It's not about giving credit. It's about the specific license. I will never use GLP, LGPL, CC, or any other viral license. I do not wish to impose a license choice onto anyone else's code just because they choose to use my code.
I always give credit wherever possible, if I fail to give it, it's just a mistake that I will fix if notified.
As you see this software copyright debate has been gong on for ages. You could have seen most of the points raised over the last few days on these forum threads on slashdot a decade ago and repeatedly since. Always the same confusion about patents, copyright, trademark, always the same views re this license or that, and so on and so on.
As for copyrighting numbers, that argument has been used to suggest that is impossible to copyright any kind of code for a long time. CSS was just a famous example. I fail to see why a piece of software should become uncopyrightable because it is a prime number as opposed to any other number.
As for the GPL vs MIT vs whatever. It just depends what your aims are in releasing the code. Choose accordingly. GPL is great if your aim is to make a vast body of software permanently free and open for the benefit of the human race as a whole for generations to come. That's a big project that many are not willing or able to contribute to, or don't see the need for, or "oh my God it's communist". MIT is great if you just want to get functionality out into the world for the benefit of all and don't care if they use it as open source or not. Draconian closed licenses are OK if you want to try and make money.
The interesting part is that I doubt any person who has participated in this thread has had their viewpoints altered an inch.
I seriously doubt that Bill's intentions were anything but positive, although I do wish he'd reconsider his license. Being recognized for your coding hasn't ever been any issue here as far as I can tell. I've used a lot of MIT code in various projects and try to go out of my way to make sure everyone is mentioned somewhere both in the code itself and in any related materials.
We've drank a lot of coffee and kicked this to almost nausea. It's time to get back to doing great things with the Prop. (Either Prop)
One argument that I heard against the MIT license is as follows:
Example, someone can take an MIT licensed program, re-license under a restricted license, and try to stop the original author from using the original work. The "Creative Commons Attribution-ShareAlike 3.0 Unported" puts a stumbling block into that process.
However, I don't believe this is correct. The MIT license states that "The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software." So doesn't that mean that any copies or modified versions must also be under the MIT license?
Who knows? I have a headache over it. That ccwiki page is good reading though. I have a lot of respect for Lessig (the principal), whose politics sometimes annoy me, but his intent generally doesn't. The focus of that effort was culture, not code, and IMHO that says a lot about the organization and it's intent to add value not make a bigger mess of things.
I will never use GLP, LGPL, CC, or any other viral license. I do not wish to impose a license choice onto anyone else's code just because they choose to use my code.
I suspect commercial licenses are usually viral, ie: Thou shalt not allow thy knowledge to be disclosed via any means.
I always give credit wherever possible, if I fail to give it, it's just a mistake that I will fix if notified.
The concern is not what you, or anyone else with noble intent, would do but what someone less so inclined may do. It's a big game of "What if?'s" in the commercial world.
... The MIT license states that "The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software." So doesn't that mean that any copies or modified versions must also be under the MIT license
I'd guess that, yes, the MIT license does have to be included, but, since it doesn't require source disclosure it can still be included with either closed or open source. Eg: Even with a CC-by-sa license on the surrounding code of a single program, the MIT code can be lifted out and reused in a closed source program.
The piece of MIT code, in my example here, would not be a derivative. It would be like the code snippets one can pick up from collections. The Parallax forums, and OBEX, being examples of such collections.
I am all too familiar with the commercial software world, having been doing commercial software of over 20 years now. Commercial software developers generally don't assign a license to their code, instead they just keep it private. If they want to get help from an external source it almost always involves and NDA. Most commercial entities avoid GPL or similar licensed software, and even go so far as to have policies that forbid their employees from viewing any such code except in special cases.
There is a great deal of code I have written that I can never share, and it's kind of a bummer, but it's part of my job.
For me, most of my prop related code is part of my hobby. The few places where it's not part of my hobby happen to be open source situations (the C/C++ Spin/PASM compiler is an example). I choose to use MIT, and I also choose avoid working on things that involve GPL licensed code. For example, I only did do a small amount of contribution to the propgcc effort, mostly feedback and comments to the guys doing the actual work. I have only looked at a small portion of the code involved, mostly the loader and kernels. Since I changed jobs, I am more free to contribute in the future efforts for Prop2, and I hope I can help, but I'll still be focusing on the non-GPL stuff. Can you tell that I really loath GPL?
