Allegro.cc - Online Community

Allegro.cc Forums » Off-Topic Ordeals » Free Software Foundation (FSF) - 2014 Giving Guide

This thread is locked; no one can reply to it. rss feed Print
 1   2   3 
Free Software Foundation (FSF) - 2014 Giving Guide
Dennis
Member #1,090
July 2003
avatar

I see I have not been thorough in my research. The statement that Linux would not exist without GNU remains wrong.

As I said earlier, my overview of the whole mess is incomplete as well. The history is probably only well known by those who are old enough to have been part of the whole development dating back all the way to the invention of programmable looms. But they probably have more important things to do to stick their heads in here to give us greenhorns an accurate overview of how things came to be.

Trent Gamblin
Member #261
April 2000
avatar

I used to be an FSF member but cancelled a few years ago. I still have the stuff they sent me. The fact is, free software is a good thing, but their attitude is hilarious and I don't take them seriously anymore. There are far more important things than source code but they believe code is God. I'm not against the GPL but I would never use it for anything I released or use any *GPL code in any of my software. Some GPL software is good (Linux itself, GCC, etc) but most of it (100% of end-user software) is awful.

As for ethics, free software is no more ethical than demanding the KFC recipe (free recipes!) It's a joke.

Arthur Kalliokoski
Second in Command
February 2005
avatar

It's easier to understand in the light of Stallman having freely exchanged source with others for years, only to have it stopped by companies wanting to increase the bottom line.

They all watch too much MSNBC... they get ideas.

Elias
Member #358
May 2000

Exactly. If you release code as MIT/BSD/zlib anyone can just take it and make it not free (you allow them to). All GPL does is try to protect the free-ness of (derivatives of) the code.

--
"Either help out or stop whining" - Evert

Trent Gamblin
Member #261
April 2000
avatar

"Me and the other ladies in town used to exchange our fried chicken recipes all the time. Why that's how I learned about adding a little bit of pepper to mine! Then KFC came along with THEIR fried chicken. The spices were just right and the meat was oh so tender. We asked the Colonel for the recipe but he said NO. I was so upset that I started the Free Recipe Foundation. Our chicken still ain't as good as KFC but one day all the worlds recipes will be free (libre) for everyone to enjoy just like God intended."

Elias said:

If you release code as MIT/BSD/zlib anyone can just take it and make it not free (you allow them to).

Yup. That's exactly why I use a similar license.

Elias
Member #358
May 2000

I'm also very much against the idea of patents (recipes might fall under that), for what it's worth. That is, the original purpose was fine, protecting inventions of inventors. But today all they do is hamper innovation and make competition against big companies impossible.

--
"Either help out or stop whining" - Evert

Trent Gamblin
Member #261
April 2000
avatar

Not having as many users != competition impossible.

Arthur Kalliokoski
Second in Command
February 2005
avatar

It seems like every time a good idea comes along, somebody eventually figures out how to game it to their own advantage through legal shenanigans.

They all watch too much MSNBC... they get ideas.

Trent Gamblin
Member #261
April 2000
avatar

I'm against most patents. To me that's completely different than free software. The difference is, with patents you are forbidden by law to make your own version of something somebody has made. With closed source software, you're still allowed to do that if it's not patented.

bamccaig
Member #7,536
July 2006
avatar

Dennis said:

Mr. McCaig, please do not write Linux when really you mean to write GNU+Linux for the scope of discussion in this thread because it does not help your request of others to do some research first before criticizing if you fail to do so yourself and by using the terms loosely like you do, you do not help to dispel the confusion around GNU, GPL and Linux but you are more likely to add to it.

Mr. Dennis, please cite where I have done so and I will correct it. I think the only time that I used Linux referring to the GNU+Linux operating system was when I referred to "a very large part of Linux distributions is GNU" (which was edited text: my original wording said "core", which was kind of nonsense when comparing the part that isn't the kernel) (but see below). It might have been misleading to have said GNU/Linux distributions since people may not realize that practically all "Linux" distributions are GNU/Linux AKA GNU+Linux distributions.

