profit from it. Even if the toolkit was open source initially.
No, you shouldn't - you're using somebody else's code, so it's not as if that "something awesome" was written entirely by you. If the library with which you're working is a requirement for your software being "awesome", then some of the "awesome" is due to the other software (if it's not a requirement, then you don't have to use it). (And whether you "should" profit from it is up to your prospective suppliers and customers; if your prospective customers don't think your software is "awesome" enough to pay a price for it that covers your costs and desired profits, that's just Too Damn Bad, and if your prospective suppliers are not willing to supply you with tools/components/whatever under terms, whether financial or licensing, that allow you to make your desired profit, that's also just Too Damn Bad.)
Do you believe in property rights for software? If so, if you're honest, then you must believe in the rights of developers of software to say "you want to rely on the software I developed in your application at run time, even if it's just by calling it from a library, you have to make your application available under the same terms as my software". (If not, you're in no position to whine if you make no money because your software gets pirated.)
At that point people would be writing their own operating systems, apis, and so on if they wanted to own a
and somebody else distributes an improved version of it
FTFY.
Yes - to clarify, you can improve GPLed software and use it only by yourself and not have to make it available to anybody, and you can improve GPLed software and use it only within your organization and not have to make it available your organization. Thanks.
So your rights if you own the code are important too.
Yes. This includes the right to require that anybody who takes the code you own and improves it make those improvements available, in source form, to anybody who gets the improved version in binary form, and that they allow anybody who gets either the source or binary form from you be allowed to give it to anybody they choose. Yes, that might get in the way of the "anybody who takes the code you own and improves it" making money from those improvements; that's not the problem of the person who put the code under the GPL, it's the problem of the person who wants to make money from code that they didn't write in their entirety.
Yes I advocate the BSD license.
If that means "Yes, I prefer using the BSD license in code I write", that's fine. Even if that means "Yes, I think other people should use the BSD license", that's fine, but it imposes no obligation, moral or otherwise, on people to use the BSD license. If they choose to use the GPL for code that would be convenient for a project for which the GPL isn't convenient, that's your problem, not theirs - they're not under any obligation to make things convenient for you.
You post is FUD of the worst sort. The GPL in no way restricts freedom to profit from GPL code, it only restricts you from taking private the hard work of the original authors.
So... if I write some software... and release it under a non-GPL license... and someone else uses it... you claim that they can prevent me from giving it away to anyone else?
No, that's not the claim. The claim is that if you write some software, and release it under a GPL license, and somebody else makes an improved version of it, they can't refuse to make source to the improved version available and can't prohibit those who have received the source or binary of the improved version from giving it away to others.
I think a lot of developers see GPL as a "taking my toys and going home" license which discourages free use. If you weren't going to make a million dollar idea with your software, why stop someone else?
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?
Probably because they want their software to be used. I'm running a one man LLC and for my current work, if something relies on GPL then I simply can't use it.
You can't, for example, use Wireshark to look at networking problems on your machine if they're getting in the way of your work? Or do you just mean "I can't incorporate GPLed code in my current work"? Software can be "used" in ways other than incorporating it in other people's software projects.
Agreed. I think the shift has occurred because of increasing corporate interest in open source. BSD is seen as more corporate-friendly than GPL, when in fact it should be the other way around--BSD allows your competitors to reap the fruit of your labor without giving you anything in return. Start-ups, however, are lured by the idea of being able to close-source everything once their product becomes a smash hit,
There may well be start-ups who are. There are other start-ups who incorporate BSD-licensed code in their software and never open-source their software in the first place. One such startup succeeded rather well.
while established companies face genuine legal issues preventing them from linking GPL'ed code with closed-source code from vendors.
What is the point of software being called open source when it is not usable by developers?
