Live data from Hacker News

Linux's GPLv2 licence is routinely violated (2015)

devever.net

71–80 of 202 posts

Re: Linux's GPLv2 licence is routinely violated (2015)

#71

Several of the points in this article are either flat-out wrong, or are correct but being read in the wrong context. > The licence is specifically GPLv2 and not “GPLv2 or later”, which is the default when applying the GPLv2 to software. Because Linus didn't and doesn't want the FSF relicensing Linux with new restrictions to fit their specific political goals[0], on the basis that adding any further restrictions to GP…

What does SCO mean? Couldn't find it in Wikipedia or lists of abbreviations. I think it is/was a company...

Re: Linux's GPLv2 licence is routinely violated (2015)

#72

Several of the points in this article are either flat-out wrong, or are correct but being read in the wrong context. > The licence is specifically GPLv2 and not “GPLv2 or later”, which is the default when applying the GPLv2 to software. Because Linus didn't and doesn't want the FSF relicensing Linux with new restrictions to fit their specific political goals[0], on the basis that adding any further restrictions to GP…

What does SCO mean? Couldn't find it in Wikipedia or lists of abbreviations. I think it is/was a company...

> "The SCO–Linux disputes were a series of legal and public disputes between the software company SCO Group (SCO) and various Linux vendors and users. The SCO Group alleged that its license agreements with IBM meant that source code IBM wrote and donated to be incorporated into Linux was added in violation of SCO's contractual rights. Members of the Linux community disagreed with SCO's claims; IBM, Novell and Red Hat filed claims against SCO."

https://en.wikipedia.org/wiki/SCO–Linux_disputes

Re: Linux's GPLv2 licence is routinely violated (2015)

#73

Earlier quoted context omitted.

I think there's some confusion in the above post. The relevant "derived work" definition for the purpose of linking isn't one stemming from copyright law, it's the one in the GPL itself. This is because the GPL defines what licensees are permitted to do with the licensed work, and it imposes restrictions on those rights. Copyright law by default lets you make very little use of the software - the GPL broadens that ou…

The main problem with the interpretation that the GPL sets conditions for use of software is that it specifically claims it does not: > Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, Now there are some ways to try to do an end-run around this clause, it does raise some questions about how you g…

I find it difficult to imagine a situation in which you could use the library without copying the library. How would you obtain the code to execute it? Someone must distribute it to you, and/or you must copy it. Those are the governed actions.

EDIT: note that the GPL broadly doesn't impose restrictions on interacting with a program over a network ("using" a server), which is why the AGPL exists.

Re: Linux's GPLv2 licence is routinely violated (2015)

#74
post #23

> It is rather ironic that in some ways, one has more freedom in how one licences a kernel driver for Windows than the freedom one supposedly has in writing a kernel module for Linux. it is not ironic at all: freedom in Free Software is the freedom of USERS of the software to be able to see and modify the source. This is a freedom granted by the writers of the software, and copyleft means that this right is ensured t…

>In this same way, the GPL and like licenses are the most permissive licenses from the perspective of every eventual user No, take for example the real world use case of distributing ZFS on Linux in binary form. Users are unable to distribute it due to conflicting copyleft licenses. Copyleft licenses limits people by creating their own exclusive ecosystems which can't be mixed.

> Users are unable to distribute

You're talking about distributors.

Re: Linux's GPLv2 licence is routinely violated (2015)

#75

> The userspace ABI is stable and intended not to cause anything targeting it to become a derived work. > the only thing forcing modules to be GPLv2-compatible is the rate of change of the kernel mode ABI making it impractical to target it via cleanroom methods. What is licensed is a specific source code version, not the history of previous versions. For a specific revision, there is no meaning to "more stable". What…

>> What makes userspace programs non-derived work is the syscall exception

And with the Google vs Oracle decision we know that APIs are not copyrightable anyway. If I can clone an API and create my own implementation then surely someone else can write programs that use my cloned implementation or the original.

Re: Linux's GPLv2 licence is routinely violated (2015)

#76

Earlier quoted context omitted.

I think there's some confusion in the above post. The relevant "derived work" definition for the purpose of linking isn't one stemming from copyright law, it's the one in the GPL itself. This is because the GPL defines what licensees are permitted to do with the licensed work, and it imposes restrictions on those rights. Copyright law by default lets you make very little use of the software - the GPL broadens that ou…

I don't see anywhere in the GPL that defines deriving, except the part where it mentions "derivative work under copyright law".