Dennis said:

Linux is not an operating system. It is a kernel, not more, not less. GNU is not an operating system either (yet, since it still lacks a stable kernel). The GNU Hurd (their kernel) has been in development since around 1990 and it does not look like it is going to be sufficiently bug free or feature complete for everyday use anytime soon. According to wikipedia even rms himself did not think optimistic about GNU Hurd in 2010 and claimed it was not crucial to finish it because a free kernel (Linux) already exists.

Linux is a kernel and drivers. GNU on the other hand is an entire operating system except for a functioning kernel [and drivers]. And yes, GNU Hurd is basically a research project at this point because the Linux kernel (and as I understand, also FreeBSD kernel) are free software (mostly) and already fill that niche. And I think the GNU Hurd people were overzealous with trying something different and barked up the wrong tree. That said, it is my understanding when last I looked that GNU Hurd does technically run, and so it is possible to have a complete GNU operating system running. However, Hurd is not considered production quality and obviously lacks many features and device drivers that are available with Linux.

Most GNU+Linux distributions ship Linux kernels with non-free software in them. The FSF project keeps track of truly free software distributions to prefer. I think that Debian is close, but I think that they still ship a non-free kernel (but a free kernel might be packaged up too). IIUC, you have to configure the kernel specifically to exclude non-free blobs when you build it if you're building it yourself. Thankfully, the kernel developers made that presumably easy to do.

Dennis said:

You claim that without GNU there would be no Linux. That is wrong. Linux was originally developed on MINIX and GNU applications later came in to replace MINIX components. So whereas it is true that Linux distributions owe much of their success to the availability of GNU software, the claim that without GNU there would be no Linux remains wrong.

I meant that Linux-based operating systems would not be complete as are the GNU+Linux distributions today. MINIX was a proprietary operating system until the year 2000. Running a free software kernel on a proprietary operating system is defeating the purpose since the system is still not free.

Perhaps the BSDs would have attempted to fill that niche, if anything, when they were eventually opened up, but then without GNU I'm skeptical that the BSDs would have ever been opened up (or MINIX for that matter). Additionally, had the BSDs been the operating system of choice for free software developers they probably would have abandoned Linux and used the BSD kernel(s) instead. It may have vastly reshaped the world we live in today. I was researching this recently and apparently the BSDs are developed as a centralized code base (i.e., the entire operating system in one source code repository).

Dennis said:

And the opposite is true as well, GNU software also owes much of their popularity, availability and distribution to the Linux kernel and Linux distributions because without the Linux kernel, there would not be any feasible way to even run GNU software.

That's debatable. Had it not been for Linux the GNU project would have needed a kernel developed, as would everybody else in the world. It's possible they could have used a BSD-derived kernel, or that GNU Hurd would have been raised to fruition instead (the Linux kernel basically drained all of the qualified and interested people, making them unavailable to help out with Hurd; I think that might have been due to mismanagement from GNU in the early days by keeping the kernel secret instead of releasing and inviting help). I highly doubt that not having a kernel would have indefinitely stopped GNU. The GNU software was already distributable, and it was already UNIX compatible. It could have run on a non-free UNIX (as it presumably did and does). It just wouldn't have satisfied the free software goals of Richard Stallman, the GNU project, and its brethren until a later date when a free kernel was otherwise developed.

Dennis said:

So there is a tight coupling going on between GNU and Linux, they both benefit from each other and it is my wish for you to stop painting a black and white picture of the situation where you seem to demonize one side and praise another. We live in a colorful world where there is no clear distinction between "good" and "evil" as you seem to would like to believe there is.

