Hello GNU folks,
this idea came to me when I was listening to the background music of a small game called "Trackball":
What abount a license which gurantees that also the music made for a software project - GNU GPL'ed or compatible, of course -is *free*?
I mean, we already a license for software (GNU GPL) and also for documentation (FDL) but no license for music (FML = Free Music License?). But you cannot compare music with software nor documentation. Music is different to them.
Just an idea...
Regards, Roland Haeder
On Sunday 14 August 2005 09:03, Roland Häder wrote:
What abount a license which gurantees that also the music made for a software project - GNU GPL'ed or compatible, of course -is *free*?
Since the music is released together with the source code, I imagine that it is covered by GPL, just like the source. IANAL.
There is already Creative Commons, a project of Lawrence Lessig which also focuses on music. Of course, it wouldn't be a bad idea to have a GNU licence for music, too.
At Sun, 14 Aug 2005 09:33:43 +0300, NSK wrote:
On Sunday 14 August 2005 09:03, Roland Häder wrote:
What abount a license which gurantees that also the music made for a software project - GNU GPL'ed or compatible, of course -is *free*?
Since the music is released together with the source code, I imagine that it is covered by GPL, just like the source. IANAL.
There is already Creative Commons, a project of Lawrence Lessig which also focuses on music. Of course, it wouldn't be a bad idea to have a GNU licence for music, too.
License proliferation is bad. Other than that, the GNU project hasn't anything to do with music. The FSF can't even write a decent documentation license, so a music license would probably be even worse than that.
So I think that it would be a bad idea.
Jeroen Dekkers
License proliferation is bad. Other than that, the GNU project hasn't anything to do with music.
The GNU project has a arts section, and collects various songs that are about free software an hacking. The Free Software song is one such example.
The FSF can't even write a decent documentation license, so a music license would probably be even worse than that.
The GNU FDL is a perfectly good license, and it would do you good not to belittle other peoples efforts.
For music, I'd say that a simple `verbatim copying' license would be enough.
On 14-Aug-2005, Alfred M. Szmidt wrote:
For music, I'd say that a simple `verbatim copying' license would be enough.
Verbatim only? That fails to grant freedom to modify and redistribute. Do you think such a license should be called "free"?
Why should music users get less freedom over a work than users of GPL programs?
For music, I'd say ithat a simple `verbatim copying' license would be enough.
Verbatim only? That fails to grant freedom to modify and redistribute. Do you think such a license should be called "free"?
Because music is not a functional work, and you don't need the right to `modify' it, you can have it, but it is not a nessecary right. Where as with software, it is.
And labeling everything with `free', when one speaks of the freedoms that govern free software, will only cloud the discussion, so please don't say `free' when you really mean `free software'.
The is the same problem that Debian faces, labeling everything as `free software', and then excluding everything that doesn't fall into that category (like free doucmnetation, free music, etc).
Why should music users get less freedom over a work than users of GPL programs?
Music and software are different, comparing the blindly will only lead to problems.
Cheers.
"Alfred M. Szmidt" ams@gnu.org wrote:
Verbatim only? That fails to grant freedom to modify and redistribute. Do you think such a license should be called "free"?
Because music is not a functional work, and you don't need the right to `modify' it, you can have it, but it is not a nessecary right. Where as with software, it is.
Who is Alfred M. Szmidt - someone who does not even see fit to attribute the previous author quoted - to decide that we don't "need" the right to modify music but do with programs?
[...]
The is the same problem that Debian faces, labeling everything as `free software', and then excluding everything that doesn't fall into that category (like free doucmnetation, free music, etc).
The problem debian faces is that software which is not free software sometimes gets included in the distribution. It's not a new problem. Even if terms such as "free documentation" can be defined in a sane way, debian doesn't currently use them.
Who is Alfred M. Szmidt - someone who does not even see fit to attribute the previous author quoted
You are quoted and attributed, in the header, see the Reference, Message-ID and To headers. A 5th attribute isn't needed in my opinion.
- to decide that we don't "need" the right to modify music but do with programs?
Who is MJ Ray who doesn't even know that he is being properly quoted and attributed, to decided that we actually need the right to modify music?
"Alfred M. Szmidt" ams@gnu.org wrote:
You are quoted and attributed, in the header, see the Reference, Message-ID and To headers. A 5th attribute isn't needed in my opinion.
I found no quote in any header and only one attribution, in In-Reply-To. Neither GNU Mailutils nor GNUMail show the In-Reply-To header by default, nor include it in the reply. Please make your mailer behave in a more common way that respects other authors.
Who is MJ Ray who doesn't even know that he is being properly quoted and attributed, to decided that we actually need the right to modify music?
I'm me and I have made no claim about what "we" need. It would seem that I'm not as arrogant as you! ;-)
At Wed, 17 Aug 2005 12:30:23 +0100, MJ Ray wrote:
"Alfred M. Szmidt" ams@gnu.org wrote:
The is the same problem that Debian faces, labeling everything as `free software', and then excluding everything that doesn't fall into that category (like free doucmnetation, free music, etc).
The problem debian faces is that software which is not free software sometimes gets included in the distribution. It's not a new problem. Even if terms such as "free documentation" can be defined in a sane way, debian doesn't currently use them.
I think that if you were to create a sane definition of free documentation, you would get very close to the definition of free software or the DFSG. And the GFDL wouldn't pass such a test either.
For me, one required freedom is to modify the whole document as I see fit. The GFDL currently fails to grant that freedom.
Jeroen Dekkers
For me, one required freedom is to modify the whole document as I see fit. The GFDL currently fails to grant that freedom.
The GNU GPL also fails to grant that freedom, if by `whole' you include the copyright notice(s). Invariant sections are not considered to be part of the main document; consider them as copyright notices that programs print when you pass the `--version' flag. The main document may be modified as you see fit.
At Wed, 17 Aug 2005 14:24:09 +0200, Alfred M. Szmidt wrote:
For me, one required freedom is to modify the whole document as I see fit. The GFDL currently fails to grant that freedom.
The GNU GPL also fails to grant that freedom, if by `whole' you include the copyright notice(s). Invariant sections are not considered to be part of the main document; consider them as copyright notices that programs print when you pass the `--version' flag. The main document may be modified as you see fit.
If they are not part of the main document, I should be able to simply remove them. But that's not the case, the invariant sections are a very strong part of the document. So strong that you can't even remove them. It's even more important than other sections of the documents, because those sections can just be removed. So that they aren't part of the document is just nonsense. It's like saying an engine isn't part of a car, because it's in front of the steering wheel.
And that you aren't allowed to remove copyright notices is something the law already says. A license can only reaffirm that. There of course might also be some other small limitations, liking marking a modified copy clearly as modified. But not being to modify or delete entire sections just isn't free. It's like saying you can edit a whole program, except two source files which do something different than the rest of the program.
There is just no reason to accept such a restriction.
Jeroen Dekkers
If they are not part of the main document, I should be able to simply remove them. But that's not the case, the invariant sections are a very strong part of the document. So strong that you can't even remove them. It's even more important than other sections of the documents, because those sections can just be removed.
Is it the non-removal bit that you object to now, or is it the verbatim bit? Or both? You keep jumping from one topic to another.
So that they aren't part of the document is just nonsense.
I didn't say that, I said the _main_ document.
It's like saying you can edit a whole program, except two source files which do something different than the rest of the program.
Now you changed the topic, once again. Either we are talking about source code, or documentation. We cannot speak about both, since they are different. There is no point in not being allowed to change source code, hence why you should have the right to change it.
There is a perfectly valid reason in not changing a non-functional work, like my memoirs, or your toughts about how much you love butterflies. You shouldn't be able to change what I or you thought about an issue.
There is also a perfectly valid reason in having invariant sections in a document, I might have written a document about how you twiddle frobs in a black box, and wish to give a note to everyone that my (non-existent) wife baked me cookies when I wrote the book and how much it helped me, and include a recipe for these cookies. I also don't want to have someone modify the recipe to produce something horrible so that everyone will think that my wife can't cook.
I also might want to make this note to my wife non-removable, since the book would not have been possible without her support.
For documentation, all this makes sense, for source code it does not.
One can agree that not being able to remove the invariant sections can be a bit troublesome if you have 1000 people who add such sections, each adding their own anecdote about how much they loved (or hated) my wife's cookies.
To me the GFDL has other problems, namely, it being a overall complex document, far to complex to be grasped in the same way that the GNU GPL can be grasped. If the GFDL has invariant sections, or not, is frankly a side issue which everyone seems to miss.
There is just no reason to accept such a restriction.
Just because you cannot see the need for such a `restriction' doesn't mean that there are good reasons to protect ones freedom to have it.
At Wed, 17 Aug 2005 15:56:53 +0200, Alfred M. Szmidt wrote:
If they are not part of the main document, I should be able to simply remove them. But that's not the case, the invariant sections are a very strong part of the document. So strong that you can't even remove them. It's even more important than other sections of the documents, because those sections can just be removed.
Is it the non-removal bit that you object to now, or is it the verbatim bit? Or both? You keep jumping from one topic to another.
I was talking about modification. Removing a section is modification. Changing text is also modification. So it's both well within the topic.
But if you were able to remove the invariant sections, the problem would go away for the biggest part: you can just remove all invariant sections and get a free document.
So that they aren't part of the document is just nonsense.
I didn't say that, I said the _main_ document.
So what is this main document? Do you normally start reading a book at page 125 or so? I normally start at page 1, and all the pages are considered part of the book by me. When I'm talking about a book, or a document, I'm talking about the whole thing, not about some arbitrary part of it.
It's like saying you can edit a whole program, except two source files which do something different than the rest of the program.
Now you changed the topic, once again. Either we are talking about source code, or documentation. We cannot speak about both, since they are different. There is no point in not being allowed to change source code, hence why you should have the right to change it.
I suggest you pick a dictionary and search for the word "analogy". Wikipedia might have some information about it too.
And why is there no point in not being allowed to modify source code and is there a point in being disallowed to modify documentation?
There is a perfectly valid reason in not changing a non-functional work, like my memoirs, or your toughts about how much you love butterflies. You shouldn't be able to change what I or you thought about an issue.
What is this valid reason?
There is also a perfectly valid reason in having invariant sections in a document, I might have written a document about how you twiddle frobs in a black box, and wish to give a note to everyone that my (non-existent) wife baked me cookies when I wrote the book and how much it helped me, and include a recipe for these cookies. I also don't want to have someone modify the recipe to produce something horrible so that everyone will think that my wife can't cook.
So how is your non-existant wife baking cookies more important than my freedom?
I also might want to make this note to my wife non-removable, since the book would not have been possible without her support.
For documentation, all this makes sense, for source code it does not.
Why?
One can agree that not being able to remove the invariant sections can be a bit troublesome if you have 1000 people who add such sections, each adding their own anecdote about how much they loved (or hated) my wife's cookies.
Yes. Certainly so when people are going to add religous and political text. So what are the reasons for having all these potential troublesome things?
To me the GFDL has other problems, namely, it being a overall complex document, far to complex to be grasped in the same way that the GNU GPL can be grasped. If the GFDL has invariant sections, or not, is frankly a side issue which everyone seems to miss.
No, because I think my freedom is more important than the complexity of a document. If a license is complex but grants everybody enough freedom, that is far smaller problem than if a license takes away necessary freedoms.
There is just no reason to accept such a restriction.
Just because you cannot see the need for such a `restriction' doesn't mean that there are good reasons to protect ones freedom to have it.
What are these good reasons? I still haven't heard this reason for taking away my freedom.
Jeroen Dekkers
On Wed, 2005-08-17 at 16:41 +0200, Jeroen Dekkers wrote:
At Wed, 17 Aug 2005 15:56:53 +0200, Alfred M. Szmidt wrote:
There is a perfectly valid reason in not changing a non-functional work, like my memoirs, or your toughts about how much you love butterflies. You shouldn't be able to change what I or you thought about an issue.
What is this valid reason?
Identity!
Jeroen you are taking this discussion to the extreme I would suggest you thinking some more before writing this way. I think you're just answering on impulsively and not thinking at all about the matter, just trying to defend your position. This is not dialog.
Would you accept me quoting you in a mail but changing your words to say "I love proprietary software, you bastards!" when instead you wrote: "I love free software my friends" ?