GPLed software is usable by some developers - those who, for whatever reason, have no problem giving source away. It's not usable by those who do. The point of licensing software under the GPL is to ensure that all work done (sufficiently directly) atop your work will be be as available to others as was your original work. I consider that every bit as legitimate (no more, no less) as choosing a license that allows others to restrict the distribution of work done atop your work. If somebody chooses a GPL-like license for their software, and that's inconvenient for somebody who wants to build something atop that software, that just means they'll need to do the work to replicate the functionality of that software, get somebody else to do so under a more convenient license, or find software that offers the same functionality under a more convenient license.
If you made something that could make you a fortune, pay for your house, student loans, and a company of your own then you would understand why.
Case in point? Look at earlier this week when someone invented something with cameras that no one else could do in/. stories? Problem was the toolkit is under GPL. How is that fair? There work was worth money but the GPL forced them to give away the algorithms to conpetitors so a whole clone of the toolkit was needed.
How is "so a whole clone of the toolkit was needed" unfair? If they used a toolkit written by somebody else, then it's not all "their work"; some of it is the work of the developer of the toolkit. You don't want to be bound by the requirements of some software, don't use the software. If that means you have to write your own replacement, so it goes; it's not the rest of the universe's responsibility to give you what you want.
No, it's worse. Raenex discovered an oversight in GPLv2's definition of "work based on the Program", meaning that if a software distribution includes at least one GPLv2 program, the whole distribution is a "work based on the Program". True, the FSF likes to claim that low coupling on a multitasking operating system turns "work based on the Program" into "mere aggregation", and this may be true of the GPLv3, but that's not what the letter of the GPLv2 says according to Raenex.
I can think of one company that provides "a software distribution includes at least one GPLv2 program", doesn't make source available to a huge chunk of that software distribution, and that has a large legal department that would probably weigh in on this, so I wouldn't assume things will necessarily go that way.
When people say 'Linux' they usually mean Linux-and-all-of-the-associated-cruft. Typically this at least includes GNU libc, GNU binutils, and GNU coreutils (which, between them, are more code than the kernel), and typically the GNU shell (bash) and GNU libstdc++. All of these have no moved to GPLv3 (in some cases with the runtime exemption).
Referring to operating systems by the userspace software they use is ridiculous.
And referring to them by a subset of the userspace software.... (Admittedly, the C library is a very significant part of the userspace software, and it is from the GNU project on most Linux distributions, albeit not the one with the green robot as I understand it.
So "Linux", in the sense of "a Linux distribution", is not under any single license:
the kernel is under GPLv2;
the C library is probably under LGPLv2;
the developer tools might now be under GPLv3;
the command-line utilities are under a bunch of licenses including GPLv2, GPLv3, and probably a pile of other licenses including BSD;
the window system, if present, is probably under the MIT license, at least for desktop distributions;
the GUI toolkits, if present, are under a mix of licenses, with what I suspect are the most popular ones being under the LGPLv2;
Sorry, I screwed up the quote tags - "Boo ho, so write it yourself" should have been in quote tags; I'm quoting it and then violently agreeing with it. (And, no, that doesn't mean I'm in the "GPL rules, everything else drools" camp, either; I've worked at a startup where we did keep our "secret sauce" private, and where we explicitly avoided using GPL code, and I work on both BSD-licensed and GPL-licensed free software.)
That's my secret sauce. If I'm a startup and trying to form a niche in an industry, why would I want to give my recipe away?
Boo hoo, so write it yourself.
Precisely. "The problem with GPLv3 is that I can't use it in an application I develop unless I release any changes/mods I make to the source code." - which presumably means "The problem with code licensed under the GPLv3 or v2 is..." - really means "the problem with code licensed under the GPLv3 or v2 is that I can't control it the way I want rather than the way the person who wrote the code wants."
That's only a "problem with the GPL" in that the GPL exists, meaning it's available for developers to use; if you have a problem with that, it's really a problem with the author of the code, who doesn't want you doing your own work atop his or her work and then not letting those who get your software get the source code to your work as well as his or her work. That's entirely their right; if you don't like it, don't use their work, use the work of authors who are fine with that.