I do not demonize Linux. Linux is for the most part amazing. The only problem is that its lead developers/overseers do not personally care about preserving its software freedoms. I think Linus Torvalds is a great man. I love it when he tells all the idiots around him that they're idiots. It's almost like a toy Dr. House in the hacker community. That said, it is problematic for me that the Linux kernel is accepting non-free binary blobs. I'd much prefer them to reverse engineer the blobs and release the code into the wild on the net separate from the Linux kernel. Or just go without support for devices whose vendors do not permit free software to power them.

Dennis said:

The FSF is not the only player around in the field of free software. No one here says they are not important and I believe much of the criticism towards the GPL is well justified for reasons that have been pointed out/repeated enough times by now. I'll add another one: it is way too wordy (which is probably one of the main reasons it confuses people and lets them get wrong ideas about it).

Now... I could be wrong but I think the MIT license and the GPL are even compatible (would have to do a thorough check/research on that though) and I do not think that if I release anything under the MIT license or use anything 3rd party under that license that there could be any occurrence of future freedom-restriction towards my code or the code that I re-use.

The GPL is as wordy as it is because it was actually drafted with the assistance of attorneys with the intent of being upheld in a court of law just like any proprietary license is. It isn't enough to write down "you can do whatever you want, you're welcome." A court of law has to interpret that in legal terms, and it is vague enough that it can be twisted around any way somebody sees fit. It then becomes the job of a judge (perhaps a bought one) to decide the outcome in court for themselves. That is the problem with those simpler licenses. They may not hold up in court which threatens everybody's safety by possibly enabling a rich and powerful entity to try to take over the software.

"MIT license" is ambiguous since there have been many. The FSF recommends these licenses for simple programs for which you don't care about copyleft. Apparently the FSF does not recommend their use for substantial programs because they don't protect users from patent treachery. The simplicity comes at a cost. Instead for those people that don't care about copyleft they recommend the Apache 2.0 license which is still permissive, but does have clauses to protect users from patent treachery. Note that it is not compatible with GPLv2, but it is compatible with GPLv3.

Compatibility is one-directional. It just means that software licensed under a compatible license can be used or distributed under the GPL of applicable versions. It does NOT mean that GPL software can be used or distributed under the more permissive license terms.

Law is difficult. Anybody opting to use a simpler license without legal advise is putting their users at risk, and depending on the license perhaps also themselves. To be fair, if you really want to be safe you should opt to get legal advice regardless, but the FSF purports to do its best to handle that step for you.

Dennis
Member #1,090
July 2003
avatar

Thank you for the extensive elaboration.

When people refer to the MIT license these days, they are in general referring to this one:
::start of license::
The MIT License (MIT)

Copyright(c) 20xx, <your name here>

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
::end of license::

I marked an important part in bold letters.

I frankly do not see how anyone could ever retro-actively take away or restrict mine and my users freedom for anything that I or anyone else releases under that license. It is short, precise, easily understandable and pretty clear on the "and to permit..." part.

In other words, if you use software or parts of software originally released under that MIT license in your product, you must give your users all the same rights for that part of your software but you still have the freedom to add other parts to your software(separate from the original re-used parts under the MIT license) which you may keep for yourself and put under any license you wish but that does not ever have the power to change the license for the re-used parts upstream and not even downstream either.

That latter piece of freedom (to choose a differen license for your own parts) is the one that the GPL denies, so the claim "free as in freedom" is actually a false claim (I feel like I am repeating myself).

On a loosely related sidenote: I am not seeing (m)any job offers for paid "free software development" positions, so I find it difficult to believe it would be possible to earn a living in doing so (I personally find the whole idea that everyone in this day and age has to "earn" a living and is controlled 100% by money and decisions purely based on "cost" and that the "needs" of financial systems are being considered more important than the needs of people ridiculous (and governments/politicians are quite powerless to make any changes because they are all controlled by those systems as well) and un-ethical as hell but that's a whole different story and does not have much to do with free software. The commonly asked question "and who is going to pay for it?" is on a global scale the single most anti-life/anti-social question anyone can ever ask. The question should always be "what is needed by people and how can we achieve it?" and the goal should be for people to work less and smarter, not more and harder... I will cut myself off here, don't want to ramble on about this the whole night, just one more thing: Money is fiction, people are real.).

bamccaig
Member #7,536
July 2006
avatar

Dennis said:

That latter piece of freedom (to choose a differen license for your own parts) is the one that the GPL denies, so the claim "free as in freedom" is actually a false claim (I feel like I am repeating myself).