Treating an NDA as not a license is just splitting hairs. It is used for the same viral purpose - to prevent knowledge spreading.
EDIT: That is unless the NDA has a reasonable expiry date and no further restrictions beyond that expiry. Lol, I've never read any I signed, which isn't many.
GPL and the likes is for commercial use. You guessed it, it has the opposite effect - to spread the knowledge commercially.
The way I understand it the MIT notice must remain intact in the source file, even in a closed source program. Of course, it would normally not be seen by others unless there was some sort of legal action that caused the code to be audited.
It seems like the attribution aspect is the sticky point for some. The MIT license states that the "above copyright notice ... shall be included in all copies or substantial portions." I interpret that to include the copyright line that most people put at the beginning of the file. That would cover attibrution in the open source case, as long as the author isn't different than the copyright holder. However, in a closed-source situation there is no requirement to disclose the copyright holder, which may be a problem for some people.
evanh,
Most of the commercial code I have worked on had no NDAs written for it since we had no external people work with it. The cases I can recall where it did have an NDA was with another company (such as Intel, nVidia, or AMD) to get them to help us find issues with their drivers/hardware. NDAs are much more than a license because they typical specify specific people who are included, specific time periods, and often more.
I suspect commercial licenses are usually viral, ie: Thou shalt not allow thy knowledge to be disclosed via any means.
I have been trying to understand you for an hour now and which ever way I read the above I still don't know what you mean.
Commercial licenses are anything but viral. You license the software from the vendor and that is pretty much the end ofthe chain, it cannot reproduce and spread from there. Also, what knowledge? You didn't get any source code with that commercial license. You only got an executable. Nothing to learn from that.
Treating an NDA as not a license is just splitting hairs. It is used for the same viral purpose - to prevent knowledge spreading.
There it is again. How is preventing the spread of knowledge viral? How is preventing the spread of copies of even executables viral? Sounds like the
total opposite to me.
Viral is normally a description of something reproducing and spreading from host to host. Commercial software licenses are exactly designed to stop that.
All I can think is that you actually meant vile, as "...vile purpose - to prevent knowledge spreading" That would make sense.
All the extra bits in a NDA aren't relevant. What's relevant is the non-disclosure part, and a possible time limitation on that.
As for in-house only, that hardly counts as you are implying no knowledge transfer at all. The question on this becomes: what is the rule if you wanted to make that code public?
Cripes Heater, we're talking about developer to developer interchange, not end user. When I say commercial here I'm talking about things like licensing sources to glue into a bigger project. Or maybe collaborative development work.
There is knowledge passed between those two commercial entities. The viral part is how that plays out down the developer chain, the knowledge is force into a tight control because the process is forced into repeating by it's very demands.
What's sticky about the MIT attribution? It's only required in the source code itself. There is no effort required to maintain that attribution.
By "sticky" I mean that it's a concern for some people. A lot of people don't care about attribution, and provide the code so it will be freely used by others.
As I said in my previous post, the MIT license provides attribution in the source code if the source is open, and the copyright line is included. I interpret the wording of the MIT license to require the "Copyright (c) <year> <name>" line. This is clear in some versions of the MIT license, but not so clear in the way it is used with Parallax.
By "sticky" I mean that it's a concern for some people. A lot of people don't care about attribution, and provide the code so it will be freely used by others.
So the concern is from the author, is that right? As in, some authors don't want to be known? Simple answer to that concern is don't give a name or lie. Who's gonna care after that?
Most commercial entities avoid GPL or similar licensed software, and even go so far as to have policies that forbid their employees from viewing any such code except in special cases
I know how that goes, but have you ever wondered how insane those policies are?
Imagine a musician gets a deal with a record company for 5 new albums and as part of the deal is forced not to listen to music for the term of the contract?
Or an author finds she can't find a publisher for her book because they all suspect she has been, well, reading books by other publishers.
I don't belive copyright prevents you from using ideas from within the works of others. If that were so there may only have been allowed one blues song, one jazz tune, one rock'n'roll song etc etc. There might only ever have been allowed one detective novel, or one cow boys and indians movie.