If your "secret sauce" is based on, or built atop, commercial code for which you'd have to negotiate a license to redistribute, so that some of your revenue has to go to the developer of the commercial code, are you going to complain that you're not going to make as much money because you'll have to pay the people who developed the commercial code in question?
Apple switching to a line of processors from a company for whom desktop/notebook computers is a major market, and that has a lot of money to throw at making those processors faster, was a bad move? How so?
and now the OS is starting to reflect the criticisms the Intel used to unfairly level on them.
"The Intel" used to unfairly level criticisms on them? Presumably you're not somehow speaking of criticisms from Windows users, as the UI has nothing to do with the instruction set architecture of the processor.
Regarding the Address Book, just go to the View menu and select "Groups". You'll be able to see your groups on the left and your contacts on the right, just like before.
"Before" I saw my groups on the left, my list of contact names in the middle, and the contact details on the right. What do I do in Address Book to see a display "just like before", with those three panes?
On the address book, if you double clip the bookmark ribbon (placeholder) graphics, you can see both contacts and groups in the left pane (Command + 3), although selecting one or the other will show you that specific view in the right side (Command +1, Command + 2, & Command + 3 toggle these views respectively).
Somebody at Apple had apparently forgotten, or never learned, about "discoverability".
The same is true of the 10.7 Address Book. It now looks like a real book (so, once again, slow page-turning animations rather than instance changes) and the two-page metaphor means that you can no longer see groups and individuals at the same time. Using groups to navigate is harder. I was going to say that they'd removed the groups functionality, but on closer inspection it is there just less discoverable
Bingo. That "click on the bookmark to switch between the two views" nonsense is a steaming crock of shit.
So what about general-purpose registers makes it difficult for them to implement those concepts
It takes more than a register and one instruction to implement heap allocation, lifetime management and garbage collection.
...which is why I asked; it's not as if a register can arrange to keep the upvalues around all by itself. Your most recent reply seems to say that it's not easy to implement a closure, register/instruction or no register/instruction, so perhaps the problem isn't that no CPU has a kind of register or instruction that easily implements a closure's upvalues - a special register or instruction might not help much, if at all.
Slashdot Mirror
Let me fix that for you: BSD is about the freedom to make choices for everyone in the distribution chain. GPL is about imposing restrictions ON THE USE OF THE CODE for everyone in the distribution chain.
BSD allows any stage in the distribution chain to impose whatever restrictions on distribution they want to upon all subsequent stages in the distribution chain.
GPL requires all stages in the distribution chain to impose the same restrictions on distribution upon all stages in the distribution chain.
This is the flip side of "BSD is about freedom of the distributor; GPL is about freedom of the code". Attaching "freedom" to one and only one of the licenses - regardless of whether it's a non-GPL-style license or a GPL-style license - comes across as an attempt to bias the discussion in favor of the freedoms offered by that license. Different people in the distribution chain may want different freedoms and restrictions at various stages in the chain.
I think the purpose of the GPL is to ensure that those that profit from your work also give back.
Nonsense. In fact, potentially completely the opposite. Someone who profits from distributing unmodified copies of your code doesn't have to give you anything back. But someone who tries to combine your code with code which they have previously written, and give away the combined result for free (or maybe even at a loss including their server hosting fees), is still required to send you a whole bunch of code unrelated to your work.
A commercial distribution license based on a percentage of sales would be more in line with your stated purpose. Or maybe a commercial license with a 100% discount for people who send you rights to use a number of lines of their own code equal to yours.
If "give back" means "give back directly to the "you" of "your work"", yes. If it means "give back to "the community"", not necessarily - you may not directly care about the code "unrelated to your work", but you might like your work to be used only in projects where the users can get the source, modify it, and redistribute the results under the terms of the GPL, so that somebody else could get code that's related to their work.
your example is wrong.
you would have to provide any improvement to the library you are using, but not code for other parts of the project.
If the library is GPLed rather than LGPLed, you would most definitely have to provide code for parts of the project linked with it, unless it falls under the "system library exception".