The GPL guarantees that the software remains free (libre) for users in every way that matters for them personally. The source must be made available, and it is permitted to modify the software, in source or binary form, and distribute exact copies or modified copies. Everyone will always be free to use, modify, and distribute copies and modified copies of the work regardless of whom derives from it, under these same terms.

The MIT license basically just says, do whatever you want, but distribute a copy of this license text with copies or modified copies the work.

You are basically arguing that:

  • The GPL licenses are not really about freedom because a user cannot choose to use or distribute somebody else's work or a derived work under his own terms.


  • The MIT license(s) are free because they permit anyone to do whatever they want, including revoking the freedom of users to use a distributed copy or modified copy of a work.

Note how the GPL licenses serve to preserve the freedom, and the MIT license(s) do not.

The MIT license doesn't even require you to release the source code or distribute copies or links to the original code. Neither the original nor modified source code. Merely including a copy of the MIT license text and copyright notice. Of course, if that text doesn't apply to your software as a whole, it loses meaning if you just find that text somewhere within a project.

your-program --stupid-no-longer-applicable-license
The following license does not apply to this software.

<MIT license text>

The GPL does not prevent authors from dual-licensing their works. In the event that you want to use GPL'd work without being bound by the terms of the GPL you could always contact the copyright holder for permission to use the work under different terms. Of course, a well versed author might refuse to grant such permission to you if they really understand the purpose of the GPL. At least unless you make it worth their while, which probably means paying money, which you probably can't afford and don't want to do, though if you care about the terms of the GPL you probably intend to ask of somebody else; and/or might limit your use of the work in some way that still makes it unusable by you, for example by not allowing you to modify, distribute, or derive from it.

I have no ill will towards the people with opposing ideas. I'm simply trying to explain my understanding of the open source licensing situation, and the logical fallacies that I believe are being used to slander the GPL license(s). I would encourage anybody with distaste for the GPL to study the writing and speaking of Richard Stallman himself. He addresses concerns, and also presents the ideas to new audiences regularly so he is not unaware of your own concerns. He has good reasons for creating these licenses and fighting for their adoption.

As for the digression on currency and slavery I think we are more or less on the same page there, except that I'd argue that representatives of governments are able to make changes, but the vast majority of those in government have no interest in doing so because the existing systems take good care of them and their people. That isn't universal though. Members of the Green Party of Canada show a healthy disdain for the corruption of government and positive steps to limiting the power of those in government (unfortunately, they don't have the power alone to enact these changes, and Canadians are too lazy to even pay attention so that they can even notice a problem, let alone acting to solve it). It occurs to me that if the majority of registered voters in Canada do not vote it should be taken as "none of the above" and the election should be declared incomplete and the function of government should be suspended until a fair election with a majority of votes for a particular representative (with non-voters still counting towards a virtual candidate of "none of the above" ad. infinitum).

Trent Gamblin
Member #261
April 2000
avatar

I got 4 letters in the mail today. Guess which ones I was happier to receive. :-*

{"name":"609063","src":"\/\/djungxnpq2nug.cloudfront.net\/image\/cache\/5\/c\/5c06fd7f606df015967c2f247eef640c.jpg","w":3456,"h":3456,"tn":"\/\/djungxnpq2nug.cloudfront.net\/image\/cache\/5\/c\/5c06fd7f606df015967c2f247eef640c"}609063

{"name":"609064","src":"\/\/djungxnpq2nug.cloudfront.net\/image\/cache\/c\/f\/cf7363d1c953e1301d828fc9385237f7.jpg","w":3456,"h":3456,"tn":"\/\/djungxnpq2nug.cloudfront.net\/image\/cache\/c\/f\/cf7363d1c953e1301d828fc9385237f7"}609064

pkrcel
Member #14,001
February 2012

It's easy, the ones that were written by an actual sender, I guess :)

It is unlikely that Google shares your distaste for capitalism. - Derezo
If one had the eternity of time, one would do things later. - Johan Halmén

Dennis
Member #1,090
July 2003
avatar

bamccaig said:

You are basically arguing that:

  • The GPL licenses are not really about freedom because a user cannot choose to use or distribute somebody else's work or a derived work under his own terms.