Of course copy large junks of code is off the table but reading it, no, I would not tolerate such a prohibition from an employer. It's nuts.
Can you tell that I really loath GPL?
I'm curious. Why do you "loath" the GPL? I can understand selecting this license or that depending on your intentions for your work. But loath is a gut reaction not a logical appraisal.
So the concern is from the author, is that right? As in, some authors don't want to be known? Simple answer to that concern is don't give a name or lie. Who's gonna care after that?
Yes, the concern is from the author. However, it was more for the case where the author is concerned that someone else will take credit for his work, or people won't acknowledge that he came up with the idea. The copyright line provides attribution.
... I don't believe copyright prevents you from using ideas from within the works of others. ...
Yup. As mentioned at least twice now
Copyright only protects copy. In music, only the melody in a song is required for copyright ... "Happy Birthday" anyone? The term "substantial" in the MIT license is fairly vague. Considering the number of ways available to achieve the same end, it beats me as to why copyright is really even valid for software unless it's completely a verbatim copy. Since RAP doesn't have a melody, it seems to be an escape for the music industry and thus grows like weeds. I'm no authority and do not want to be ... it's just common sense to me.
As they apply to copyrighted material, licenses are basically nothing more than mere tools in the form of a contract. Each of these tools has a particular purpose of disseminating information in a particular fashion. Licenses should be selected or drafted according to the desired means of dissemination. No single license fits all possible needs.
Correct, NDA's are anoher subject, not necessaryilly anything to do with copyright licenses. In fact they are all about things that may not be
copyrightable. For example, the mere fact that a company is developing some product may be something they don't want known to their competitors. NDA's can be more about trade secrets than anything else. Copyright covers the text of the code you are producing, NDA's cover the ideas behind the code.
Cripes Heater, we're talking about developer to developer interchange, not end user.
Are we? I thought this thread was about posting code in the forums. I make no distinction between end users and developers. Someone who uses can
be a developer, a developer can sometimes just use. Same as books and authors, authors read as well, readers can become authors.
The viral part is how that plays out down the developer chain, the knowledge is force into a tight control because the process is forced into repeating by it's very demands.
Generally these commercial developer chains are not very long and strictly limited by the licences (and NDS's). Hardly viral for the code itself.
Viral is normally a term applied to GPL'ed and other code because it almost forces itself to be published publicly.
Copyright covers the text of the code you are producing, NDA's cover the ideas behind the code.
I would think NDA is also used as a pretext to licensing. So, in context to this discussion, they're equivalent.
I make no distinction between end users and developers. Someone who uses can be a developer, ...
Which, for this conversation, makes all parties developers.
Generally these commercial developer chains are not very long and strictly limited by the licences (and NDS's). Hardly viral for the code itself.
And that matters? When a closed source license/NDA is written it very much intends to apply itself downstream. I'm sure there's plenty of shallow GPL'd code also.
Viral is normally a term applied to GPL'ed and other code because it almost forces itself to be published publicly.
When a closed source license/NDA is written it very much intends to apply itself downstream
Yes but that stream is very short. Perhaps only one licensee. The licence is designed to keep it short.
"Viral" has connotations of reproducing wildly and hopping from host to host freely. Hence the common use of the word to describe the wild fire spread of news, stories, jokes and such through the internet.
Commercial licences are exactly designed to stop that viral behavior of copy after copy hopping from host to host. Commercial licences stop source code spreading outside the company, except as allowed for subcontractors etc under NDA as we have discussed. Commercial licences stop the spread of binary executables except if the user wants to pay up. Copying your binary and selling it on is verboten. They would even like to stop selling on the only binary copy you have.
So, I don't see how "viral" describes commercial licenses. I do see how it applies to GPL, for example, that why GPL is described as viral. Nothing to do with marketing.
It's always a point of view isn't it. The marketers want to give a negative connotation to the competition so they chose to label the competition's product with a negative name while at the same time being careful not to associate it with their own product. Either process, as I've shown, can be seen from either side as negative.
The very nature of what makes the GPL a positive for some has been branded as negative, in the casual observers eyes, by the competition.