I don't think so, and this is a stupid example. Political thoughts are delicate and not being allowed to change MY thoughts cannot be considered a limitation of YUOR freedom, unless you consider freedom as being able to do whatever you like disregarding others freedom.
There is also a perfectly valid reason in having invariant sections in a document, I might have written a document about how you twiddle frobs in a black box, and wish to give a note to everyone that my (non-existent) wife baked me cookies when I wrote the book and how much it helped me, and include a recipe for these cookies. I also don't want to have someone modify the recipe to produce something horrible so that everyone will think that my wife can't cook.
So how is your non-existant wife baking cookies more important than my freedom?
It is a personal thought of the author, something intimate, something that have nothing to do with your freedom.
I also might want to make this note to my wife non-removable, since the book would not have been possible without her support.
For documentation, all this makes sense, for source code it does not.
Why?
This is a question I would do too, why ... Now before you can even think about the answer you should first ask yourself, what limit to my freedom do something like that mean in the case of a _documentation_ book ? Are they acceptable? If you accept the limit on freedom posed by the GNU GPL (freedom to take parts of the software without sticking to the license) as a good balance you should think if an invariant section may be a good balance in the case of a document, a book, an article, etc ...
One can agree that not being able to remove the invariant sections can be a bit troublesome if you have 1000 people who add such sections, each adding their own anecdote about how much they loved (or hated) my wife's cookies.
Yes. Certainly so when people are going to add religous and political text. So what are the reasons for having all these potential troublesome things?
Being Humans ? I think we can discuss the idea of letting people use the variant sections in other works, I do not see why you should be able to strip out all the invariant section from a book and still refer to the new work as it was the previous book. What kind of freedom to you get form that? Would you really do that? In which case? Why?
To me the GFDL has other problems, namely, it being a overall complex document, far to complex to be grasped in the same way that the GNU GPL can be grasped. If the GFDL has invariant sections, or not, is frankly a side issue which everyone seems to miss.
No, because I think my freedom is more important than the complexity of a document. If a license is complex but grants everybody enough freedom, that is far smaller problem than if a license takes away necessary freedoms.
Can you express which freedom of yours is being getting out by the GFDL? What about the author freedom? Where's the balance? Why people that whines against the GFDL has not yet been able to: A) explain clearly which is the problem in a public article so that all people can understand your illumination? B) Propose a better license for documentation that address the same problem that the GFDL try to solve?
No, saying: "the GNU GPL is the right one because it grants my freedom" is not enough, sorry.
There is just no reason to accept such a restriction.
Just because you cannot see the need for such a `restriction' doesn't mean that there are good reasons to protect ones freedom to have it.
What are these good reasons? I still haven't heard this reason for taking away my freedom.
You should first clearly express which freedom is taken away from you, which freedom is taken away from the author, and why our freedom is more important then the freedom of expression of the author.
Please try to honestly answer to my question, if you can effectively explain me these things, I'm willing to think about them and perhaps change my mind. Please do not just rant for the sake of it, it makes no sense and just weaken your position.
Simo.
At Thu, 18 Aug 2005 10:57:39 +0200, Simo Sorce wrote:
On Wed, 2005-08-17 at 16:41 +0200, Jeroen Dekkers wrote:
At Wed, 17 Aug 2005 15:56:53 +0200, Alfred M. Szmidt wrote:
There is a perfectly valid reason in not changing a non-functional work, like my memoirs, or your toughts about how much you love butterflies. You shouldn't be able to change what I or you thought about an issue.
What is this valid reason?
Identity!
Would you accept me quoting you in a mail but changing your words to say "I love proprietary software, you bastards!" when instead you wrote: "I love free software my friends" ?
No, but there is no reason to forbid this in a license. It's already forbidden by defamation laws. Other than that, the problem might also be solved in other ways.
I don't think so, and this is a stupid example. Political thoughts are delicate and not being allowed to change MY thoughts cannot be considered a limitation of YUOR freedom, unless you consider freedom as being able to do whatever you like disregarding others freedom.
It can be solved very easy: just make a requirement that modified versions must be marked as such.
There is also a perfectly valid reason in having invariant sections in a document, I might have written a document about how you twiddle frobs in a black box, and wish to give a note to everyone that my (non-existent) wife baked me cookies when I wrote the book and how much it helped me, and include a recipe for these cookies. I also don't want to have someone modify the recipe to produce something horrible so that everyone will think that my wife can't cook.
So how is your non-existant wife baking cookies more important than my freedom?
It is a personal thought of the author, something intimate, something that have nothing to do with your freedom.
If this thought is forced upon me and the rest of the readers and co-writers, then it is.
I also might want to make this note to my wife non-removable, since the book would not have been possible without her support.
For documentation, all this makes sense, for source code it does not.
Why?
This is a question I would do too, why ... Now before you can even think about the answer you should first ask yourself, what limit to my freedom do something like that mean in the case of a _documentation_ book ? Are they acceptable? If you accept the limit on freedom posed by the GNU GPL (freedom to take parts of the software without sticking to the license) as a good balance you should think if an invariant section may be a good balance in the case of a document, a book, an article, etc ...
What do I get back? With the GPL it's pretty easy to see: more Free Software. With the invariant sections, it's not really easy to see for me. I know that g++ probably wouldn't have been there if the GPL didn't require to free it. But would the GCC manual exist without invariant sections? I think it would. And maybe my imagination isn't good enough, but I really can't imagine that we would get less free documentation if we wouldn't allow invariant sections to be removed at least (which might be an acceptable compromise - although I'm not sure about it).
One can agree that not being able to remove the invariant sections can be a bit troublesome if you have 1000 people who add such sections, each adding their own anecdote about how much they loved (or hated) my wife's cookies.
Yes. Certainly so when people are going to add religous and political text. So what are the reasons for having all these potential troublesome things?
Being Humans ? I think we can discuss the idea of letting people use the variant sections in other works, I do not see why you should be able to strip out all the invariant section from a book and still refer to the new work as it was the previous book. What kind of freedom to you get form that? Would you really do that? In which case? Why?
Nobody says we should be referring to the new book as the previous book. But it's actually the GFDL which requires that! If I take the GCC manual, modify it however I want, and apply the front-cover clauses, I *must* put "a GNU manual" on the cover. So I'm even forced to do it.
To me the GFDL has other problems, namely, it being a overall complex document, far to complex to be grasped in the same way that the GNU GPL can be grasped. If the GFDL has invariant sections, or not, is frankly a side issue which everyone seems to miss.
No, because I think my freedom is more important than the complexity of a document. If a license is complex but grants everybody enough freedom, that is far smaller problem than if a license takes away necessary freedoms.
Can you express which freedom of yours is being getting out by the GFDL? What about the author freedom? Where's the balance? Why people that whines against the GFDL has not yet been able to: A) explain clearly which is the problem in a public article so that all people can understand your illumination?
It's pretty strange to hear on this list that caring about your freedom is whining. But anyway, your statement that the people not been able to do this is just false. Some Debian Developers wrote a pretty clear article about the GFDL, to give an example: http://people.debian.org/~srivasta/Position_Statement.html
To quote one part of it:
In addition to the simple restrictions of freedoms imposed by the Invariant Sections, they also cause practical problems:
* Incompatibility with the GPL. It's GPL-incompatible in both directions. This means that you can't legally extract text from a GFDL'ed manual and put it into integrated help strings in a GPL'ed program. And you can't extract code or comments from a GPL'ed program and put it into a GFDL'ed manual. (Without getting explicit permission to relicense from every copyright-holding contributor, that is.)
* Being forced to retain inaccurate Invariant Sections (or Cover Texts, or Dedications).
* Being forced to retain obsolete Invariant Sections (or Cover Texts, or Dedications). Dated invariant sections would exacerbate this problem.
* Possibility of obsolescence, via changes in technologies (such as the disappearance of printed matter, or the particulars of file formats and access restrictions).
* Being forced to retain technically inappropriate Invariant Sections or Cover Texts, etc.
* Being forced to retain Invariant Sections even in extremely space-tight environments (such as a rescue disks, reference card, PDA's, or embedded devices).
* Being forced to retain untranslated Invariant Sections in a translation.
* Being unable to use material from the document for a new document whose primary topic is that of an Invariant Section (because the Invariant Section must be retained, and must be Secondary, but would no longer be Secondary). This issue, where it can be very difficult or impossible to repurpose chunks (eg copy a chapter about regular expressions when one copies the just the regexp code), is a significant degradation of freedom.
* Invariant Section "bloat". The natural response to several of the above problems is to add new Invariant Sections, saying "I think the old Invariant Section is inaccurate/obsolete/offensive" or "This is a translation of the old Invariant Section". These will accumulate and will also be non-removable.
* Difficulty in modifying invariant sections of GFDLed documents not under unified central control (need permission from many contributors or their estates/agents, which becomes more difficult with time).
B) Propose a better license for documentation that address the same problem that the GFDL try to solve?
I've actually been thinking about it. The main problem is lack of time and it wouldn't solve the current problems with all current GFDL works. I also heard that the FSF might fix the GFDL.
No, saying: "the GNU GPL is the right one because it grants my freedom" is not enough, sorry.
Can you please say why we have this really big need for invariant sections and why this need is bigger than all the problems just mentioned? Because I haven't heard it and I asked for it in my previous mail.
There is just no reason to accept such a restriction.
Just because you cannot see the need for such a `restriction' doesn't mean that there are good reasons to protect ones freedom to have it.
What are these good reasons? I still haven't heard this reason for taking away my freedom.
You should first clearly express which freedom is taken away from you, which freedom is taken away from the author, and why our freedom is more important then the freedom of expression of the author.
I don't see any freedom of expression taken away from the author. The author still has the freedom to write whatever he wants. I'm not forbidding the author to express himself. I'm just saying that I don't see a need to give the author of a manual the power to force his expressions on all future co-authors and readers of that manual.
Please try to honestly answer to my question, if you can effectively explain me these things, I'm willing to think about them and perhaps change my mind. Please do not just rant for the sake of it, it makes no sense and just weaken your position.
My freedom to do with the manual what I want. It's the same thing as with free software. There you also need to freedom to do with the things I want.
I can give an example. I have a manual of a program. Now I want to make a reference card based on the manual (so it's a derived work and also falls under the GFDL). But this reference card has to include all invariants sections, as a whole, because if it doesn't, I'm violating the GFDL.
Or what if the author of this manual writes that Bush is great and talks about that everybody should support the Republician Party. Then my freedom is restricted, because I can't remove it.
Now a question to you. If I write a political text supporting Bush and print if everytime my programs starts up and write a license which says that you can't remove that text. Would it be Free Software? And if wouldn't, why would the same license be Free Documentation? Is a documentation writer more important than a software writer? It's just some questions I have asked myself. Personally I've come to the conclusion that documentation might be treated a bit different because it is a bit different than software, but that the basic freedoms shouldn't be that much different.
Jeroen Dekkers
On Thu, 2005-08-18 at 22:12 +0200, Jeroen Dekkers wrote:
At Thu, 18 Aug 2005 10:57:39 +0200, Simo Sorce wrote:
This is a question I would do too, why ... Now before you can even think about the answer you should first ask yourself, what limit to my freedom do something like that mean in the case of a _documentation_ book ? Are they acceptable? If you accept the limit on freedom posed by the GNU GPL (freedom to take parts of the software without sticking to the license) as a good balance you should think if an invariant section may be a good balance in the case of a document, a book, an article, etc ...
What do I get back? With the GPL it's pretty easy to see: more Free Software. With the invariant sections, it's not really easy to see for me. I know that g++ probably wouldn't have been there if the GPL didn't require to free it. But would the GCC manual exist without invariant sections? I think it would. And maybe my imagination isn't good enough, but I really can't imagine that we would get less free documentation if we wouldn't allow invariant sections to be removed at least (which might be an acceptable compromise - although I'm not sure about it).
Ok, I buy out the argument that the objective of both licenses is to foster more production of works, (I'm not sure this is the main objective of the GFDL, but let's take it as an assumption). Given that is documentation and software comparable in this regard ? What's the culprit in the GPL? To protect freedom of the software against proprietarization (protection) and to spread use of the GPL license through usage of other GPL licensed code in new projects (inheritance). Is it the same for documentation? I think the protection wise the GFDL is the same as the GPL. What about inheritance? Well I do not think inheritance is so important for documentation as it is for software. This different weight in the importance of this factor, make it possible to reconsider the which incentives are more important. An invariant section, might be a better incentive for some authors to write documentation then the chance of reuse of the writing. I do not know if this is a plausible thing, but we should think some more on it. I'm not saying you're wrong, I'm just saying that as free software programmers we may have a distorted way to look at things.