  • The MIT license(s) are free because they permit anyone to do whatever they want, including revoking the freedom of users to use a distributed copy or modified copy of a work.

Note how the GPL licenses serve to preserve the freedom, and the MIT license(s) do not.

That is wrong on two levels. Wrong because you are claiming I say things which I do not say. Wrong because of false facts about the MIT license and about the GPL license.

what I am actually saying is:
I am saying the levels of freedom granted by the GPL are incomplete(reread the thread from the start if you don't believe this to be true) and since the meaning of "freedom" should not be watered down(restricted) like that(because restrictions of any type basically go against the very definition of freedom), "free as in freedom" is a false claim. It may sound nice on the surface but if you look closely it does not hold up to match the definition of freedom.

The MIT license does not include the right to revoke any already given freedom of users to use anything. It even explicitly and impossible-to-misunderstand says "and to permit... to do so".

The MIT license does not at all say anywhere to "do whatever the fuck you want"(like you seem to think it does when it does not). It is in fact pretty clear about which freedoms are granted and does not allow anyone to take those freedoms away (that's where you seem to misunderstand the MIT license as you keep claiming that anyone could basically at any moment revoke any rights previously granted by stuff under MIT license which is, again, not true: There is no possibility for revoking rights anywhere in it.).

I have a feeling our positions on this matter are set in stone for now and the exchange of thoughts is running in circles which is why I no longer find it desirable to continue on this topic. I learned a lot about the GPL and the FSF in the process.

Quote:

I have no ill will towards the people with opposing ideas. I'm simply trying to explain my understanding of the open source licensing situation, and the logical fallacies that I believe are being used to slander the GPL license(s).

Neither do I have an "ill will" nor do I think anyone else here who criticizes the GPL/FSF does (seriously, where do you get those ideas?). Likewise, I am trying to explain my understanding of the open source licensing situation and the logical fallacies that I believe are being used to blindly worship the GPL, the FSF, rms and slander everyone else who does not agree with it/him and spread false claims and instill fear about other licenses such as the MIT license.

I will happily continue to use GNU based software in my daily use of computers and I am, like I said before, very thankful for all the tools they provide and for the existence of open/free software in general. I just will not use GPL'ed components in my software or use the GPL for my software because I like my freedom and I do not want to impose unreasonable restrictions towards the freedom of others who might be interested in (re)using my software.

My opinion may even change some day, maybe with a new revision of the GPL, one which respects ALL levels of freedom and does not force certain rules upon any new software which re-uses any existing GPL'ed components. Maybe I still do not fully understand the GPL though... that's still a possibility.

Elias
Member #358
May 2000

Dennis: You misunderstood the MIT license:

"Permission is hereby granted... to deal in the Software without restriction, including without limitation the rights to... sell copies of the Software, and to permit persons to whom the Software is furnished to do so..."

It grants you, the receiver of software with the license, the right to permit others the same rights. It's the right to sub-license basically, which for classic licenses often is not the case and so they felt they had to explicitly mention it. In no way does this imply that you have to or even should do so. It is perfectly fine to put a more restrictive license on and then sue whoever makes modifications to the original software :(

--
"Either help out or stop whining" - Evert

Dennis
Member #1,090
July 2003
avatar

Nah, I think you misunderstood it because you stopped reading too early when it was about to come to another important fraction of the text which explicitly mentions something which, if done so as the license demands, will grant all the same rights and permissions upstream (what it does not demand though is that you grant and permit the same rights downstream for the parts that you add which must be separate from the re-used(under MIT license conditions) upstream parts).