Comments
It's not about giving credit. It's about the specific license. I will never use GLP, LGPL, CC, or any other viral license. I do not wish to impose a license choice onto anyone else's code just because they choose to use my code.
I always give credit wherever possible, if I fail to give it, it's just a mistake that I will fix if notified.
As for copyrighting numbers, that argument has been used to suggest that is impossible to copyright any kind of code for a long time. CSS was just a famous example. I fail to see why a piece of software should become uncopyrightable because it is a prime number as opposed to any other number.
As for the GPL vs MIT vs whatever. It just depends what your aims are in releasing the code. Choose accordingly. GPL is great if your aim is to make a vast body of software permanently free and open for the benefit of the human race as a whole for generations to come. That's a big project that many are not willing or able to contribute to, or don't see the need for, or "oh my God it's communist". MIT is great if you just want to get functionality out into the world for the benefit of all and don't care if they use it as open source or not. Draconian closed licenses are OK if you want to try and make money.
I seriously doubt that Bill's intentions were anything but positive, although I do wish he'd reconsider his license. Being recognized for your coding hasn't ever been any issue here as far as I can tell. I've used a lot of MIT code in various projects and try to go out of my way to make sure everyone is mentioned somewhere both in the code itself and in any related materials.
We've drank a lot of coffee and kicked this to almost nausea. It's time to get back to doing great things with the Prop. (Either Prop)
Jeff
http://forums.parallax.com/showthread.php?144677-Announcing-CLMM-(pronounced-as-Clem)-Execute-Code-from-the-CLUT
C.W.
Hey, there is this!
http://sam.zoy.org/wtfpl/
http://wiki.creativecommons.org/FAQ#Can_I_use_a_Creative_Commons_license_for_software.3F
CC really is intended for data products more than code, which supports the basic incompatibility we've been discussing here.
LOL. That pretty much settles it. I wonder if going against the CC recommendation voids it's use
The concern is not what you, or anyone else with noble intent, would do but what someone less so inclined may do. It's a big game of "What if?'s" in the commercial world.
The piece of MIT code, in my example here, would not be a derivative. It would be like the code snippets one can pick up from collections. The Parallax forums, and OBEX, being examples of such collections.
I am all too familiar with the commercial software world, having been doing commercial software of over 20 years now. Commercial software developers generally don't assign a license to their code, instead they just keep it private. If they want to get help from an external source it almost always involves and NDA. Most commercial entities avoid GPL or similar licensed software, and even go so far as to have policies that forbid their employees from viewing any such code except in special cases.
There is a great deal of code I have written that I can never share, and it's kind of a bummer, but it's part of my job.
For me, most of my prop related code is part of my hobby. The few places where it's not part of my hobby happen to be open source situations (the C/C++ Spin/PASM compiler is an example). I choose to use MIT, and I also choose avoid working on things that involve GPL licensed code. For example, I only did do a small amount of contribution to the propgcc effort, mostly feedback and comments to the guys doing the actual work. I have only looked at a small portion of the code involved, mostly the loader and kernels. Since I changed jobs, I am more free to contribute in the future efforts for Prop2, and I hope I can help, but I'll still be focusing on the non-GPL stuff. Can you tell that I really loath GPL?
Roy
EDIT: That is unless the NDA has a reasonable expiry date and no further restrictions beyond that expiry. Lol, I've never read any I signed, which isn't many.
GPL and the likes is for commercial use. You guessed it, it has the opposite effect - to spread the knowledge commercially.
It seems like the attribution aspect is the sticky point for some. The MIT license states that the "above copyright notice ... shall be included in all copies or substantial portions." I interpret that to include the copyright line that most people put at the beginning of the file. That would cover attibrution in the open source case, as long as the author isn't different than the copyright holder. However, in a closed-source situation there is no requirement to disclose the copyright holder, which may be a problem for some people.
Most of the commercial code I have worked on had no NDAs written for it since we had no external people work with it. The cases I can recall where it did have an NDA was with another company (such as Intel, nVidia, or AMD) to get them to help us find issues with their drivers/hardware. NDAs are much more than a license because they typical specify specific people who are included, specific time periods, and often more.