What is "profit from" also? If I borrow some internationalization code as part of a huge project, this saves the company money from having to buy a propriety product perhaps. More likely though it means we don't implement it ourselves and have less bugs down the road than if we rolled our own. Now does that mean we profited and the original author now gets access to every single line of our code as the GPL would imply?
No, it means that (whether you profited or not), all recipients of your huge project get to give that project away to anybody they want (under the terms of the GPL, so they can't further restrict its redistribution) and get to ask for and get the source to your huge project, and that anybody they give it away to gets the same rights.
Even if the code is proprietary and our competitors are anxious to get a peek at it?
Yes, in which case you would be advised not to use GPLed internationalization code. Sorry.
Even if various government agencies disallow giving away the code or allowing end user customization of the machines?
Yes, in which case you would be advised not to use GPLed internationalization code. Sorry.
With GPL the return payment is that you must also be GPL in absolutely everything you do. With BSD the return payment is that you give recognition to the author and keep the copyright notices intact. The first type of payment is too high for most companies unless they've got a software model that fits (ie, dynamic libraries, separately loaded programs, kernel modules, multiple cpus). The second payment is much easier but many companies don't know of it and they associate all free or open source software as GPL tainted.
Then the right answer isn't "complain about the GPL", it's "promote non-GPL-style licenses and code that uses them, so that more companies know about it and use non-GPLed software in those cases where it works better for them".
So the result is many smaller companies reinvent small pieces of code or libraries all the time, not the result desired if the author wanted to share code.
Perhaps the desire of authors of at least some GPLed code isn't simply "to share code"; perhaps it's "to share code in a fashion that keeps it, and any improvements to it, under the GPL", in which case that might well be the result desired - they'd rather have people reinvent the wheel than to use their wheel design in a fashion they don't like. People who want their code used even in cases where the GPL gets in the way, including people whose sole goal is "to share code", would presumably use a BSD license or an MIT license or some other non-GPL-style license. Perhaps smaller companies might not realize that they won't get trapped by the GPL if they use that software; if so, that is, as noted, an evangelism problem.
So if I invent a robot with my own ai code but it uses GPL apis
Presumably you mean "GPLed libraries"; I don't know of any interfaces that are GPLed in any sense, e.g. somebody could reimplement a GPLed library, duplicating all its APIs, and release it under the BSD license without violating the GPL. I guess you could see legal arguments about, say, header files for the library, but those arguments may already have come up and been resolved.
What if I onvested 5,000,000 making the code? Whoops competitors now use my code and undercut me because they didnt have to invest the 5,000,000. I go out of business.
Then whoops maybe you should have invested a little more to implement that functionality yourself, or paying somebody else to implement it, or even just to try to find a free-as-in-beer-at-least implementation with a non-GPL license, rather than using the work of somebody else who doesn't want you to use it in a fashion where you don't have to make source code available under the terms of the GPL.
Google gets a free ride because they are not redistributing. For everyone else who makes smart appliances you are screwed.
If a smart appliance maker is going to invest a lot of money in the software, they're not "screwed", they're just required to invest a little more money making or buying or finding non-GPL software. If their "secret sauce" is the hardware, and giving away the source to the software doesn't make it significantly easier for others to make that hardware - or if they have a patent on the hardware - they may not be "screwed" at all.
Small business owners are too. You cant sell your company as that too counts as redistribution.
A small business using GPLed software most definitely can sell their company. It's not as if using GPLed software GPLs your entire company, including the office desks and the software you bought to run on your computers and your customer lists and the software you've developed without incorporating GPLed code into it.
Just google router xompanies? Gnu went after them
...because they used GPLed code without complying with the requirements for doing so. A company that makes Wi-Fi access points managed to find an alternative to GPLed software for their access points.
Lets take your argument to the next level then?
Let's not, unless the developers of the GPLed software you're complaining about do so as well; otherwise, you're just constructing a strawman.