Elias said:

It is perfectly fine to put a more restrictive license on and then sue whoever makes modifications to the original software

I do not think so (about the retroactively putting on a more restrictive license to the re-used parts and about the sue-ing (I mean, you can try but the license is pretty clear on that being a violation of the terms), because of "and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software."

In other words, if one uses anything under the MIT license without granting the same permissions (due to the "subject to the following conditions..." part), they are in direct violation of the terms of the MIT license and therefore not allowed to use those pieces of software in the first place.

They can not possibly restrict use of what they were are not allowed to use by violating the original terms of usage.

update/append:
I will prepare a structured view of the important parts of the MIT license here which will make this even more clear.

full text passage:
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

structured text passage:
PREFIX, ENUMERATION(of permissions/rights), CONDITIONS(under which the rights are granted)

PREFIX:
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"),

ENUMERATION:
to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so

CONDITIONS:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

Note these two and very important statements about the conditions:
a) if you do not follow the conditions you are not granted any of the rights in the ENUMERATION thus are not allowed to use the MIT licensed piece of code at all

b) if you do adhere to the conditions (most importantly by including the copyright notice and by including the same permission notice) you are allowed to use the MIT licensed piece of code and by including the license/permission notice recursively grant your users the same rights/permissions for the re-used portion of MIT licensed code

There simply is no loophole there which would allow anyone to revoke/restrict the granted rights and permissions for the re-used pieces upstream for then they would be in violation of the conditions of the license and so are not allowed to re-use the pieces from upstream which they are trying to restrict so the attempt to do so would be simply, well: illegal.

Elias
Member #358
May 2000

My scenario was, you write software A, with MIT license. Someone modifies it with modification B to create a new software A+B, released with a restrictive license, for example non-commercial (or even something like GPL).

Now you incorporate the same modification B into your original project. And they sue you for copyright violation as you do not have the right to use B. (Or license violation since you can't re-release GPL code as MIT.)

And so my fear is that once I release something under MIT, and someone happens to make a bug fix or add a feature before me, I couldn't fix the bug or add the feature anymore without violating their copyright/license.

--
"Either help out or stop whining" - Evert

Dennis
Member #1,090
July 2003
avatar

I think that fear is basically the same one which bamccaig has about it.

A bugfix or a feature falls under the grant for the right to modify. And if one does not include the permission notice as required by the conditions, one is not granted that right to modify in the first place.

So if someone makes a bugfix or adds a feature, they basically have to allow anyone(by inclusion of the unmodified copyright and permission notice) to add the same bugfix or feature upstream or they'd violate the original terms of the license.

I think the MIT license is solid in that regard of protecting your freedoms and rights. It is (as is the GPL) an OSI approved license ( http://opensource.org/licenses ) and that means it fulfills the conditions of the OSD ( http://opensource.org/osd ).

Well... of course, some paid(not to say bought) lawyers and judges might (voluntarily)interpret things differently in court but it clearly is not my fault if they fail at even the most basic elements of propositional calculus and so I would personally find their ruling in favor of an unwarranted suing process to restrict anyones open source software rights/freedoms ridiculous and I would be unable to take said lawyers and judges seriously and I would suggest or rather demand of them to switch career for something which does not require the ability to read, understand and apply logical connectives.

Elias
Member #358
May 2000

Dennis said:

A bugfix or a feature falls under the grant for the right to modify. And if one does not include the permission notice as required by the conditions, one is not granted that right to modify in the first place.

But the permission notice only applies to the original, not to the modifications. Yes, you are required to include the original license. But that doesn't mean that it will apply to anything but the original code.

--
"Either help out or stop whining" - Evert

Dennis
Member #1,090
July 2003
avatar

Elias said:

Yes, you are required to include the original license. But that doesn't mean that it will apply to anything but the original code.