Nobody says we should be referring to the new book as the previous book. But it's actually the GFDL which requires that! If I take the GCC manual, modify it however I want, and apply the front-cover clauses, I *must* put "a GNU manual" on the cover. So I'm even forced to do it.
I recognize this is a big problem. One of the worst problem of the GFDL is the inability to break down a work into pieces, like printing a chapter of the book each month on a paper etc ...
Can you express which freedom of yours is being getting out by the GFDL? What about the author freedom? Where's the balance? Why people that whines against the GFDL has not yet been able to: A) explain clearly which is the problem in a public article so that all people can understand your illumination?
It's pretty strange to hear on this list that caring about your freedom is whining. But anyway, your statement that the people not been able to do this is just false. Some Debian Developers wrote a pretty clear article about the GFDL, to give an example: http://people.debian.org/~srivasta/Position_Statement.html
Thanks for point me at this draft, it is indeed clear, and I was delighted to finally see a clear concise text about the perceived problems with the GFDL.
To quote one part of it:
In addition to the simple restrictions of freedoms imposed by the Invariant Sections, they also cause practical problems:
- Incompatibility with the GPL. It's GPL-incompatible in both directions. This means that you can't legally extract text from a GFDL'ed manual and put it into integrated help strings in a GPL'ed program. And you can't extract code or comments from a GPL'ed program and put it into a GFDL'ed manual. (Without getting explicit permission to relicense from every copyright-holding contributor, that is.)
I do not think this is a problem, it is the same problem you have between many other free software licenses. Yes it is unfortunate that they both come from the FSF but are incompatible, but it is just a matter of taste. Plus I think it is wrong to say you need the permission of every copyright-holder to extract a portion of a work, you need permission only from the contributor(s) that make that portion of the work imho, but I do not want to get in to legal as IANAL.
Being forced to retain inaccurate Invariant Sections (or Cover Texts, or Dedications).
Being forced to retain obsolete Invariant Sections (or Cover Texts, or Dedications). Dated invariant sections would exacerbate this problem.
These are annoying.
Possibility of obsolescence, via changes in technologies (such as the disappearance of printed matter, or the particulars of file formats and access restrictions).
Being forced to retain technically inappropriate Invariant Sections or Cover Texts, etc.
This have to do more with the quality and proper use of the license, but I agree these are annoyances.
- Being forced to retain Invariant Sections even in extremely space-tight environments (such as a rescue disks, reference card, PDA's, or embedded devices).
This may be a real problem even if rare.
- Being forced to retain untranslated Invariant Sections in a translation.
This is annoying.
- Being unable to use material from the document for a new document whose primary topic is that of an Invariant Section (because the Invariant Section must be retained, and must be Secondary, but would no longer be Secondary). This issue, where it can be very difficult or impossible to repurpose chunks (eg copy a chapter about regular expressions when one copies the just the regexp code), is a significant degradation of freedom.
I do not understand this point very well.
- Invariant Section "bloat". The natural response to several of the above problems is to add new Invariant Sections, saying "I think the old Invariant Section is inaccurate/obsolete/offensive" or "This is a translation of the old Invariant Section". These will accumulate and will also be non-removable.
I think this is fiction, the natural response to a problem in an invariant section is to add a NON-invariant section that can be updated to say how much obsolete/wrong/whatever the original invariant section is. While it is possible that people may add other invariant sections I thin it is unlikely and a clear desire of the authors to trash their own work (which is also unlikely).
- Difficulty in modifying invariant sections of GFDLed documents not under unified central control (need permission from many contributors or their estates/agents, which becomes more difficult with time).
Potentially a problem.
B) Propose a better license for documentation that address the same problem that the GFDL try to solve?
I've actually been thinking about it. The main problem is lack of time and it wouldn't solve the current problems with all current GFDL works. I also heard that the FSF might fix the GFDL.
Yeah, it's a tough matter, now I recognize I was being provocative with this question, I apologize.
No, saying: "the GNU GPL is the right one because it grants my freedom" is not enough, sorry.
Can you please say why we have this really big need for invariant sections and why this need is bigger than all the problems just mentioned? Because I haven't heard it and I asked for it in my previous mail.
Sorry, I'm not advocating invariant sections, what I say is that we should carefully think if they are really so bad, if the balance in the case of documentation can be different and if invariant sections may be a good thing or not.
Based on this discussion I'm more and more convinced that being able to strip out invariant sections (but being forced to change the name of the resulting document and refer to the original one in a note) would be a good thing, also being able to use excerpts of the main text in GPLed works would be desirable (but not a must in my eyes).
I don't see any freedom of expression taken away from the author. The author still has the freedom to write whatever he wants. I'm not forbidding the author to express himself. I'm just saying that I don't see a need to give the author of a manual the power to force his expressions on all future co-authors and readers of that manual.
I tend to agree with you, but I can also see why an author may want to protect his ideas also on a written text. And as documentation does not have the same properties as software and we refer to secondary sections only I still can't decide whether invariant section are bad per se or just bad in their long term consequences, I tend to favor the second opinion right now, but more comments are welcome.
Please try to honestly answer to my question, if you can effectively explain me these things, I'm willing to think about them and perhaps change my mind. Please do not just rant for the sake of it, it makes no sense and just weaken your position.
My freedom to do with the manual what I want. It's the same thing as with free software. There you also need to freedom to do with the things I want.
I think that freedom is not constrained more then the way they are constrained with the GNU GPL.
I can give an example. I have a manual of a program. Now I want to make a reference card based on the manual (so it's a derived work and also falls under the GFDL). But this reference card has to include all invariants sections, as a whole, because if it doesn't, I'm violating the GFDL.
Yeah, I think this is one of the main problems of the GFDL, but a reference card is easy to rewrite compared to software, so the problem is much less serious than with software imho.
Or what if the author of this manual writes that Bush is great and talks about that everybody should support the Republician Party. Then my freedom is restricted, because I can't remove it.
Yes this is an example of political views expressed by an author, while I may disagree with his view I'm not sure I should be entitle to remove them, I can always add a comment about the fact I do not agree with them (and I'm not obliged to put my comment in an invariant section).
Now a question to you. If I write a political text supporting Bush and print if everytime my programs starts up and write a license which says that you can't remove that text. Would it be Free Software?
This is a hard question. If the original program prints out Copyrights notes by default, you can't change this thing in a GNU GPL program, but it is considered a Free Software License. As this text is a non functional part of the program I should think very carefully about not considering it a Free Software License, even if on a first look I may say it is not.
And if wouldn't, why would the same license be Free Documentation?
Well I tend to think that different works have different characteristics and freedom can not be claimed blindly.
Is a documentation writer more important than a software writer?
I do not think it is a question about the importance of the author, it is a question about the balance between the freedom of the user and the rights of the author. With software the functional aspects of a program are so important that the freedom to change it has more weight than the right of an author to veto any change, that's different for a document and it's non-functional aspects (like a dedication) imho.
It's just some questions I have asked myself. Personally I've come to the conclusion that documentation might be treated a bit different because it is a bit different than software, but that the basic freedoms shouldn't be that much different.
I fully agree, the difficult thing here is to balance freedoms, as the balance change in different kind of works imho.
Simo.
On Wed, Aug 17, 2005 at 03:56:53PM +0200, Alfred M. Szmidt wrote as answer to Jeroen Dekkers:
If they are not part of the main document, I should be able to simply remove them. But that's not the case, the invariant sections are a very strong part of the document.
There is also a perfectly valid reason in having invariant sections in a document, I might have written a document about how you twiddle frobs in a black box, and wish to give a note to everyone that my (non-existent) wife baked me cookies when I wrote the book and how much it helped me, and include a recipe for these cookies. I also don't want to have someone modify the recipe to produce something horrible so that everyone will think that my wife can't cook.
I also might want to make this note to my wife non-removable, since the book would not have been possible without her support.
One detail is often forgotten, but might is interesting in the debate: The GNU FDL only allows sections to be invariant only for "secondary sections".
So let the book's main topic be about "twiddle frobs in a black box" and let it be an unrelated section about Alfred's wife and her cookies. It needs to be unrelated, otherwise it would not be a secondary section. Also it probably needs to be short compared to the rest of the book otherwise the main topic would be the wife and the cookies.
Now maximising freedom would mean: Many users get a good book about the frobs in the black box. They might not care about the secondary section, but it also does not hurt. The question is more: Are the frobs described well?
If I want to update the book and reprint, I can change all I like about the frobs, the box and the twiddling. I will only do that if the first book suited me and I found the quality to be okay. Otherwise I would just write a new book with the knowledge I have learned from the old one.
I guess we should spend more time on rating the quality and selection of a free documentation under GNU FDL.
There is another small legal advantage of the GNU FDL: The opportunity to have invariant section helps a lot when you are in court explaning how the main part of your document was meant. Otherwise an author could not grant unlimited use for all purposes, he or she would just claim: This was not the indented use according to my personal rights!
At Thu, 18 Aug 2005 16:38:51 +0200, Bernhard Reiter wrote:
On Wed, Aug 17, 2005 at 03:56:53PM +0200, Alfred M. Szmidt wrote as answer to Jeroen Dekkers:
If they are not part of the main document, I should be able to simply remove them. But that's not the case, the invariant sections are a very strong part of the document.
There is also a perfectly valid reason in having invariant sections in a document, I might have written a document about how you twiddle frobs in a black box, and wish to give a note to everyone that my (non-existent) wife baked me cookies when I wrote the book and how much it helped me, and include a recipe for these cookies. I also don't want to have someone modify the recipe to produce something horrible so that everyone will think that my wife can't cook.
I also might want to make this note to my wife non-removable, since the book would not have been possible without her support.
One detail is often forgotten, but might is interesting in the debate: The GNU FDL only allows sections to be invariant only for "secondary sections".
So let the book's main topic be about "twiddle frobs in a black box" and let it be an unrelated section about Alfred's wife and her cookies. It needs to be unrelated, otherwise it would not be a secondary section. Also it probably needs to be short compared to the rest of the book otherwise the main topic would be the wife and the cookies.
What if I want to distribute only one chapter of the book? If there are 3 invariant sections, then it might well be that the invariant sections are three quarters of the whole. Is it still considered secondary? Or may not just distribute one chapter only?
Now maximising freedom would mean: Many users get a good book about the frobs in the black box. They might not care about the secondary section, but it also does not hurt. The question is more: Are the frobs described well?
I can tell the same about software: Many users get windows with their PC. Since they aren't programmers, it doesn't hurt that they can't modify it. The question is: Does windows do what they want?
If I want to update the book and reprint, I can change all I like about the frobs, the box and the twiddling. I will only do that if the first book suited me and I found the quality to be okay. Otherwise I would just write a new book with the knowledge I have learned from the old one.
Yes, and you add your own invariant section. And the persons which gets your new book also changes stuff and add their own little invariant section. Then somebody translates your book in another language, and adds their own invariant section in their language to the invariant section in your language (because they may not translate invariant section). And then we're about 20 years later, but we have still have the first invariant section which has actually no relevance anymore and even isn't in the same language the main book is.
There is another small legal advantage of the GNU FDL: The opportunity to have invariant section helps a lot when you are in court explaning how the main part of your document was meant. Otherwise an author could not grant unlimited use for all purposes, he or she would just claim: This was not the indented use according to my personal rights!
Huh? I don't follow this. A section not having anything to do with the main part describes how the main part was meant?!?
Jeroen Dekkers
"Alfred M. Szmidt" ams@gnu.org wrote:
There is a perfectly valid reason in not changing a non-functional work, like my memoirs, or your toughts about how much you love butterflies. You shouldn't be able to change what I or you thought about an issue.
Changing what you wrote does not change what you thought. Changing what you wrote does not claim you thought it. Attributing something you didn't write to you might.
The real problem is misattribution and invariant sections are a very blunt tool for attacking that. If I want to misattribute something to you, I don't need to use your expressions. I can just plain make stuff up (as people often do to me) and attribute it to you and your copyright licence won't matter.
Using a copyright licence for this only hurts friendly people who like your work and want to re-use it. Bad people who want to make you look bad will still make you look bad.