Do you use Visual Studio at work? Hell, did you use any include statement for any program you wrote while you were in school?
Therefore MS owns all your code right?
Wrong. If, however, you include any of the "Distributable Code" in your code when you distribute it, then you must do a number of things, including "for any Distributable Code having a filename extension of .lib, distribut[ing] only the results of running such Distributable Code through a linker with your program", "distribut[ing] Distributable Code included in a setup program only as part of that setup program without modification", and not "modify[ing] or distribut[ing] the source code of any Distributable Code so that any part of it becomes subject to an Excluded License" (where "Excluded License" includes, and is probably targeted at, licenses such as the GPL).
Similarly, if you use GCC at work, the FSF doesn't ipso facto own your code. If, however, you use GPLed "distributable code" as a part of your software, they impose constraints on its redistribution.
Microsoft's terms might be less problematic for some developers than the GPL's terms, but nobody who develops software is under any moral or legal obligation to avoid the GPL's terms - don't like those terms, don't use their code. Now, you might have to pay somebody if you're buying non-GPLed code from them to use in your product (e.g., getting a "developer license" from Essential Objects for their .NET widgets, so that you can redistribute those widgets in your application), or you might have to pay somebody to write non-GPLed code, or you might have to spend time and energy writing it yourself, or you might have to spend time and energy searching for non-GPLed free-as-in-beer-at-least code, but, well, that's life.
Under your own definition a simple link means that entity or person owns the code.
No. Under the GPL, a simple link means that the entity or person who owns the code with which you're linking gets to impose certain requirements on the redistribution of your code.
If I write something awesome for a specialized task with R&D then I should keep it and profit from it. Even if the toolkit was open source initially.
No, you shouldn't - you're using somebody else's code, so it's not as if that "something awesome" was written entirely by you. If the library with which you're working is a requirement for your software being "awesome", then some of the "awesome" is due to the other software (if it's not a requirement, then you don't have to use it). (And whether you "should" profit from it is up to your prospective suppliers and customers; if your prospective customers don't think your software is "awesome" enough to pay a price for it that covers your costs and desired profits, that's just Too Damn Bad, and if your prospective suppliers are not willing to supply you with tools/components/whatever under terms, whether financial or licensing, that allow you to make your desired profit, that's also just Too Damn Bad.)
Do you believe in property rights for software? If so, if you're honest, then you must believe in the rights of developers of software to say "you want to rely on the software I developed in your application at run time, even if it's just by calling it from a library, you have to make your application available under the same terms as my software". (If not, you're in no position to whine if you make no money because your software gets pirated.)
At that point people would be writing their own operating systems, apis, and so on if they wanted to own a
and somebody else distributes an improved version of it
FTFY.
Yes - to clarify, you can improve GPLed software and use it only by yourself and not have to make it available to anybody, and you can improve GPLed software and use it only within your organization and not have to make it available your organization. Thanks.
So your rights if you own the code are important too.
Yes. This includes the right to require that anybody who takes the code you own and improves it make those improvements available, in source form, to anybody who gets the improved version in binary form, and that they allow anybody who gets either the source or binary form from you be allowed to give it to anybody they choose. Yes, that might get in the way of the "anybody who takes the code you own and improves it" making money from those improvements; that's not the problem of the person who put the code under the GPL, it's the problem of the person who wants to make money from code that they didn't write in their entirety.
Yes I advocate the BSD license.
If that means "Yes, I prefer using the BSD license in code I write", that's fine. Even if that means "Yes, I think other people should use the BSD license", that's fine, but it imposes no obligation, moral or otherwise, on people to use the BSD license. If they choose to use the GPL for code that would be convenient for a project for which the GPL isn't convenient, that's your problem, not theirs - they're not under any obligation to make things convenient for you.
You post is FUD of the worst sort. The GPL in no way restricts freedom to profit from GPL code, it only restricts you from taking private the hard work of the original authors.
So... if I write some software... and release it under a non-GPL license... and someone else uses it... you claim that they can prevent me from giving it away to anyone else?