But it kind of does as the license explicitly says the rights/grants/freedoms and conditions apply to "the software" and not just to "a specific piece of" the software.

So if you use A and combine/bugfix/modify/feature enrich it with B, you can not exempt B from the rights/grants/freedoms conditions of A by putting A+B under a more restrictive license, because A+B becomes "the software" then.

What you could do at most would be to use an unmodified A and your own B which makes use of A (further refered to as the B(A) way of re-use) and then put B under a different license and include A under the original license but in that case (putting B under a different license), you are not allowed to modify/bugfix/feature enrich A itself (because that would violate the conditions of the license of A), so the freedoms of the author of A are protected and the freedoms of the author of B are protected as well as they are not required to use the same license as A in the B(A) way of re-using the software.

The GPL on the other hand, as I understand it, requires both the A+B type of reuse and the B(A) type of reuse, to also be made available under the GPL terms thus limiting the freedom of the author of B in the B(A) re-use scenario.

Feel free to correct me if you think I got that wrong.

bamccaig
Member #7,536
July 2006
avatar

Elias said:

And so my fear is that once I release something under MIT, and someone happens to make a bug fix or add a feature before me, I couldn't fix the bug or add the feature anymore without violating their copyright/license.

It wouldn't necessarily prevent you from fixing the same bug or adding the same feature, but if their license wasn't compatible with yours (e.g., theirs is some version of GPL and yours is some variation of MIT) then you wouldn't be able to copy their changes exactly. You'd have to reinvent them yourself. And if you were afraid of them coming after you you would have to take care to make sure there was no question that you didn't take it from them. That said, MIT says nothing of software patents, so presumably even if the other software is MIT licensed you might still not be able to implement their bug fix or feature if they happen to own a software patent for it (even by producing your own changes!).

Dennis said:

A bugfix or a feature falls under the grant for the right to modify. And if one does not include the permission notice as required by the conditions, one is not granted that right to modify in the first place.

So if someone makes a bugfix or adds a feature, they basically have to allow anyone(by inclusion of the unmodified copyright and permission notice) to add the same bugfix or feature upstream or they'd violate the original terms of the license.

I think the MIT license is solid in that regard of protecting your freedoms and rights. It is (as is the GPL) an OSI approved license ( http://opensource.org/licenses ) and that means it fulfills the conditions of the OSD ( http://opensource.org/osd ).

You're erroneously arguing that MIT licenses are "infectious" like the GPL. MIT licenses are not, which is exactly the reason that some people prefer them. It gives users the freedom to change the terms for their own users.

Dennis said:

Well... of course, some paid(not to say bought) lawyers and judges might (voluntarily)interpret things differently in court but it clearly is not my fault if they fail at even the most basic elements of propositional calculus and so I would personally find their ruling in favor of an unwarranted suing process to restrict anyones open source software rights/freedoms ridiculous and I would be unable to take said lawyers and judges seriously and I would suggest or rather demand of them to switch career for something which does not require the ability to read, understand and apply logical connectives.

Lawyers and judges don't care what you consider fair or logical. They care about the letter of the law and precedence, much of which is created by corruption and money. If you choose to disregard them then you might as well just abuse GPL'd software from day zero too and not worry about it. It would be the same thing.

Dennis said:

But it kind of does as the license explicitly says the rights/grants/freedoms and conditions apply to "the software" and not just to "a specific piece of" the software.

So if you use A and combine/bugfix/modify/feature enrich it with B, you can not exempt B from the rights/grants/freedoms conditions of A by putting A+B under a more restrictive license, because A+B becomes "the software" then.

Your interpretation of MIT licenses is as restrictive as GPL. You would not be able to use it for traditional commercial software because you'd be required to grant all of these rights to everyone (i.e., use, modify, distribute). That is not how MIT licenses work. MIT licenses do not enforce any such thing. The only thing they enforce is that the original user of the original work has those rights and that you include a copy of the original license and copyright notice (not that they apply to your own work or derivative work).