[...]
I also might want to make this note to my wife non-removable, since the book would not have been possible without her support.
For documentation, all this makes sense, for source code it does not.
Unmodifiable unremovable advertising, even for your wife and cookies, does not make any more sense for documentation. Why not have a program that pops up a window with the recipe and thanks at certain times? Those things are called "nag boxes" for a reason.
[...]
There is just no reason to accept such a restriction.
Just because you cannot see the need for such a `restriction' doesn't mean that there are good reasons to protect ones freedom to have it.
You may have reasons, but they're probably not going to persuade some that they're worthwhile when we make a different argument for software in general. "good" is the pivot there.
Regards,
The real problem is misattribution and invariant sections are a very blunt tool for attacking that. If I want to misattribute something to you, I don't need to use your expressions. I can just plain make stuff up (as people often do to me) and attribute it to you and your copyright licence won't matter.
Then those people are liable for defamnation. They are not if I change what you wrote and gave explicit permission to do so.
Using a copyright licence for this only hurts friendly people who like your work and want to re-use it. Bad people who want to make you look bad will still make you look bad.
The GFDL doesn't hurt anyone who is friendly, only those who are unfriendly.
Unmodifiable unremovable advertising, even for your wife and cookies, does not make any more sense for documentation. Why not have a program that pops up a window with the recipe and thanks at certain times? Those things are called "nag boxes" for a reason.
You obvioulsy don't get the difference between non-functional and functional works, please try and figure out why they are different before continuting with these comparisons.
At Wed, 24 Aug 2005 19:26:20 +0200, Alfred M. Szmidt wrote:
The real problem is misattribution and invariant sections are a very blunt tool for attacking that. If I want to misattribute something to you, I don't need to use your expressions. I can just plain make stuff up (as people often do to me) and attribute it to you and your copyright licence won't matter.
Then those people are liable for defamnation. They are not if I change what you wrote and gave explicit permission to do so.
Can you give me the clause in the law that says that people are only liable for defamation if they also violate copyright law?
Unmodifiable unremovable advertising, even for your wife and cookies, does not make any more sense for documentation. Why not have a program that pops up a window with the recipe and thanks at certain times? Those things are called "nag boxes" for a reason.
You obvioulsy don't get the difference between non-functional and functional works, please try and figure out why they are different before continuting with these comparisons.
That's because you are just making arbitrary differences.
Jeroen Dekkers
That's because you are just making arbitrary differences.
And you are making arbitrary indifferences.
"Alfred M. Szmidt" ams@gnu.org wrote:
The real problem is misattribution and invariant sections are a very blunt tool for attacking that. If I want to misattribute something to you, I don't need to use your expressions. I can just plain make stuff up (as people often do to me) and attribute it to you and your copyright licence won't matter.
Then those people are liable for defamnation. They are not if I change what you wrote and gave explicit permission to do so.
If and only if you gave permission to attribute incorrect statements to you, surely? In general, copyright and reputation are two mostly independent things and rightly so.
Using a copyright licence for this only hurts friendly people who like your work and want to re-use it. Bad people who want to make you look bad will still make you look bad.
The GFDL doesn't hurt anyone who is friendly, only those who are unfriendly.
For example, I consider having to print the adverts ("invariant sections") with a subset of the work painful, to the needlessly wasted trees as well as me. Generally too much pain, like using BSD with the obnoxious ad clause code, but worse.
[...]
You obvioulsy don't get the difference between non-functional and functional works, please try and figure out why they are different before continuting with these comparisons.
You obviously don't get that your distinction is entirely arbitrary.
Then those people are liable for defamnation. They are not if I change what you wrote and gave explicit permission to do so.
If and only if you gave permission to attribute incorrect statements to you, surely?
Yes, and the GPL does this, and so does the GFDL in non-invariant sections. Hence the need for verbatim-only sections.
In general, copyright and reputation are two mostly independent things and rightly so.
Sure, one is law, the other one isn't.
For example, I consider having to print the adverts ("invariant sections") with a subset of the work painful, to the needlessly wasted trees as well as me. Generally too much pain, like using BSD with the obnoxious ad clause code, but worse.
Your comparison doesn't make sense, the old BSD license is for software, the GFDL is for books, manuals, etc.
You obvioulsy don't get the difference between non-functional and functional works, please try and figure out why they are different before continuting with these comparisons.
You obviously don't get that your distinction is entirely arbitrary.
The same can be said for you who consider them the same.
Alfred M. Szmidt wrote:
In general, copyright and reputation are two mostly independent things and rightly so.
Sure, one is law, the other one isn't.
Surely both are covered by laws; reputation is covered by slander/libel laws, not copyright law. Trying to protect reputation through creative use of copyright law has malign side-effects in limiting freedom, and is in any case not very effective.
Cheers Graham
Surely both are covered by laws; reputation is covered by slander/libel laws, not copyright law.
Reputation and slander/libel are different things. Slander is when one tries to defame someone on purpose. Reputation is when you are `famous' for something.
On Wed, Aug 17, 2005 at 12:30:23PM +0100, MJ Ray wrote:
"Alfred M. Szmidt" ams@gnu.org wrote:
Verbatim only? That fails to grant freedom to modify and redistribute. Do you think such a license should be called "free"?
Because music is not a functional work, and you don't need the right to `modify' it, you can have it, but it is not a nessecary right. Where as with software, it is.
Who is Alfred M. Szmidt - someone who does not even see fit to attribute the previous author quoted - to decide that we don't "need" the right to modify music but do with programs?
MJ Ray,
I fail to see the usefulness or connection to the debate of this comment of yours.
Bernhard
Bernhard Reiter bernhard@intevation.de wrote:
On Wed, Aug 17, 2005 at 12:30:23PM +0100, MJ Ray wrote:
"Alfred M. Szmidt" ams@gnu.org wrote:
Because music is not a functional work, and you don't need the right to `modify' it, you can have it, but it is not a nessecary right. Where as with software, it is.
=20 Who is Alfred M. Szmidt - someone who does not even see fit to attribute the previous author quoted - to decide that we don't "need" the right to modify music but do with programs?
MJ Ray,
I fail to see the usefulness or connection to the debate of this comment of yours.
I don't see the usefulness of telling others what they need. The irritation about non-attribution was just in passing.
Please send any further personal messages for me off-list.
Thanks,
Ben Finney wrote:
On 14-Aug-2005, Alfred M. Szmidt wrote:
For music, I'd say that a simple `verbatim copying' license would be enough.
Verbatim only? That fails to grant freedom to modify and redistribute. Do you think such a license should be called "free"?
Hello :)
I don't know if it can be usefull to the discussion but I believe CC-by-sa (attribution- sharealike) is fine for artists who want a "GPL-like" licence for arts.
An example of what is possible thanks to the "GPL-like freedoms" for music: http://francois.schnell.free.fr/wordpress/?p=9
An example of a website with CC-by-sa musics and derivatives works: http://www.opsound.org/
francois
Why should music users get less freedom over a work than users of GPL programs?
Discussion mailing list Discussion@fsfeurope.org https://mail.fsfeurope.org/mailman/listinfo/discussion
On 14-Aug-2005, Jeroen Dekkers wrote:
License proliferation is bad.
When existing licenses do the job just as well, I agree.
Other than that, the GNU project hasn't anything to do with music.
Heh. I agree -- I've heard the Free Software Song too :-)
The FSF can't even write a decent documentation license
The GPL is fine for documentation, for my money.
Hi Jeroen,
On Sun, Aug 14, 2005 at 12:24:51PM +0200, Jeroen Dekkers wrote:
License proliferation is bad. Other than that, the GNU project hasn't anything to do with music.
true, though Stallman has made some statements about copyright and music in the past which I believe were substantial contributions to the debate.
The FSF can't even write a decent documentation license, so a music license would probably be even worse than that.
I disagree with you, the GNU FDL is a very decent documentation license. It touches some points that others would not have thought about before.
However it is not the last word on what freedom a documentation needs and how to archive this. Many have expected more, but for the focus of transporting technical instructions in books, the GNU FDL is among the best license I know, because clearly aimed at more freedom for society.
Bernhard
* Bernhard Reiter bernhard@intevation.de [050815 12:15]:
I disagree with you, the GNU FDL is a very decent documentation license. It touches some points that others would not have thought about before.
However it is not the last word on what freedom a documentation needs and how to archive this. Many have expected more, but for the focus of transporting technical instructions in books, the GNU FDL is among the best license I know, because clearly aimed at more freedom for society.
I think that's the main problem with the GNU FDL, it is a very nice free book license. It just is no free software license, an thus in my eyes totally unsuitable for documentation accompanying programs.
Hochachtungsvoll, Bernhard R. Link
On Tue, Aug 16, 2005 at 10:22:18AM +0200, Bernhard R. Link wrote:
- Bernhard Reiter bernhard@intevation.de [050815 12:15]:
the GNU FDL is a very decent documentation license. It touches some points that others would not have thought about before.
However it is not the last word on what freedom a documentation needs and how to archive this. Many have expected more, but for the focus of transporting technical instructions in books, the GNU FDL is among the best license I know, because clearly aimed at more freedom for society.
I think that's the main problem with the GNU FDL, it is a very nice free book license. It just is no free software license, an thus in my eyes totally unsuitable for documentation accompanying programs.
It was meant to not be a Free Software license, because it is for free documentation.
If you are documenting a program, it depends what the purpose of the documentation is to find out if the GNU FDL fits it.
API definition and source code comments, especially is used to create documentation automatically would better be considered source code and covered by the GNU GPL.
If you make an introductary course book or or an in-depth guide to usage, the GNU FDL can be best, because this basically is a free book then.
Bernhard
* Bernhard Reiter bernhard@intevation.de [050817 13:10]:
API definition and source code comments, especially is used to create documentation automatically would better be considered source code and covered by the GNU GPL.
If you make an introductary course book or or an in-depth guide to usage, the GNU FDL can be best, because this basically is a free book then.
But if I have only a free book, all the other things I can make with it are lost. There are not only books and API definitions, there are tutorials, how-tos, references and manpages. There might be interactive help screens or tooltips, literate programming, cups or mouse pads showing the most important commands.
Each of these adds another access to the program or system being documented, none of them and not even two of them together can alone be sattisfactory.
And while they have all obvious and subtle differences, and may be best when written from scratch, this is just impossible. By sheer lack of time I can neigther rewrite my Operating System every time I it for some other target, nor can I write enough documentation without heavily using everything else documentating this program including the source code of it.
Thus the information and its copyrightable containers in forms of sentences and paragraphs must be able to flow between all these different forms and ideally the source code of the machine interpreted code of the program, too.
Adding unnecessary limits to this flow between different aspects (What is a title page of a manpage? What that of a cup? Where do I "include a unaltered copy" of the 3278 words license on a mouse pad? Does every manpage count as "document" and has to contain the license and all those possible invariant sections?) or between the documentation and the program (Are the texts of the tooltips only mere aggregation though all of this logik when to show which? Or do I have to write them in a clear-room implementation on my own?) keeps us away from getting free programs with free documentation.
So please, whenever you have or feel to release something under GFDL, please consider dual-license it adding the permissions of the GNU GPL or any other free software compatible with the program documented.
Hochachtungsvoll, Bernhard R. Link
P.S: Nice visiting when you want the other parts of the license in action is: http://www.gnu.org/software/make/manual/make.html (visited Wed Aug 17 16:42:56 2005, my browser telling me last changed Tue, 28 Oct 2003 20:40:35 GMT):
the html in one page version: The only place where I find "permission is granted" (even in the html source) is in the description within the licence how to apply it.
The ascii text version telling: Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.1 or any later version published by the Free Software Foundation; with no Invariant Sections, with no Front-Cover Texts, and with no Back-Cover Texts. A copy of the license is included in the section entitled "GNU Free Documentation License".
But the best is the (hopefully old and forgotten) postscript version: Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.1 or any later version published by the Free Software Foundation; with the Invariant Sections being "GNU General Public License", the Front-Cover Texts being "A GNU Manual", and with the Back-Cover Texts being as in (a) below. A copy of the license is included in the section entitled "GNU Free Do cumen tation License". (a) The FSF's Back-Cover Text is: You have freedom to copy and modify this GNU Manual, like GNU software. Copies published by the Free Software Foundation raise funds for GNU development.
On Wed, 2005-08-17 at 17:07 +0200, Bernhard R. Link wrote:
And while they have all obvious and subtle differences, and may be best when written from scratch, this is just impossible. By sheer lack of time I can neigther rewrite my Operating System every time I it for some other target, nor can I write enough documentation without heavily using everything else documentating this program including the source code of it.
Rewriting an operating system or rewriting a document/book are 2 entirely different tasks, and you either do not write OSs or do not write books if you try to make such a silly comparison. Why do you think each year get printed many thousands of new books and no OSs ?
Thus the information and its copyrightable containers in forms of sentences and paragraphs must be able to flow between all these different forms and ideally the source code of the machine interpreted code of the program, too.
And you can do that for sentences, and probably small paragraphs too.
Adding unnecessary limits to this flow between different aspects (What is a title page of a manpage? What that of a cup? Where do I "include a unaltered copy" of the 3278 words license on a mouse pad? Does every manpage count as "document" and has to contain the license and all those possible invariant sections?) or between the documentation and the program (Are the texts of the tooltips only mere aggregation though all of this logik when to show which? Or do I have to write them in a clear-room implementation on my own?) keeps us away from getting free programs with free documentation.
When will people stop whining about things like this? It is just not the truth, don't you know that you can cite other authors works? Citation is permitted by copyright law, so you can extract a little portion of even a GFDL covered book and put it on a Mouse Pad without problems. Do you also know you can contact the author and ask for permission to get substantial parts of a text and agree on a different license?
So please, whenever you have or feel to release something under GFDL, please consider dual-license it adding the permissions of the GNU GPL or any other free software compatible with the program documented.
No, you can't generalize like that. If I express my political views in a paragraph, I want them to stay as they are, verbatim, that's why invariant sections exists. If they didn't exist I would be forced to put all the work under a verbatim license as I do not agree others can change MY thoughts. So the GFDL is a lot more Free then a verbatim license for instance and give enough freedom to change technical parts, correct errors, update info in a text without falling in revisionism.
I would consider a 1.new license that give the permission to use the variant parts of the text in other works under the GNU GPL license. That probably makes sense.
Simo.
Simo Sorce wrote:
On Wed, 2005-08-17 at 17:07 +0200, Bernhard R. Link wrote:
So please, whenever you have or feel to release something under GFDL, please consider dual-license it adding the permissions of the GNU GPL or any other free software compatible with the program documented.
No, you can't generalize like that. If I express my political views in a paragraph, I want them to stay as they are, verbatim, that's why invariant sections exists. If they didn't exist I would be forced to put all the work under a verbatim license as I do not agree others can change MY thoughts. So the GFDL is a lot more Free then a verbatim license for instance and give enough freedom to change technical parts, correct errors, update info in a text without falling in revisionism.
I've long had difficulty with this rationale for invariant sections.
It's the nature of digital works that they can be manipulated, and it seems to me to be appropriate that people recognize that a text that is not directly from the author, or attested to by the author, or PGP-stamped by the author, may have been modified. I much prefer simply PGP-stamping text as a way of providing means to ascertain that the author's actual words are represented. I always say what I write is out there in a malleable form; go ahead and do what you wish with it; just don't go misrepresenting me. This is the basic principle that we apply when people relay messages verbally, and it's part of the logic of paraphrasing, where you're expected to "go to the source" before taking somebody's version of what somebody else said at face value. We continue to assume somebody's paraphrasing unless they indicate a direct quote with quotation marks.
Invariant sections are a strange concept when compared to traditional copyright jurisprudence -- they seem to project the idea that you have the ability to directly control what people do with the content of your expressive works once they're released into the wild, which is not the real nature of copyright, WIPO and DMCA notwithstanding. Copyright is a recourse that can be taken when infringements occur; it's not really about direct control.
Seth Johnson
On Thu, 2005-08-18 at 05:02 -0400, Seth Johnson wrote:
It's the nature of digital works that they can be manipulated,
True, nobody can object to that.
and it seems to me to be appropriate that people recognize that a text that is not directly from the author, or attested to by the author, or PGP-stamped by the author, may have been modified. I much prefer simply PGP-stamping text as a way of providing means to ascertain that the author's actual words are represented. I always say what I write is out there in a malleable form; go ahead and do what you wish with it; just don't go misrepresenting me.
Yes you are true, but the difference is that you say that people can, on purpose, misrepresent you. That's possible of course, that's why there are laws that sanction that.
This is the basic principle that we apply when people relay messages verbally, and it's part of the logic of paraphrasing, where you're expected to "go to the source" before taking somebody's version of what somebody else said at face value. We continue to assume somebody's paraphrasing unless they indicate a direct quote with quotation marks.
I think invariant section are not there to prevent people to willfully change your words, but to do that by mistake or in good faith.
Invariant sections are a strange concept when compared to traditional copyright jurisprudence -- they seem to project the idea that you have the ability to directly control what people do with the content of your expressive works once they're released
Control the content?? As copyright is about expression they can cover only the expression of my ideas not the content. You can go and say whatever you want of my ideas, in a GFDL text you can also add a new paragraph and "tell the truth" about the invariant section you do not like, it is just silly but possible.
into the wild, which is not the real nature of copyright, WIPO and DMCA notwithstanding. Copyright is a recourse that can be taken when infringements occur; it's not really about direct control.
You can't honestly compare an invariant section to a DRM mechanism, please. Do you think the same of Verbatim copy texts ?
I think we can get away with copyright if we wish, and state clearly that everything written in the digital age has the same value as oral transmission, as modification is now trivial, I can accept that without any problem. But if you want to speak about a license you should assume all the legal framework and it's consequences.
Simo.
* Simo Sorce simo.sorce@xsec.it [050818 10:55]:
And while they have all obvious and subtle differences, and may be best when written from scratch, this is just impossible. By sheer lack of time I can neigther rewrite my Operating System every time I it for some other target, nor can I write enough documentation without heavily using everything else documentating this program including the source code of it.
Rewriting an operating system or rewriting a document/book are 2 entirely different tasks, and you either do not write OSs or do not write books if you try to make such a silly comparison. Why do you think each year get printed many thousands of new books and no OSs ?
Thanks for calling my text silly. May I answer with a plea to read what I wrote before answering? Thanks.
As I wrote, I'm not speaking about books (And you assume right, I never wrote a book). I'm speaking about documentation for free software. And I hope you will agree that there if definitly a large lack of it.
Thus the information and its copyrightable containers in forms of sentences and paragraphs must be able to flow between all these different forms and ideally the source code of the machine interpreted code of the program, too.
And you can do that for sentences, and probably small paragraphs too.
I may be able to do that for some stentences or for a small passages, as that might not be enough to be counted as protected by copyright law, but I will never know which jurisdiction will allow me how much?
Adding unnecessary limits to this flow between different aspects (What is a title page of a manpage? What that of a cup? Where do I "include a unaltered copy" of the 3278 words license on a mouse pad? Does every manpage count as "document" and has to contain the license and all those possible invariant sections?) or between the documentation and the program (Are the texts of the tooltips only mere aggregation though all of this logik when to show which? Or do I have to write them in a clear-room implementation on my own?) keeps us away from getting free programs with free documentation.
[...offensive sentence interjection removed...] It is just not the truth, don't you know that you can cite other authors works? Citation is permitted by copyright law, so you can extract a little portion of even a GFDL covered book and put it on a Mouse Pad without problems.
Well, first of all citation is a local property of the law, i.e. there are many different laws in different countries. Some people even claim there are countries with no such fair use at all. Even if citation is allowed, law normaly uses this word for a very specific meaning of citation. As far as I as non-lawyer understand it, the German law for example asks for the citing work to have enough independent creation, to be verbatim or marking all modifications including skipping of words. Things are compressing whole sentences changing notations to fit with the rest are definitly no citation.
[...offensive prephrase removed..] you can contact the author and ask for permission to get substantial parts of a text and agree on a different license [...]
This might be good for a work from a single author, but gets harder the more the freedom is used and the more people contribute. Asking for another license mostly means that the license does not fit. Needings authors to agree later means the authors (or copyright owners) can allow or forbid me as they see fit. They have this right, but I'm reluctant to call anything requiring that "free".
So please, whenever you have or feel to release something under GFDL, please consider dual-license it adding the permissions of the GNU GPL or any other free software compatible with the program documented.
No, you can't generalize like that. If I express my political views in a paragraph, I want them to stay as they are, verbatim, that's why invariant sections exists. If they didn't exist I would be forced to put all the work under a verbatim license as I do not agree others can change MY thoughts. So the GFDL is a lot more Free then a verbatim license for instance and give enough freedom to change technical parts, correct errors, update info in a text without falling in revisionism.
Well, invariant sections are another building site. The GFDL makes it easy in this case to produce non-free documents that are more free than a verbatim license. In my eyes that is a disadvantage of it, as it keeps people from using even more free possibilities, like adding removeable invariant sections.
Hochachtungsvoll, Bernhard R. Link
On Thu, 2005-08-18 at 19:16 +0200, Bernhard R. Link wrote:
Rewriting an operating system or rewriting a document/book are 2 entirely different tasks, and you either do not write OSs or do not write books if you try to make such a silly comparison. Why do you think each year get printed many thousands of new books and no OSs ?
Thanks for calling my text silly. May I answer with a plea to read what I wrote before answering? Thanks.
Sure, you could if I were not in topic, can't I say your text is silly if I think it is?
As I wrote, I'm not speaking about books (And you assume right, I never wrote a book). I'm speaking about documentation for free software.
That's why I wrote "document/book", I know you're speaking about writing documentation for free software. But your arguments can apply only when the docs are so voluminous they can also be called a book, in other cases I do not think there's any problem in rewriting a document if the GFDL does not meet your taste.
And I hope you will agree that there if definitly a large lack of it.
Yes there's a lack of it, but having seen how documentation is built I can say I've never seen the licensing problems you cite, they are theoretical problems, but I have never heard a real story of people being unable to extend suitably an GFDL text because of license incompatibilities, if you have such stories then make them public, they will help in debating over real problems of the GFDL.
Thus the information and its copyrightable containers in forms of sentences and paragraphs must be able to flow between all these different forms and ideally the source code of the machine interpreted code of the program, too.
And you can do that for sentences, and probably small paragraphs too.
I may be able to do that for some stentences or for a small passages, as that might not be enough to be counted as protected by copyright law, but I will never know which jurisdiction will allow me how much?
It's enough it is ok in your jurisdiction as the Berne convention applies for others most probably and I do not think it is a big problem to ask the single author of a sentence the permission to cite it. I do not consider plausible a scenario where you want to cite thousands of sentences, that means you're basically copying over a text not citing it.
[...offensive sentence interjection removed...]
Here and later, I do not see offense in my sentence, I apologize if that's what it ended up, but I see that removing the sentence and replacing it as you do may give the wrong impression to the casual reader, I do not like tricks like this.
[...offensive prephrase removed..] you can contact the author and ask for permission to get substantial parts of a text and agree on a different license [...]
This might be good for a work from a single author, but gets harder the more the freedom is used and the more people contribute. Asking for another license mostly means that the license does not fit. Needings authors to agree later means the authors (or copyright owners) can allow or forbid me as they see fit. They have this right, but I'm reluctant to call anything requiring that "free".
So you are reluctant to call the GNU GPL free because you are required to ask permission to the author if you want to copy part of the GPL covered work in your BSD license covered work without being forced to move to the GPL license yourself ?
Interesting.
So please, whenever you have or feel to release something under GFDL, please consider dual-license it adding the permissions of the GNU GPL or any other free software compatible with the program documented.
No, you can't generalize like that. If I express my political views in a paragraph, I want them to stay as they are, verbatim, that's why invariant sections exists. If they didn't exist I would be forced to put all the work under a verbatim license as I do not agree others can change MY thoughts. So the GFDL is a lot more Free then a verbatim license for instance and give enough freedom to change technical parts, correct errors, update info in a text without falling in revisionism.
Well, invariant sections are another building site. The GFDL makes it easy in this case to produce non-free documents that are more free than a verbatim license. In my eyes that is a disadvantage of it, as it keeps people from using even more free possibilities, like adding removeable invariant sections.
So if invariant section are "another" problem, which is the real problem? Just license incompatibility with GPLed code? I think invariant sections are the real problem and being able to remove them may be good, there's obviously room for improvements in the GFDL License, but it has born for a real need and I think your arguments against it (when used for the right thing of course) are weak.
Simo.
* Simo Sorce simo.sorce@xsec.it [050818 19:56]:
Sure, you could if I were not in topic, can't I say your text is silly if I think it is?
You can say whatever you want. But you are hard to distinguish from a troll the way you write.
That's why I wrote "document/book", I know you're speaking about writing documentation for free software. But your arguments can apply only when the docs are so voluminous they can also be called a book, in other cases I do not think there's any problem in rewriting a document if the GFDL does not meet your taste.
I recently (well, for fitting values of recently) rewrote a manpage due to this. And you can be assured that my arguments are not only theoretical but influenced from the utter frustration while doing so.
Yes there's a lack of it, but having seen how documentation is built I can say I've never seen the licensing problems you cite, they are theoretical problems, but I have never heard a real story of people being unable to extend suitably an GFDL text because of license incompatibilities, if you have such stories then make them public, they will help in debating over real problems of the GFDL.
I won't answer to such a troll paragraph. Sorry.
I may be able to do that for some stentences or for a small passages, as that might not be enough to be counted as protected by copyright law, but I will never know which jurisdiction will allow me how much?
It's enough it is ok in your jurisdiction as the Berne convention applies for others most probably and I do not think it is a big problem to ask the single author of a sentence the permission to cite it. I do not consider plausible a scenario where you want to cite thousands of sentences, that means you're basically copying over a text not citing it.
I'm sorry to inform you that the Berne convention is no law, nowhere. It's a treaty between states what kind of laws they want to make, and only those laws are relevant. (Unless you are a country and want to sue another country before the WIPO or things like that).
German law for example only allows me eighter to take portions and fragments too small to be protected. (Which I think hardly applies to taking half a sentence from each of several dozend or hundred of program options) or to cite, which I only discussed in the mail you are answering).
So you are reluctant to call the GNU GPL free because you are required to ask permission to the author if you want to copy part of the GPL covered work in your BSD license covered work without being forced to move to the GPL license yourself ?
No, but I'd consider a license non-free that only allows to me to use the rights to use, copy, modify and so forth to make e.g. an OCR software. Or if the license only allows no translations into other languages and only compilation into arm-code. (Also when someone arguments people might use obscure languages to remove the freedom to use it and compilation might be use compilation to translate it to another language, and that the program in question is only usefull on that one embedded device only having an arm processor).
So if invariant section are "another" problem, which is the real problem? Just license incompatibility with GPLed code?
The uselessness for free software documentation due to this (incompatibility with GPLed code) and lacking support for most documentation formats but printed and electronic books, as I wrote in my previous mails...
Bernhard R. Link
Jeroen Dekkers jeroen@vrijschrift.org writes:
The FSF can't even write a decent documentation license
Richard has mentioned publicly that the community's issues with the GFDL will be fixed. He didn't go into the details though. The GPLv3 is a higher priority, so the revising of the GFDL will probably happen in 2007.
Ciaran O'Riordan ciaran@fsfe.org wrote:
Richard has mentioned publicly that the community's issues with the GFDL will be fixed. He didn't go into the details though. [...]
When/where?
Thanks,
MJ Ray mjr@phonecoop.coop writes:
Ciaran O'Riordan ciaran@fsfe.org wrote:
Richard has mentioned publicly that the community's issues with the GFDL will be fixed. He didn't go into the details though. [...]
When/where?
In the Q&A after one of his talks. I heard it in a recording I found on line. I don't remember which one though.
I think it was one recorded some time in the last year, but it wasn't the June 1st swpat conference, and it wasn't the copyright talk he gave in Taiwan.
'fraid I don't recall any more detail than that. He just gave a one sentence answer to a question.
If I remember or come across it I'll let you know.
At 16 Aug 2005 22:40:19 +0100, Ciaran O'Riordan wrote:
MJ Ray mjr@phonecoop.coop writes:
Ciaran O'Riordan ciaran@fsfe.org wrote:
Richard has mentioned publicly that the community's issues with the GFDL will be fixed. He didn't go into the details though. [...]
When/where?
In the Q&A after one of his talks. I heard it in a recording I found on line. I don't remember which one though.
I think it was one recorded some time in the last year, but it wasn't the June 1st swpat conference, and it wasn't the copyright talk he gave in Taiwan.
'fraid I don't recall any more detail than that. He just gave a one sentence answer to a question.
If I remember or come across it I'll let you know.
I've heard the same thing from a Debian developer, who probably knows it from the FSF-Debian committee which was formed to talk about the GFDL or maybe he just knows it from the same talk.
Jeroen Dekkers
On Sun, Aug 14, 2005 at 09:33:43AM +0300, NSK wrote:
There is already Creative Commons, a project of Lawrence Lessig which also focuses on music.
Note that Creative Commons publishes a license construction set. A CC license can easily make your software non-free.
It is good that they publish such a set of licensing components, because it enables people to publish work with more degrees of freedom easily and it help the debate.
The difference to GNU for Software is, that Creative Commons does not give a good answer to what freedom is actually needed and why. For Software the debate has been lasting decates, with software and music this is more complicated.
Bernhard
On 14-Aug-2005, Roland Häder wrote:
I mean, we already a license for software (GNU GPL) and also for documentation (FDL) but no license for music (FML = Free Music License?). But you cannot compare music with software nor documentation. Music is different to them.
It's not different under copyright law though. If you want to grant the same freedoms, you can use the same license for *all* your software: programs, documentation, music, images, data, ...
The GPL talks about "program" and "source code"; so long as you make it clear what you consider those terms to apply to, the GPL should not be problematic to apply to any copyrighted work if you want.
Ben Finney ben@benfinney.id.au wrote:
On 14-Aug-2005, Roland H=E4der wrote:
[...] But you cannot compare music with software nor documentation. Music is different to them.
It's not different under copyright law though. If you want to grant the same freedoms, you can use the same license for *all* your software: programs, documentation, music, images, data, ...
I agree with the general point: we shouldn't rely on distinguishing bit streams to determine which standard of freedom applies. Sadly, some copyright laws do distinguish documentation and music, but I'm still not sure whether it's enough to require different licences. The best situation for music in a GPL package is the GPL, which is quite clever in its definitions of "program" and "source code" and lets it cover any software, including music and documentation.
A possible alternative is the Open Audio License from EFF, but I don't remember if that's a free software licence or not. CC-scotland is a big improvement on CC 2 and I hope CC 3 will include similar fixes to make some variants (like CC-by) free software friendly.
I'll be discussing free software other than programs at Scundog in Yorkshire on Tuesday 30 August, as well as demo'ing the koha library system. See http://www.scundog.org/ for details and directions. Please let me know if you're coming and have a question I should prepare for!
Best wishes,
On Sun, Aug 14, 2005 at 12:11:39PM +0100, MJ Ray wrote:
Ben Finney ben@benfinney.id.au wrote:
On 14-Aug-2005, Roland H=E4der wrote:
[...] But you cannot compare music with software nor documentation. Music is different to them.
It's not different under copyright law though. If you want to grant the same freedoms, you can use the same license for *all* your software: programs, documentation, music, images, data, ...
I agree with the general point: we shouldn't rely on distinguishing bit streams to determine which standard of freedom applies. Sadly, some copyright laws do distinguish documentation and music, but I'm still not sure whether it's enough to require different licences.
From the legal point of view, I am quite convinced that there is enough difference to warrant different licenses.
Getting to the social point and the freedom: This is an unanswered question with contributions from several hundered years of debate. With Software finding the balance intellectually, is fairly easy, because it has a very new level of being funcational.
Melodies, Songs, Lyrics, Literature, News reports, law texts, essays they are have a different role in our society that might require a slightly different balance. Saying this all can be represented as a bitstream and thus should be treated the same is makeing life too simpley and disregards the knowledge about their role thate we have already aquired.
Bernhard
On Sun, 2005-08-14 at 18:40 +1000, Ben Finney wrote:
The GPL talks about "program" and "source code"; so long as you make it clear what you consider those terms to apply to, the GPL should not be problematic to apply to any copyrighted work if you want.
I'm afraid I think the GPL would be a terrible licence to apply to music. I can't think of a sensible definition of "source" and "program" that would apply to music without trying to fudge things (i.e., trying to treat a performance the same as compilation). The GPL also doesn't recognise the separate, severable copyrights that subsist in most music - all of which must be licensed. Trying to licence all those copyrights simultaneously under the GPL would result in a dog's breakfast, in my opinion - I don't know how you could define what was "source" with regards a performance, for example, in a way that made sense wrt. the obligations of the GPL.
Music is a special case - the law treats it very differently (i.e., having several copyrights subsist in a single work - you don't usually have different rights to the same thing) and I think that demands a different licensing scheme.
Cheers,
Alex.
On Sun, 14 Aug 2005, Ben Finney wrote:
On 14-Aug-2005, Roland Häder wrote:
I mean, we already a license for software (GNU GPL) and also for documentation (FDL) but no license for music (FML = Free Music License?). But you cannot compare music with software nor documentation. Music is different to them.
It's not different under copyright law though. If you want to grant the same freedoms, you can use the same license for *all* your software: programs, documentation, music, images, data, ...
The GPL talks about "program" and "source code"; so long as you make it clear what you consider those terms to apply to, the GPL should not be problematic to apply to any copyrighted work if you want.
The definition of "Program" in 0. is also broader as they use work as definition :
"The "Program", below, refers to any such program or work,..."
So as you said, if you are able to make a clear separation between "work" and "the preferred form of the work for making modifications to it." the GPL can be used for any copyrighted work.
Keeping a single free license in the free community is better for the dynamic of the community.
On Mon, Aug 15, 2005 at 10:31:45AM +0200, Alexandre Dulaunoy wrote:
On Sun, 14 Aug 2005, Ben Finney wrote:
On 14-Aug-2005, Roland Häder wrote:
I mean, we already a license for software (GNU GPL) and also for documentation (FDL) but no license for music (FML = Free Music License?). But you cannot compare music with software nor documentation. Music is different to them.
It's not different under copyright law though. If you want to grant the same freedoms, you can use the same license for *all* your software: programs, documentation, music, images, data, ...
The GPL talks about "program" and "source code"; so long as you make it clear what you consider those terms to apply to, the GPL should not be problematic to apply to any copyrighted work if you want.
The definition of "Program" in 0. is also broader as they use work as definition :
"The "Program", below, refers to any such program or work,..."
So as you said, if you are able to make a clear separation between "work" and "the preferred form of the work for making modifications to it." the GPL can be used for any copyrighted work.
Keeping a single free license in the free community is better for the dynamic of the community.
Notwithstanding the possibility to bend the text of the GNU GPL to cover musical works, it should be noticed that the latter are subject to a variety of rights (so-called "neihbouring rights") which are hardly applicable to computer programs, and which the GNU GPL does not consider as far as I can understand - which is completely right, given that such license was meant primarily for software.
Among such rights, we can consider the rights of performers, of phonogram producers, of broadcasters, etc.
The Convention of Rome and the WPPT are a good reference on this topic, as well as the relevant national/regional legislation.
So, even assuming you would want to apply the GNU GPL to musical works, you would need to manage such neighbouring rights in some way or another. The GNU GPL alone will not suffice (nor will CC licenses, for that).
I will not dwelve into considerations of whether the business and social dynamics that are possible and beneficial for software can be applied to music production and consumption (I believe not, but it is a large topic and I am not really equipped to discuss it with the necessary precision).
Cheers,
Andrea
Keeping a single free license in the free community is better for the dynamic of the community.
Software, music and documentation, articles, ..., are all different kind of works, and need different kind of protections. Applying a hammer when you wish to screw in a screw, is well... Silly, to say the least. You don't need to be change my article about who much I love feries, to say that I don't love them. Fo functional works (software, for example) one needs the right for modification, fo non-functional (non-technical written works) one doesn't. Music, and a novel, are quite similar, and a simple `verbatim copying allowed' is quite satisifactory for such works. A manual, or a program, is where you need more freedom since it is useful to change a manual, or a program. But a manual is very different from a program, so a manual needs to define a different set of freedoms.
Keeping a single license, that is applied to everything, will only hurt the community in the long run.
Cheers.
On Mon, 15 Aug 2005, Alfred M. Szmidt wrote:
Keeping a single free license in the free community is better for the dynamic of the community.
Software, music and documentation, articles, ..., are all different kind of works, and need different kind of protections. Applying a hammer when you wish to screw in a screw, is well... Silly, to say the least. You don't need to be change my article about who much I love feries, to say that I don't love them. Fo functional works (software, for example) one needs the right for modification, fo non-functional (non-technical written works) one doesn't. Music, and a novel, are quite similar, and a simple `verbatim copying allowed' is quite satisifactory for such works. A manual, or a program, is where you need more freedom since it is useful to change a manual, or a program. But a manual is very different from a program, so a manual needs to define a different set of freedoms.
The subject is a very difficult matter. We got a very similar discussion in the fsf-france mailing-list sometimes ago without any conclusion.
http://lists.gnu.org/archive/html/fsfe-france/2005-05/msg00001.html (sorry mainly in french but maybe some references could be useful)
I was thinking that the differentiation between functional and non-functional was easy. But working on various projects, I found that is not the case following the evolution of usage and distribution of "digital object". If we want to keep them free, the right of modification is very important... A nice example is digital archiving. In digital archiving, you have to protect the data and meaning of the data. If we take into consideration just the data (without the meaning) it's already difficult to keep free a work without making modification (because conversion is often (always?) modification in digital format).
Another nice example is the representation of an object in 3D for a free software a game. If you make a modification in the description of the object (a vertex for example because the render is unable to parse it), the result (representation) can be the same but the work has been modified. You can also make modification or conversion of fonts and keep the same representation of the work.
A lot of non-functional work has a functional description behind. Can we disallow modification of "non-functional" free work ? or does the separation make sense today ?
{For me, the questions are still open...}
Keeping a single license, that is applied to everything, will only hurt the community in the long run.
Making new "virtual" boundaries between digital free works will limit their current and future usage.
IMHO,
adulau
On Mon, Aug 15, 2005 at 02:33:29PM +0200, Alexandre Dulaunoy wrote:
I was thinking that the differentiation between functional and non-functional was easy. But working on various projects, I found that is not the case following the evolution of usage and distribution of "digital object".
It might be difficult, but that does not mean it cannot be done.
If we want to keep them free, the right of modification is very important...
I agree, but there are limit where the compromise in the other direction actually helps the freedom in the society.
A nice example is digital archiving. In digital archiving, you have to protect the data and meaning of the data. If we take into consideration just the data (without the meaning) it's already difficult to keep free a work without making modification (because conversion is often (always?) modification in digital format).
We could consider a mere transformation just that and not a modification.
A lot of non-functional work has a functional description behind.
Can you give an example for this?
Can we disallow modification of "non-functional" free work ? or does the separation make sense today ?
{For me, the questions are still open...}
You to throw the motivations of the authors into the ring. Each bit of information needs to be written or assembled which is effrot. If we want quality works we need to compensate for this effort.
I remember an article that claimed that around in the French revolution, copyright was completely abandoned and the quality of publications went down significantly. Unfortunately I do not have the reference on this one. It was a few years ago.
Keeping a single license, that is applied to everything, will only hurt the community in the long run.
Making new "virtual" boundaries between digital free works will limit their current and future usage.
If the aim is: How do we bring in more freedom into a human and democratic world, we can make different compromises. I think that categories will help the process a lot more than burning all copy-rights on the same bonfire, like legislators tend to do.
To make realistic suggestions for the politicians, we need to accept that many people accept the idea of economic reward.
Bernhard
On Wed, 2005-08-17 at 13:03 +0200, Bernhard Reiter wrote:
On Mon, Aug 15, 2005 at 02:33:29PM +0200, Alexandre Dulaunoy wrote:
I was thinking that the differentiation between functional and non-functional was easy. But working on various projects, I found that is not the case following the evolution of usage and distribution of "digital object".
It might be difficult, but that does not mean it cannot be done.
If we want to keep them free, the right of modification is very important...
I agree, but there are limit where the compromise in the other direction actually helps the freedom in the society.
The perfect compromise would be to say: "As long as you preserve the meaning you can modify the work to fit technical needs".
Except that we have a problem in defining both "meaning" and "technical", so to ease things we should try to simplify b categorizing things and proposing license for each category.
You to throw the motivations of the authors into the ring. Each bit of information needs to be written or assembled which is effrot. If we want quality works we need to compensate for this effort.
Are you sure? It's true that compensation give someone a chance to dedicate more time to a matter, but are you sure there's a direct link between compensation and quality?
I'd say no, the link is between motivation and skills, and quality. Now granted you have the right skills, in some cases compensation can be the right motivator. But compensation is not enough in most of the cases, and does not matte at all in many cases.
I remember an article that claimed that around in the French revolution, copyright was completely abandoned and the quality of publications went down significantly. Unfortunately I do not have the reference on this one. It was a few years ago.
Even if such data exist I would like to see how they have been put in relation to each other. What kind of measure for quality have been adopted, how the general economical and political status have been weighted and so on, it is to take to set of data and link them together through a time relation (they happened at the same time so they are linked), but timing is not enough to explain most facts.
I think that categories will help the process a lot more than burning all copy-rights on the same bonfire, like legislators tend to do.
I agree, oversimplification some times is bad.
To make realistic suggestions for the politicians, we need to accept that many people accept the idea of economic reward.
I do not see the link, under this line of reasoning you just advocate the status quo as that is accepted by most people ...
Simo.
Hi Simo,
On Thu, Aug 18, 2005 at 11:16:51AM +0200, Simo Sorce wrote:
On Wed, 2005-08-17 at 13:03 +0200, Bernhard Reiter wrote:
I agree, but there are limit where the compromise in the other direction actually helps the freedom in the society.
The perfect compromise would be to say: "As long as you preserve the meaning you can modify the work to fit technical needs".
I expect a different compromise for each category of works to be a good solution. For software, the four freedoms seem to be a good compromise to me, for literature, music, law texts or instruction books I am missing this. There are examples where meaning and technical needs are intervoven and there are examples where I believe that the right to change the meaning of the author is a good thing.
Except that we have a problem in defining both "meaning" and "technical", so to ease things we should try to simplify b categorizing things and proposing license for each category.
I agree.
You to throw the motivations of the authors into the ring.
have
Each bit of information needs to be written or assembled which is effrot. If we want quality works we need to compensate for this effort.
Are you sure? It's true that compensation give someone a chance to dedicate more time to a matter, but are you sure there's a direct link between compensation and quality?
I am sure there is a connection, though it is not a simple relation. Note that I have deliberately not wrote "paied" or "money" as I wanted a more neutral tone.
I'd say no, the link is between motivation and skills, and quality. Now granted you have the right skills, in some cases compensation can be the right motivator. But compensation is not enough in most of the cases, and does not matte at all in many cases.
Your "motivation" seems to be similiar to what I called "compensation". In admitting that money can be the right motivatior in a fraction of cases, you admitted a connection. Also I am not ready to completely throw away the benefits of competition and a market. It has its limits, but also its advantages.
A lot of the high quality "content" that I make use of is done by professionals which are in a competitive market. An example are the newspaper I read. Of course there is also other high quality "contents".
I remember an article that claimed that around in the French revolution, copyright was completely abandoned and the quality of publications went down significantly. Unfortunately I do not have the reference on this one. It was a few years ago.
Even if such data exist I would like to see how they have been put in relation to each other. What kind of measure for quality have been adopted, how the general economical and political status have been weighted and so on, it is to take to set of data and link them together through a time relation (they happened at the same time so they are linked), but timing is not enough to explain most facts.
Probably this is a broad subject and I would love to find that article again.
To make realistic suggestions for the politicians, we need to accept that many people accept the idea of economic reward.
I do not see the link, under this line of reasoning you just advocate the status quo as that is accepted by most people ...
No, I recommend that our recipe should include an answer for those people that have accepted the idea of the economic stimulation. Anything we propose will not be heard until we have a good answer to them.
Bernhard
On Thu, 2005-08-18 at 16:11 +0200, Bernhard Reiter wrote:
On Thu, Aug 18, 2005 at 11:16:51AM +0200, Simo Sorce wrote:
On Wed, 2005-08-17 at 13:03 +0200, Bernhard Reiter wrote:
I agree, but there are limit where the compromise in the other direction actually helps the freedom in the society.
The perfect compromise would be to say: "As long as you preserve the meaning you can modify the work to fit technical needs".
I expect a different compromise for each category of works to be a good solution.
I agree
Are you sure? It's true that compensation give someone a chance to dedicate more time to a matter, but are you sure there's a direct link between compensation and quality?
I am sure there is a connection, though it is not a simple relation. Note that I have deliberately not wrote "paied" or "money" as I wanted a more neutral tone.
Ok I assumed you were speaking of economic compensation == money, if you give a broader meaning to compensation I mostly agree with you.
I'd say no, the link is between motivation and skills, and quality. Now granted you have the right skills, in some cases compensation can be the right motivator. But compensation is not enough in most of the cases, and does not matte at all in many cases.
Your "motivation" seems to be similiar to what I called "compensation".
Yes, I used motivation because some times people do things (even great ones) without expecting any compensation (think of anonymous works in the Public Domain).
In admitting that money can be the right motivatior in a fraction of cases, you admitted a connection. Also I am not ready to completely throw away the benefits of competition and a market. It has its limits, but also its advantages.
How can you throw away a market? The only way is through complete monopolization.
A lot of the high quality "content" that I make use of is done by professionals which are in a competitive market. An example are the newspaper I read.
Oh the more I read newspapers the more I'd say the opposite about their quality :-), but I agree on the concept underlying this example.
To make realistic suggestions for the politicians, we need to accept that many people accept the idea of economic reward.
I do not see the link, under this line of reasoning you just advocate the status quo as that is accepted by most people ...
No, I recommend that our recipe should include an answer for those people that have accepted the idea of the economic stimulation. Anything we propose will not be heard until we have a good answer to them.
Ok I agree on this, it's just that I do not see why we should ban the idea of economic stimulation, it's just about balancing it with other fundamental rights.
Simo.
On Thu, Aug 18, 2005 at 04:42:20PM +0200, Simo Sorce wrote:
On Thu, 2005-08-18 at 16:11 +0200, Bernhard Reiter wrote:
On Thu, Aug 18, 2005 at 11:16:51AM +0200, Simo Sorce wrote:
On Wed, 2005-08-17 at 13:03 +0200, Bernhard Reiter wrote:
In admitting that money can be the right motivatior in a fraction of cases, you admitted a connection. Also I am not ready to completely throw away the benefits of competition and a market. It has its limits, but also its advantages.
How can you throw away a market? The only way is through complete monopolization.
I wrote this, because: If you accept the benefits of a market, you also accept that economic motivation can work.
I believe they can work.
I was thinking that the differentiation between functional and non-functional was easy.
It isn't easy, if it was, then we wouldn't have these problems. The license for each functional, and non-functional work should be based on the situation for what it should be used for. For example, Ogg Vorbis was licensed under the BSD license to allow for wider use, instead for being licensed under the GNU GPL; and this was done with the blessing of the FSF and the GNU project. A music sample might be better licensed under a license like the BSD license, my memoairs might be better licensed under a verbatim-copying-only license, your instructions on how to make great cookie might be better to release into the public domain. For non-functional works (say a painting, a music piece) a verbatim license is quite good, for the most time, if it isn't one should pick a different licensing model.
It all depends on the situation, and one should thing carefully before licensing and not blindly apply a license to all types of works.
"Alfred M. Szmidt" ams@gnu.org wrote:
Software, music and documentation, articles, ..., are all different kind of works, and need different kind of protections. [...]
*Programs*, music and documentation, articles are different kinds of work. Software is a representation. It can be a representation of programs, music, documentation or whatever. Arguing that it is different is rather like arguing that paper and music are different kinds of work: it's not very illuminating.
The argument that novels are fine as verbatim copying is rather unfortunate. For example, see the problems with works like "the Wind Done Gone" which I think might also exist for verbatim-copy works. Basically, the freedoms to use, study, adapt and copy are still desirable and "verbatim copying only" isn't that. The author of a verbatim-copy opinion piece is cutting their own nose off to spite their face, too: if I need to adapt an idea to my audience and I can't adapt their expression, then I don't use their expression and I'll probably only cite the primary sources. If their article isn't a primary source (which most opinion pieces aren't), they lost a credit.
There may be other copyrights not addressed by the GPL which are interesting for some software. I'm not that familiar with them and no-one in this thread has yet given useful references (the aforementioned volumes of legislation are, er, voluminous). I'm interested in the claim that CC doesn't address them either: if that's true, what are projects like Free Culture and Remix doing about performances?
I may also be attending FAVE/3avc this weekend in Bristol, where we might find people who know more. Anyone got a good list of questions I could take? http://www.fave.org.uk/
On Mon, Aug 15, 2005 at 07:21:42PM +0100, MJ Ray wrote:
There may be other copyrights not addressed by the GPL which are interesting for some software. I'm not that familiar with them and no-one in this thread has yet given useful references (the aforementioned volumes of legislation are, er, voluminous). I'm interested in the claim that CC doesn't address them either: if that's true, what are projects like Free Culture and Remix doing about performances?
They are not "copyrights", they are "neighbouring rights" (they do not, per se, impact on the right to authorize or deny copying).
I'm not sure what you want to know about these rights - could you please be more specific?
The texts I referenced are in my opinion quite useful. They are very short (you can read both treaties in two hours, I think). Of course, you are supposed to know a bit about law, and copyright law specifically, but any reference I could give needs that anyway.
I don't know precisely what Free Culture and Remix are doing with regards to performances, any link or explanation would be good (I know both projects by name, so feel free to skip any description :).
I may also be attending FAVE/3avc this weekend in Bristol, where we might find people who know more. Anyone got a good list of questions I could take? http://www.fave.org.uk/
I was planning to come, since I will be in London in those days. But honestly, it strongly depends on my mood at that time, because these days are supposed to be my vacation, and I really don't need discussing about copyright and licenses (even though I love doing it :).
Ciao, andrea
Andrea Glorioso andrea@digitalpolicy.it wrote:
They are not "copyrights", they are "neighbouring rights" (they do not, per se, impact on the right to authorize or deny copying).
Now I'm really confused. So need a music file's copyright licence cover those rights at all?
[...]
The texts I referenced are in my opinion quite useful. They are very short (you can read both treaties in two hours, I think). [...]
Maybe some time that I have two hours spare... :-/
I don't know precisely what Free Culture and Remix are doing with regards to performances, any link or explanation would be good [...]
As I understand it, they are planning to record performances of CC'd works and combine them in various ways. Do they need to license the neighbouring rights and, if so, what licences are suitably free?
Thanks,
On Tue, 2005-08-16 at 10:11 +0100, MJ Ray wrote:
Andrea Glorioso andrea@digitalpolicy.it wrote:
They are not "copyrights", they are "neighbouring rights" (they do not, per se, impact on the right to authorize or deny copying).
Now I'm really confused. So need a music file's copyright licence cover those rights at all?
I think Andrea is right and wrong. They aren't copyrights in the literary sense, and if you look on the back of a CD or something you'll often find that alongside the usual (C) symbol you also have the (P) right mark.
This is basically because performing music is radically different to, say, compiling software - the performance itself has a creative quality, and the performance right is supposedly attached to that aspect. The duration of that right is only 50 years too, which is why Elvis recordings are nearing the public domain in this country.
However, at least in the UK they are still called copyrights (sometimes "mechanical copyright", which is hysterical). And, over here, you do need to license those rights in order to copy and perform the work.
I don't know precisely what Free Culture and Remix are doing with regards to performances, any link or explanation would be good [...]
As I understand it, they are planning to record performances of CC'd works and combine them in various ways. Do they need to license the neighbouring rights and, if so, what licences are suitably free?
They need to licence the (P) rights as well as the (C) rights. I've never seen a licence which addresses those rights (some CC licences mention performance, but I'm not sure that's good enough - I suspect in many instances the licence isn't coming from the right person/people, to begin with; you'd need separate licences to make sure you knew which rights were coming from who in any case).
Plus, I *think* in this country the score, lyrics and composition of music are all separately copyrightable. So, again, large quantities of permission required.
Cheers,
Alex.
*Programs*, music and documentation, articles are different kinds of work. Software is a representation. It can be a representation of programs, music, documentation or whatever.
Sorry, you are wrong. Software and programs are equivialent; and software is not music, documentation or whatever. Atleast in the world of computing science.
[...] The author of a verbatim-copy opinion piece is cutting their own nose off to spite their face, too: if I need to adapt an idea to my audience and I can't adapt their expression, then I don't use their expression and I'll probably only cite the primary sources.
You can use their expression, aslong as you don't change it. You can also quote pieces (fair use).
"Alfred M. Szmidt" ams@gnu.org wrote:
Sorry, you are wrong. Software and programs are equivialent; and software is not music, documentation or whatever. Atleast in the world of computing science.
The word "software" was first used in print in January 1958 in American Mathematical Monthly by John W Tukey as a sort-of opposite to "hardware" (the machine parts) for the situation when "stored program" did not cover all the things held in memory being described. It was used *because* it is not equivalent to programs. At least in the world of mathematics.
If you define software and programs as equivalent, then you lose detail and must start claiming that data can never be software!
[...] The author of a verbatim-copy opinion piece is cutting their own nose off to spite their face, too: if I need to adapt an idea to my audience and I can't adapt their expression, then I don't use their expression and I'll probably only cite the primary sources.
You can use their expression, aslong as you don't change it. You can also quote pieces (fair use).
The situation given was I need to adapt the idea to the audience, so using their expression unchanged is not useful. I could quote it, but why?
The word "software" was first used in print in January 1958 in American Mathematical Monthly by John W Tukey as a sort-of opposite to "hardware" (the machine parts) for the situation when "stored program" did not cover all the things held in memory being described.
In 1958, a program was stored on cards, and then read into memory and stored to be run (where it became `soft'ware). Today everything is in memory (a drive is a type of memory, but a slow one).
It was used *because* it is not equivalent to programs.
In 1958 you had this distinction, today you don't have the same distiniction.
If you define software and programs as equivalent, then you lose detail and must start claiming that data can never be software!
Data is any type of information that can be expressed in binary. So, data is software. I fail to see how you could draw the opposite conclusion.
You can use their expression, aslong as you don't change it. You can also quote pieces (fair use).
The situation given was I need to adapt the idea to the audience, [...]
So use the idea, ideas aren't copyrighted. I'm quite sure you can come up with a expression that presents the idea in a good way.
On Tue, 2005-08-16 at 10:24 +0100, MJ Ray wrote:
"Alfred M. Szmidt" ams@gnu.org wrote:
Sorry, you are wrong. Software and programs are equivialent; and software is not music, documentation or whatever. Atleast in the world of computing science.
The word "software" was first used in print in January 1958 in American Mathematical Monthly by John W Tukey as a sort-of opposite to "hardware" (the machine parts) for the situation when "stored program" did not cover all the things held in memory being described. It was used *because* it is not equivalent to programs. At least in the world of mathematics.
If you define software and programs as equivalent, then you lose detail and must start claiming that data can never be software!
Thanks for the etymology lesson, unfortunately, commonly programs and software are synonymous and I think that's true even for the term "Free Software". If that's the problem than you better ask FSF what does Software mean in their definition of Free Software and propose to add a FAQ on that topic.
[...] The author of a verbatim-copy opinion piece is cutting their own nose off to spite their face, too: if I need to adapt an idea to my audience and I can't adapt their expression, then I don't use their expression and I'll probably only cite the primary sources.
You can use their expression, aslong as you don't change it. You can also quote pieces (fair use).
The situation given was I need to adapt the idea to the audience, so using their expression unchanged is not useful. I could quote it, but why?
Grow up please! Copyright covers expression not meaning, so you can use another author concept using your words as much as you want, you can't just ascribe the new expression to the author that make you think the new analogy, but you can, of course credit him as the originator of your concept. (And do not forget freedom of expression rights).
Simo.
Simo Sorce simo.sorce@xsec.it wrote: [software and programs]
Thanks for the etymology lesson, unfortunately, commonly programs and software are synonymous and I think that's true even for the term "Free Software".
Commonly they are not synonymous, but when asked for an example of "software" most people will give a type of program.
If that's the problem than you better ask FSF what does Software mean in their definition of Free Software and propose to add a FAQ on that topic.
I know what RMS means by it, I know I disagree and I don't recall any easy way to contact all of FSF, nor do I see the point. FSF are not controllers of the English language.
[...] The author of a verbatim-copy opinion piece is cutting their own nose off to spite their face, too: if I need to adapt an idea to my audience and I can't adapt their expression, then I don't use their expression and I'll probably only cite the primary sources.
You can use their expression, aslong as you don't change it. You can also quote pieces (fair use).
The situation given was I need to adapt the idea to the audience, so using their expression unchanged is not useful. I could quote it, but why?
Grow up please!
...because shouting "Grow up" is such adult debate(!) Hello pot!
Copyright covers expression not meaning, so you can use another author concept using your words as much as you want, you can't just ascribe the new expression to the author that make you think the new analogy, but you can, of course credit him as the originator of your concept. (And do not forget freedom of expression rights).
I mentioned he's not the primary source: that is, he's not the originator of the concept. Sorry if that's too hard for you to understand.
"Freedom of expression rights" have little to override the desire to adapt prose, compared with the desire to adapt programs. Some programmers argue that their work is so elegant and personal, we shouldn't be allowed to modify it. I'm amazed to see it seriously suggested as a reason for the FDL.
I know what RMS means by it, I know I disagree and I don't recall any easy way to contact all of FSF, nor do I see the point. FSF are not controllers of the English language.
Neither are you, in English, software and programs are exactly the same.
"Alfred M. Szmidt" ams@gnu.org wrote:
Neither are you, in English, software and programs are exactly the same.
Insist however you want, but many of us disagree, including some (many?) English people and this isn't a new position. You may care about editing only programs, but others care about editing the data too, with similar arguments and reasons.
At Wed, 24 Aug 2005 19:16:15 +0200, Alfred M. Szmidt wrote:
I know what RMS means by it, I know I disagree and I don't recall any easy way to contact all of FSF, nor do I see the point. FSF are not controllers of the English language.
Neither are you, in English, software and programs are exactly the same.
From WordNet (r) 2.0 (August 2003) [wn]:
software n : (computer science) written programs or procedures or rules and associated documentation pertaining to the operation of a computer system and that are stored in read/write memory; "the market for software is expected to expand" [syn: {software system}, {software package}, {package}] [ant: {hardware}]
Jeroen Dekkers
On Wed, 2005-08-24 at 19:47 +0200, Jeroen Dekkers wrote:
Neither are you, in English, software and programs are exactly the same.
From WordNet (r) 2.0 (August 2003) [wn]:
I'm not sure people are going to get anywhere arguing over terms :o)
There are people who use the word "software" to describe computer programs (as the counterpart to the computer electronics, the "hardware"), and there are others who use the word "software" to describe (digital?) data. Some people may like the latter because it allows software + hardware to be a complete set, whereas the former is not.
(I would argue both senses of the word are useless as taxonomy since neither definition offers a mutually exclusive meaning, but still people ought to recognise that both definitions are in use and neither is more 'correct').
If I can draw an analogy with the GFDL debate, I would say that people on either side of the divide are both arguing for a greater understanding of freedom of data/information. People oppose the GFDL because a computer is unlikely to care much for the difference between different bitstreams (e.g., "text document" versus "Perl script"). Others like the GFDL because it doesn't care much for the difference between different representations of the same work (e.g., "text document" versus "hardback book"). Again, it's not an either-or situation: you can quite sensibly argue from both points of view, and neither is "right".
Cheers,
Alex.
Hi Roland,
On Sun, Aug 14, 2005 at 08:03:58AM +0200, Roland Häder wrote:
What abount a license which gurantees that also the music made for a software project - GNU GPL'ed or compatible, of course -is *free*?
for practical purposes I would put sound effects and musical scores under GNU GPL if they are closely tight to a software under GNU GPL.
I mean, we already a license for software (GNU GPL) and also for documentation (FDL) but no license for music (FML = Free Music License?). But you cannot compare music with software nor documentation. Music is different to them.
There have been a few initiatives.
E.g. the CD done by Linuxtag e.V. in 2001: http://openmusic.linuxtag.org/modules/freecontent/content/openmusic/ it uses Green and Yellow OpenMusic Licenses. I do not remember all the details.
There are more, it would be interesting to first make a survey about the licenses already used. Then we would need to find answer to the question: What freedom does music need and why?
Bernhard