No, that's not the claim. The claim is that if you write some software, and release it under a GPL license, and somebody else makes an improved version of it, they can't refuse to make source to the improved version available and can't prohibit those who have received the source or binary of the improved version from giving it away to others.
I think a lot of developers see GPL as a "taking my toys and going home" license which discourages free use. If you weren't going to make a million dollar idea with your software, why stop someone else?
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?
Probably because they want their software to be used. I'm running a one man LLC and for my current work, if something relies on GPL then I simply can't use it.
You can't, for example, use Wireshark to look at networking problems on your machine if they're getting in the way of your work? Or do you just mean "I can't incorporate GPLed code in my current work"? Software can be "used" in ways other than incorporating it in other people's software projects.
Agreed. I think the shift has occurred because of increasing corporate interest in open source. BSD is seen as more corporate-friendly than GPL, when in fact it should be the other way around--BSD allows your competitors to reap the fruit of your labor without giving you anything in return. Start-ups, however, are lured by the idea of being able to close-source everything once their product becomes a smash hit,
There may well be start-ups who are. There are other start-ups who incorporate BSD-licensed code in their software and never open-source their software in the first place. One such startup succeeded rather well.
while established companies face genuine legal issues preventing them from linking GPL'ed code with closed-source code from vendors.
Unless it's an established company that doesn't link GPL'ed code with closed-source code.
What is the point of software being called open source when it is not usable by developers?
GPLed software is usable by some developers - those who, for whatever reason, have no problem giving source away. It's not usable by those who do. The point of licensing software under the GPL is to ensure that all work done (sufficiently directly) atop your work will be be as available to others as was your original work. I consider that every bit as legitimate (no more, no less) as choosing a license that allows others to restrict the distribution of work done atop your work. If somebody chooses a GPL-like license for their software, and that's inconvenient for somebody who wants to build something atop that software, that just means they'll need to do the work to replicate the functionality of that software, get somebody else to do so under a more convenient license, or find software that offers the same functionality under a more convenient license.
If you made something that could make you a fortune, pay for your house, student loans, and a company of your own then you would understand why.
Case in point? Look at earlier this week when someone invented something with cameras that no one else could do in /. stories? Problem was the toolkit is under GPL. How is that fair? There work was worth money but the GPL forced them to give away the algorithms to conpetitors so a whole clone of the toolkit was needed.
How is "so a whole clone of the toolkit was needed" unfair? If they used a toolkit written by somebody else, then it's not all "their work"; some of it is the work of the developer of the toolkit. You don't want to be bound by the requirements of some software, don't use the software. If that means you have to write your own replacement, so it goes; it's not the rest of the universe's responsibility to give you what you want.
No, it's worse. Raenex discovered an oversight in GPLv2's definition of "work based on the Program", meaning that if a software distribution includes at least one GPLv2 program, the whole distribution is a "work based on the Program". True, the FSF likes to claim that low coupling on a multitasking operating system turns "work based on the Program" into "mere aggregation", and this may be true of the GPLv3, but that's not what the letter of the GPLv2 says according to Raenex.
Raenex also says "Anyways, what really counts is what the license says, what the copyright holder thinks (only he has grounds to sue), and what a judge decides." (emphasis mine), so it's not settled (common) law yet that "if a software distribution includes at least one GPLv2 program, the whole distribution is a "work based on the Program"". (I don't know how this works in non-common-law jurisdictions.)
I can think of one company that provides "a software distribution includes at least one GPLv2 program", doesn't make source available to a huge chunk of that software distribution, and that has a large legal department that would probably weigh in on this, so I wouldn't assume things will necessarily go that way.
When people say 'Linux' they usually mean Linux-and-all-of-the-associated-cruft. Typically this at least includes GNU libc, GNU binutils, and GNU coreutils (which, between them, are more code than the kernel), and typically the GNU shell (bash) and GNU libstdc++. All of these have no moved to GPLv3 (in some cases with the runtime exemption).