The relevant clauses read as follows:

     b) You must cause any work that you distribute or publish, that in whole or
        in part contains or is derived from the Program or any part thereof, to 
        be licensed as a whole at no charge to all third parties under the terms 
        of this License.

     [...]

     These requirements apply to the modified work as a whole. If identifiable 
     sections of that work are not derived from the Program, and can be 
     reasonably considered independent and separate works in themselves, then 
     this License, and its terms, do not apply to those sections when you 
     distribute them as separate works. But when you distribute the same 
     sections as part of a whole which is a work based on the Program, the 
     distribution of the whole must be on the terms of this License, whose 
     permissions for other licensees extend to the entire whole, and thus to 
     each and every part regardless of who wrote it. 
The post on which we're commenting says that "derived" comes from copyright law, and not from the text of the GPL, because the GPL cannot supercede copyright law. What I'm pointing out is that's incorrect - the definition of "derived" comes from what the license itself considers derived. This is because the license puts conditions on "derived from" works in the above term. This is true even if the license says "see copyright law for definition of term X".

When trying to determine what "derived works" means, the intent of the license is relevant. The license cannot restrict your rights beyond what copyright law allows it to do, but it can grant you additional rights on terms of its own choosing, including very restrictive terms.

Importantly for this conversation, something that is derived from "any part" of the program may or may not be a "derived work" of the whole under US copyright law, but may be a "derived work" for the purposes of the above clauses regardless. Plus the latter paragraph has an AND addendum - "AND can be reasonably considered independent and separate works in themselves" - which can restrict beyond the referenced legal definition of "derived".

Re: Linux's GPLv2 licence is routinely violated (2015)

#77
by using copyright law as an enforcement mechanism, these licenses make terms which govern behavior controlled by users -- terms which are practically unenforceable

this article is a ridiculous charade, starting out saying that people like Linus don't have the power to interpret the GPL then he goes on for many paragraphs with a more outlandish interpretation of the GPL than i've seen in a long time.

Re: Linux's GPLv2 licence is routinely violated (2015)

#78

Earlier quoted context omitted.

>In this same way, the GPL and like licenses are the most permissive licenses from the perspective of every eventual user No, take for example the real world use case of distributing ZFS on Linux in binary form. Users are unable to distribute it due to conflicting copyleft licenses. Copyleft licenses limits people by creating their own exclusive ecosystems which can't be mixed.

> Users are unable to distribute You're talking about distributors.

Going with the GPL, a user can be a distributor too and are ofte encouraged to distribute libre software.

Re: Linux's GPLv2 licence is routinely violated (2015)

#79

Earlier quoted context omitted.

The main problem with the interpretation that the GPL sets conditions for use of software is that it specifically claims it does not: > Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, Now there are some ways to try to do an end-run around this clause, it does raise some questions about how you g…

I find it difficult to imagine a situation in which you could use the library without copying the library. How would you obtain the code to execute it? Someone must distribute it to you, and/or you must copy it. Those are the governed actions. EDIT: note that the GPL broadly doesn't impose restrictions on interacting with a program over a network ("using" a server), which is why the AGPL exists.

It's ridiculous to expect that you can go after individual users for making single copies while not following your terms. You're going to sue someone for zero damages? For bruising your ego? Good luck with that.

For the same reason, an AGPL item could be enforceable, but only if the violator is juicy enough.

Re: Linux's GPLv2 licence is routinely violated (2015)

#80
post #23

> It is rather ironic that in some ways, one has more freedom in how one licences a kernel driver for Windows than the freedom one supposedly has in writing a kernel module for Linux. it is not ironic at all: freedom in Free Software is the freedom of USERS of the software to be able to see and modify the source. This is a freedom granted by the writers of the software, and copyleft means that this right is ensured t…

> the GPL and like licenses are the most permissive licenses This is a deliberately confusing statement. It's fine to prefer the GPL over MIT/BSD-style licenses, but the term "permissive" is well-understood in reference to non-copyleft licenses. Misapplying that term just muddles the issue unnecessarily. See also: [the wikipedia page on permissive licenses]( https://en.wikipedia.org/wiki/Permissive_software_license )…

Yeah, it's already confusing enough with "free" and "open" having too many competing meanings.
Post reply on HN