Commercial licenses are anything but viral. You license the software from the vendor and that is pretty much the end ofthe chain, it cannot reproduce and spread from there. Also, what knowledge? You didn't get any source code with that commercial license. You only got an executable. Nothing to learn from that. There it is again. How is preventing the spread of knowledge viral? How is preventing the spread of copies of even executables viral? Sounds like the
total opposite to me.
Viral is normally a description of something reproducing and spreading from host to host. Commercial software licenses are exactly designed to stop that.
All I can think is that you actually meant vile, as "...vile purpose - to prevent knowledge spreading" That would make sense.
As for in-house only, that hardly counts as you are implying no knowledge transfer at all. The question on this becomes: what is the rule if you wanted to make that code public?
There is knowledge passed between those two commercial entities. The viral part is how that plays out down the developer chain, the knowledge is force into a tight control because the process is forced into repeating by it's very demands.
As I said in my previous post, the MIT license provides attribution in the source code if the source is open, and the copyright line is included. I interpret the wording of the MIT license to require the "Copyright (c) <year> <name>" line. This is clear in some versions of the MIT license, but not so clear in the way it is used with Parallax.
Imagine a musician gets a deal with a record company for 5 new albums and as part of the deal is forced not to listen to music for the term of the contract?
Or an author finds she can't find a publisher for her book because they all suspect she has been, well, reading books by other publishers.
I don't belive copyright prevents you from using ideas from within the works of others. If that were so there may only have been allowed one blues song, one jazz tune, one rock'n'roll song etc etc. There might only ever have been allowed one detective novel, or one cow boys and indians movie.
Of course copy large junks of code is off the table but reading it, no, I would not tolerate such a prohibition from an employer. It's nuts.
I'm curious. Why do you "loath" the GPL? I can understand selecting this license or that depending on your intentions for your work. But loath is a gut reaction not a logical appraisal.
Yup. As mentioned at least twice now
Copyright only protects copy. In music, only the melody in a song is required for copyright ... "Happy Birthday" anyone? The term "substantial" in the MIT license is fairly vague. Considering the number of ways available to achieve the same end, it beats me as to why copyright is really even valid for software unless it's completely a verbatim copy. Since RAP doesn't have a melody, it seems to be an escape for the music industry and thus grows like weeds. I'm no authority and do not want to be ... it's just common sense to me.
copyrightable. For example, the mere fact that a company is developing some product may be something they don't want known to their competitors. NDA's can be more about trade secrets than anything else. Copyright covers the text of the code you are producing, NDA's cover the ideas behind the code. Are we? I thought this thread was about posting code in the forums. I make no distinction between end users and developers. Someone who uses can
be a developer, a developer can sometimes just use. Same as books and authors, authors read as well, readers can become authors. Generally these commercial developer chains are not very long and strictly limited by the licences (and NDS's). Hardly viral for the code itself.
Viral is normally a term applied to GPL'ed and other code because it almost forces itself to be published publicly.
Which, for this conversation, makes all parties developers.
And that matters? When a closed source license/NDA is written it very much intends to apply itself downstream. I'm sure there's plenty of shallow GPL'd code also.
That's marketing for ya.
I'm must be dense I'm still not with you.
Yes but that stream is very short. Perhaps only one licensee. The licence is designed to keep it short.
"Viral" has connotations of reproducing wildly and hopping from host to host freely. Hence the common use of the word to describe the wild fire spread of news, stories, jokes and such through the internet.
Commercial licences are exactly designed to stop that viral behavior of copy after copy hopping from host to host. Commercial licences stop source code spreading outside the company, except as allowed for subcontractors etc under NDA as we have discussed. Commercial licences stop the spread of binary executables except if the user wants to pay up. Copying your binary and selling it on is verboten. They would even like to stop selling on the only binary copy you have.
So, I don't see how "viral" describes commercial licenses. I do see how it applies to GPL, for example, that why GPL is described as viral. Nothing to do with marketing.
It's always a point of view isn't it. The marketers want to give a negative connotation to the competition so they chose to label the competition's product with a negative name while at the same time being careful not to associate it with their own product. Either process, as I've shown, can be seen from either side as negative.
The very nature of what makes the GPL a positive for some has been branded as negative, in the casual observers eyes, by the competition.