GNU libc? Not so.
Referring to operating systems by the userspace software they use is ridiculous.
And referring to them by a subset of the userspace software.... (Admittedly, the C library is a very significant part of the userspace software, and it is from the GNU project on most Linux distributions, albeit not the one with the green robot as I understand it.
So "Linux", in the sense of "a Linux distribution", is not under any single license:
and so on.
Sorry, I screwed up the quote tags - "Boo ho, so write it yourself" should have been in quote tags; I'm quoting it and then violently agreeing with it. (And, no, that doesn't mean I'm in the "GPL rules, everything else drools" camp, either; I've worked at a startup where we did keep our "secret sauce" private, and where we explicitly avoided using GPL code, and I work on both BSD-licensed and GPL-licensed free software.)
Boo hoo, so write it yourself.
Precisely. "The problem with GPLv3 is that I can't use it in an application I develop unless I release any changes/mods I make to the source code." - which presumably means "The problem with code licensed under the GPLv3 or v2 is..." - really means "the problem with code licensed under the GPLv3 or v2 is that I can't control it the way I want rather than the way the person who wrote the code wants."
That's only a "problem with the GPL" in that the GPL exists, meaning it's available for developers to use; if you have a problem with that, it's really a problem with the author of the code, who doesn't want you doing your own work atop his or her work and then not letting those who get your software get the source code to your work as well as his or her work. That's entirely their right; if you don't like it, don't use their work, use the work of authors who are fine with that.
If your "secret sauce" is based on, or built atop, commercial code for which you'd have to negotiate a license to redistribute, so that some of your revenue has to go to the developer of the commercial code, are you going to complain that you're not going to make as much money because you'll have to pay the people who developed the commercial code in question?
No you do not have to give back to the community. If you use a GPL library in your application you owe nothing!
Presumably you mean "if you use an LGPL library in your application"; if you use a GPL library in your application, your application must be under the GPL. Now, if you only use your GPL-licensed application internally, you don't have to make the source available outside your organization.
Apple switching to Intel was a bad move
Apple switching to a line of processors from a company for whom desktop/notebook computers is a major market, and that has a lot of money to throw at making those processors faster, was a bad move? How so?
and now the OS is starting to reflect the criticisms the Intel used to unfairly level on them.
"The Intel" used to unfairly level criticisms on them? Presumably you're not somehow speaking of criticisms from Windows users, as the UI has nothing to do with the instruction set architecture of the processor.
Regarding the Address Book, just go to the View menu and select "Groups". You'll be able to see your groups on the left and your contacts on the right, just like before.
"Before" I saw my groups on the left, my list of contact names in the middle, and the contact details on the right. What do I do in Address Book to see a display "just like before", with those three panes?
On the address book, if you double clip the bookmark ribbon (placeholder) graphics, you can see both contacts and groups in the left pane (Command + 3), although selecting one or the other will show you that specific view in the right side (Command +1, Command + 2, & Command + 3 toggle these views respectively).
Somebody at Apple had apparently forgotten, or never learned, about "discoverability".
The same is true of the 10.7 Address Book. It now looks like a real book (so, once again, slow page-turning animations rather than instance changes) and the two-page metaphor means that you can no longer see groups and individuals at the same time. Using groups to navigate is harder. I was going to say that they'd removed the groups functionality, but on closer inspection it is there just less discoverable
Bingo. That "click on the bookmark to switch between the two views" nonsense is a steaming crock of shit.
So what about general-purpose registers makes it difficult for them to implement those concepts
It takes more than a register and one instruction to implement heap allocation, lifetime management and garbage collection.
...which is why I asked; it's not as if a register can arrange to keep the upvalues around all by itself. Your most recent reply seems to say that it's not easy to implement a closure, register/instruction or no register/instruction, so perhaps the problem isn't that no CPU has a kind of register or instruction that easily implements a closure's upvalues - a special register or instruction might not help much, if at all.