See here: https://tldrlegal.com/license/mit-license

By definition of the MIT licenses being compatible with the GPL licenses, that means that I can take your MIT licensed work and relicense it under the GPL. You would not be able to touch any modifications that I made without my GPL infecting your work. Conversely, your MIT licensed work will not infect my work.

Append:

And here's everything that GPL v2 and v3 does (as far as I can see, v3 just seeks to make it explicit that you're free to use the licensors' patents without worry):

https://tldrlegal.com/license/gnu-general-public-license-v2
https://tldrlegal.com/license/gnu-general-public-license-v3-(gpl-3)

Dennis
Member #1,090
July 2003
avatar

bamccaig said:

You're erroneously arguing that MIT licenses are "infectious" like the GPL.

No I am not. Clearly if you can not comprehend the fine differences in the different ways of re-use and their respective implications about what one is and is not allowed to do in each of the licenses then that is not a basis on which it is possible for me to continue this discussion with you. I do not actually think you are incapable of understanding it though, it is more like you do not want to understand it for the sake of argument and frankly I am tired of the way you keep suggesting that I am saying things which I am not saying.

Quote:

Lawyers and judges don't care what you consider fair or logical. They care about the letter of the law and precedence, much of which is created by corruption and money.

And as I already said, it is impossible to take such lawyers and judges seriously.

Quote:

If you choose to disregard them then you might as well just abuse GPL'd software from day zero too and not worry about it. It would be the same thing.

No that is a different thing entirely. I respect very well what the GPL does to software licensed under it, which is exactly the reason I avoid it.

Quote:

Your interpretation of MIT licenses is as restrictive as GPL.

Only in one out of the two described different cases of software re-use.

Quote:

hat means that I can take your MIT licensed work and relicense it under the GPL. You would not be able to touch any modifications that I made without my GPL infecting your work.

So much for the GPL protecting and promoting that level of my freedom. Thank you for proving my point.

bamccaig
Member #7,536
July 2006
avatar

Dennis said:

So much for the GPL protecting and promoting that level of my freedom. Thank you for proving my point.

It isn't your (1) freedom the GPL serves. It is everyone's freedom (N). It is impossible to give everyone absolute freedom. Necessarily the freedom of one collides with the freedom of others. It is absolutely nonsense to argue that the GPL is not freedom-centric because of this.

For example, if you were absolutely free to do whatever you wanted then you could pick up a brick and beat an infant to death with it (regardless of what it would say about you to want to do it). It sounds like you are arguing that laws that prevent you from doing so are anti-freedom... When really they serve to prevent you from encroach upon the rights or freedoms of others.

The GPL exists to protect people generally from evil corporations. The MIT licenses do not accomplish this at all. A selfish, greedy entity can take MIT licensed software, make some malicious modifications to it, and release it against unsuspecting users. They have no obligation to share the code with us and they can restrict our right to reverse engineer their code by law so we have no legal way to check if they do or don't have malicious features. Albeit, these days you can just assume it, but you can't prove it so the vast majority of people remain oblivious victims.

Whereas the GPL would force the entity, by law (to the extent it can), to release their modifications and allow us all to see these malicious features. We'd be able to remove them, and also be able to hold the entity accountable. The entity would not even attempt to add those features knowing they would not be allowed to hide them from us. The GPL exists to let we the people beat selfish, greedy entities (typically corporations) at their own game. Primarily, it exists to free our computing.

Elias
Member #358
May 2000

What I've seen projects using MIT code do is just have something like "Parts of this software use code from A, see A_license.txt for its license". Which then has a copy of the original license. But it does not say which code exactly falls under it. And that's all the MIT license requires, retain the original license text somewhere. But mainly for the copyright notice, as you are free to license the derived software under any license you wish. (You don't have to include any source code at all either, but if you do, it all is under the new license not under MIT.)

--
"Either help out or stop whining" - Evert

 1   2   3 


Go to: