Talk:GNU General Public License/Archive 6

From Wikipedia, the free encyclopedia
Jump to: navigation, search
← Archive 5 Archive 6 Archive 7 →

Examples of output that is derivative?

Note that the copyleft applies only to the software and not to its output (unless that output is itself a derivative work of the program); for example, a public web portal running a modified derivative of a GPL'ed content management system is not required to distribute its changes to the underlying software.

Could examples be given for GPL software whose output is a derivative? Would a bitmap rendering from a GPL'ed SVG text file be a derivative work? What about a program that draws a circle, and an image of the circle drawn? Program source code that paints a free texture map onto a flat surface and the output of such code when interpreted? 84user (talk) 03:01, 11 December 2008 (UTC)

Parsers generated by GNU bison are examples of outputs that are derivative works. The output of a bison run is a set of tables based on the input syntax grammar (which is of course not derived from bison), and a program that uses the tables (which is a part of bison). As the bison documentation says:
"The output of the Bison utility—the Bison parser file—contains a verbatim copy of a sizable piece of Bison, which is the code for the parser's implementation. (The actions from your grammar are inserted into this implementation at one point, but most of the rest of the implementation is not changed.) When we applied the GPL terms to the skeleton code for the parser's implementation, the effect was to restrict the use of Bison output to free software."
RossPatterson (talk) 04:11, 11 December 2008 (UTC)

Thank you, I have added GNU Bison as a good counter example. I am still interested in any sources that better define the boundary between derivative and non-derivative . Is it only if the output contains code, and what kind of code? Maybe I will ask at the Reference Desk. 84user (talk) 21:20, 11 December 2008 (UTC)

If the output includes (sizeable) parts of the work, then the output is a derivative. E.g. drawing programs may offer artwork, which can be used only to the extent explicitely permitted. Free software is in no special position, so a general legal discussion would do as source. --LPfi (talk) 10:01, 23 February 2011 (UTC)

Text related to GPL

I'd like to see a clear statement about what people can do with text such as that on a wiki about a program. Example Wikia:Freeciv, which is a wiki devoted to the GPL game Freeciv and which says on its edit page that all contributions are under the GPL. Under what conditions, if any, could some of that wiki be legally copied to a CC-BY-SA wiki? Robin Patterson (talk) 03:50, 22 February 2011 (UTC)

GPL (and GDFL also) is incompatible with CC-BY-SA so neither text under GPL can be relicenced to CC-BY-SA, nor the other way.1exec1 (talk) 10:33, 22 February 2011 (UTC)
Or, more correctly: they can be relicensed only by the right owners (usually the authors). Quoting minor parts of the discussion is usually permitted as fair use, and referencing it is unproblematic. If you want to copy larger parts of the discussion, you should create a separate GPL licensed section in (or outside) the wiki. --LPfi (talk) 10:01, 23 February 2011 (UTC)

I still cannot appreciate the Open Source Concept

I have gone through enough information about the Open Source Initiative and the GPL licenses. It is more confusing than clarifying, the original authors intent. Since nobody would be able to make software products that could sell in the market using the Open Source as a base, to what end is this reaching out? Is this initiative not defeating the basic business sense? Can somebody kindly throw some light? —Preceding unsigned comment added by 59.92.193.158 (talk) 17:41, 28 January 2009 (UTC)

First of all, there are other reasons to use open source besides those involving business. For some, using open source is partially based on a principle of public participation and contribution. Needless to say, there are many reasons developers choose to participate in or create open source projects. As far as business is concerned, some companies profit based on a model of selling services and support for their software. If a company develops open source software, there are sometimes clients who are willing to pay for technical support, expertise and customisation from the original developer. It's not too surprising you didn't get a sense of the debate surrounding open source software and business in this particular article, since the article isn't focused around those specific issues. Business models for open source software would probably be a better article to look at for more information about corporations and open source software. --Xaliqen (talk) 12:00, 6 March 2010 (UTC)

misspeled word license as licence

in the criticism section license is spelled as licence many times Jnw222 (talk) 23:39, 18 March 2009 (UTC)

well correct it! KenThomas (talk) 19:26, 17 October 2009 (UTC)

Note that in British English, "licence" is the noun and "to license" is the verb. — Preceding unsigned comment added by 147.188.194.245 (talk) 16:02, 13 September 2012 (UTC)

Section numbering style

When referring to sections in the GPL, it is important to note that the sections start at 0 in the hacker manner. However, perhaps "section 0" should be the "first section", while "section 5" is the "sixth session". I don't care what we choose, but it'd be better if we were consistent. How about we avoid "zero section" and go with the serial format -- "first", "second".

Another option is to avoid mentioning numbers all together, since this seems to be a nomenclature that the FSF and other GPL specialists only use. We could just use the section titles. --Ashawley (talk) 16:55, 27 July 2009 (UTC)

market price

Since we mention market price in the criticism section, it would be better if we had a section on the commercialization aspect for gpl in the article. Software has several different commercialization aspects and only mentioning the one where "company X produce product Y and tries to sell it to the general public" is very limited. —Preceding unsigned comment added by Belorn (talkcontribs) 10:22, 27 September 2009

That whole discussion smacks of unacceptable Original Research into economics. There are plenty of companies selling software that is GPLed (e.g. Oracle/Sun/MySQL), plus it's how the FSF funded itself for years. {{Citation needed}}. RossPatterson (talk) 16:39, 27 September 2009 (UTC)

A survey that shows massive GPL usage

I've stumbled upon this survey:

It's from 2001. Section 5.1.2 has a graph showing licences used and puts GPL at what looks like about 90%. That's quite high. Is there something that skews their sample group or is that the way things were back in 2001? Gronky (talk) 02:19, 30 September 2009 (UTC)

NPOV dispute

Initiating NPOV discussion. This page seems to me deeply embedded with a POV aligned with the Open Source/Free Software community and the non-neutral assertions of the GPL itself; it is based almost entirely on admitting as fact, or neutral, statements and opinions that would be seen as true by members of this movement, but are not clear from the POV of non-members of this community. There is a tendency to hero-worship of the GPL and Stallman. Othertimes the article either grossly oversimplifies complex matters to the point of error, or is very close to being unreferenced bunk or the repetition of common but unsubstantiated myths.

For instance: under 'Licensing and contractual issues': the article states: "Those who do not agree to the GPL's terms and conditions do not have permission, under copyright law, to copy or distribute GPL licensed software or derivative works." This is unreferenced; it is also grossly ignorant of the details of "copyright law." In a jurisdiction where some sort of copyright law applies to a work allegedly under the GPL, for instance, "copyright law" may authorize use under fair use provisions or other 'exceptions;' it may not apply to the GPLed work at all, because it is not copyrightable (does not constitute a work subject to local copyright); and the reference to "derivative works"(*) is likely deceptive insomuch as no clear definition of 'derivative' is proferred, and the general article seems to thing that this is a far wider category than supposed. (*) Where linked elsewhere, the definition of "derivative work" on WikiPedia is rightly marked as dealing primary with North American Civil Law jurisdictions-- if that.

The section "Compatibility and Multi-licensing" is unsourced, and largely a repetition of the (non-neutral) assertions of the GPL. A phrase such as "Also see the list of FSF approved software licenses for examples of compatible and incompatible licenses" is using a single, non-neutral source, ie, essentially repeating a non-neutral source, and its assertion that there are objective "compatible and incompatible licences, as if it's assertion were truth (in the sense of "gospel"). "Criticism" appears to take a neutral form, but chooses its example to make the opposition look stupid, really says "MicroSoft asserted this bunk, but fortunately, the good side came back ... and MS...".

Paragraph 4 of this section, beginning "While the GPL does allow..." (and continuing) is unsourced/unreferenced/unsubstantiated, and employs non-neutral language which essentially repeats the non-independent assertions of the GPL. Better phrasing might be, "The GPL asserts that licencees (or: adherents) may not...". Regardless, "the GPL does not allow" is neither neutral or accurate language. The GPL is not a live actor which can prevent "what it does not allow;" it is a written document with questionable legal, contractual or other validity in the situations described, particularly when these situations occur in jurisdictions outside the United States.

"Linking and derived works" is largely unsourced and trades mainly in the obscure local myths of the FS/OS community, for instance, >>A key dispute related to the GPL is whether or not non-GPL software can be dynamically linked to GPL libraries. The GPL is clear in requiring that all derivative works of code under the GPL must themselves be under the GPL. While it is understood that static linking produces derivative works, it is not clear whether an executable that dynamically links to a GPL code should be considered a derivative work (see Weak Copyleft). << This section is essentially a result of original (like "participant observer") research into the OS/FS community, and, at best, forwards statements that are highly suspect from an independent, factual point-of-view.

In "The GPL in court," first paragraph engages in an attempted cherry-picking of facts to show a "win" for validity of the GPL: >>At the hearing, Judge Saris "saw no reason" that the GPL would not be enforceable.[34]<< This attempts to give the impression that the GPL was in someway positively affirmed by this comment on the part of Judge Saris; what in fact (seems to have occurred), was that on a preliminary hearing in the matter, the Judge expressed that she saw no immediate and obvious barriers to enforcement of the GPL in this particular case; in a legal context, this non-negative allowance for the matter to proceed is in no way equivalent to a positive judgment that the matter might proceed. —Preceding unsigned comment added by KenThomas (talkcontribs) 16:26, 18 October 2009 (UTC)

And so on-- most paragraphs in the article can be picked apart, and are a compilation of non-neutral, mythical, unsourced material, or sourced material being used in a non-neutral manner in order to support ideological statements and/or irrational beliefs. KenThomas (talk) 19:22, 17 October 2009 (UTC)

Hi Ken. Thanks for this review. I don't agree with all your points, but for the points I don't agree with, I accept that if you came to these conclusions, then the article needs to be clearer.
I came here to add something about GPL exceptions and I don't see a good place to add that info, so a bit of a restructure is needed anyway. I'll read from your comments when reviewing it. Gronky (talk) 23:34, 23 October 2009 (UTC)
...I don't have the time to do the review necessary afterall :-( Gronky (talk) 01:33, 2 December 2009 (UTC)
Ken, you're absolutely right. There are peacock terms all over this article. In the second paragraph, it says "...the GPL is a powerful public license...." If that's not a violation of WP:NPOV, I don't know what is. —Preceding unsigned comment added by 67.180.128.108 (talk) 00:25, 22 March 2010 (UTC)

Can I get a GPLv2 project, modify it, and release it under GPLv3?

That's it! I want to get a project under GPLv2 which the copyright owners don't want to change it to GPLv3. Can I modify this project and release it under GPLv3 (with a new name of course)?

Also, there are many projects who says "GPLv2 or, at your own choice, any later version" while others says "exclusively GPLv2". Can i do the same with these projects too? I think so, after all, if I get an "exclusive GPLv2 project", modify it, give it a new name and release it under GPLv2, this project's owner is automatically me, and if it's "my" project, I can choose to move it from GPLv2 to GPLv3, can't I? —Preceding unsigned comment added by 189.101.77.22 (talk) 03:45, 15 February 2010 (UTC)

Copyright law reserves all rights (including the right to distribute derivative works) to the initial author. When that author releases it under GPLv2, he/she conditionally grants you the right to distribute derivative works. The condition is that you have to make your work available under the same license. Releasing your derivative work under GPLv3 does not satisfy that condition because it is not the same license as GPLv2. Thus, you have no legal right to distribute a derivative work, unless you obtain that right through some other mechanism (such as the author explicitly saying "or any later version"). You are the author of the parts you wrote, so you can release those parts under any terms you wish. To become the legal owner of the parts you didn't write, you need a contractual transfer of copyright ownership. The GPL is not a contractual transfer of copyright ownership. Generally, open source developers are anxious to have derivative works. Odds are good they will be flexible if you contact them. Maybe they'll add the "or any later version" clause just for you.--128.187.80.2 (talk) 18:29, 23 June 2010 (UTC)

Renaming Linux to GNU/Linux

All places where Linux has been used to mean the entire OS has been correctly renamed to GNU/Linux. Linux is the kernel while GNU along with kernel Linux is actually what we incorrectly term Linux. See gnu.org for more details. If anyone has any query, please leave them here before editing. —Preceding unsigned comment added by Maitraya (talkcontribs) 08:36, 2 March 2010 (UTC)

v3's Additional Terms

The article only mentions briefly this new concept (section 7 of v3). I think it's a very important addition. Attribution (7b), for example, is a revolutionary step for GPL. Is there any source out there that gives real or simulated examples for v3 licenses that have additional terms?

Also, what do those terms even worth if

When you convey a copy of a covered work, you may at your option remove any additional permissions from that copy

-79.176.200.211 (talk) 00:16, 16 April 2010 (UTC)

GPL and the public domain?

The article doesn't mention how GPL and the public domain fit together. For example, can public domain code exist within a GPL project? Is it legal to create a GPL fork of public domain code? pgr94 (talk) 19:37, 26 May 2010 (UTC)

According to http://www.gnu.org/licenses/license-list.html

Public Domain: Being in the public domain is not a license; rather, it means the material is not copyrighted and no license is needed. Practically speaking, though, if a work is in the public domain, it might as well have an all-permissive non-copyleft free software license. Public domain material is compatible with the GNU GPL.

Is that worth adding to the article?
Also, is it possible to create a GPL fork of public domain material? A GPL fork would require subsequent contributions to be shared, while the public domain version would not place restrictions on the contributions. pgr94 (talk) 21:13, 26 May 2010 (UTC)

Misleading initial explanation

The very start of this article says "...which means that it requires derived works to be available under the same license terms". I propose this should be changed to "...which means that it only grants distribution rights if derivative works are made available under the same license terms". As a license, the GPL has no power to make demands. It can only grant rights that are otherwise limited by Copyright law. Granting these rights conditioned upon making derivative works available is not the same as "requiring derived works to be available...".--128.187.80.2 (talk) 18:13, 23 June 2010 (UTC)

Proposed rename to "GPL"

The following discussion is an archived discussion of the proposal. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section.

No consensus to move. Vegaswikian (talk) 00:35, 29 August 2010 (UTC)

GNU General Public LicenseGPLWP:COMMONNAME says it all. Almost exclusively, people refer to these licenses as simply "GPL", to the point that many people don't actually know what the acronym means. -- intgr [talk] 12:18, 22 August 2010 (UTC)

  • Comment - I have no clear opinion but would like to share some thoughts. While GPL may be the common name for this topic - that acronym is still used by others. Granted this is by far the most popular use of that acronym - however, actually titling the article with the full-name may be preferable as it communicates unambiguously what the article is about. Naming it GPL on the other hand could invite greater confusion on the part of people searching for articles found on GPL (disambiguation). Overall I understand the points made in favor of the common name but feel that the current title's accuracy and specificity in relating the contents of this article have value too. Solid State Survivor (talk) 04:10, 23 August 2010 (UTC)
  • I object. The full name is commonly used in many places, including the standard GPL boilerplate, legal websites, and FOSS project descriptions. Heck, even PNG and PDF are at their common names, and the average reader has much less incentive to know what they stand for than it is the case for the GPL. - Sikon (talk) 04:18, 23 August 2010 (UTC)
    How about "GNU GPL"? -- intgr [talk] 07:55, 23 August 2010 (UTC)
  • Oppose. What positive effects do you see in the first place? Take into account the fact that GPL already redirects to this article.1exec1 (talk) 09:31, 23 August 2010 (UTC)
    The primary problem is people linking to this article with the text "GNU General Public License"; First, it's not as descriptive/recognizable as just stating "GPL" would be, and second it's unwieldy — especially on infoboxes where it doesn't fit on a single line. Even if these links were pointing to the GPL redirect, there's always the "anti-redirect crew" that would undo that. And I really don't like piping if "GPL" would be a so much more obvious target. Also, WP:COMMONNAME is policy and has good reasons and practice behind it. I realize now that I should have started my nomination with these justifications, sorry about that. -- intgr [talk] 20:02, 23 August 2010 (UTC)
  • Oppose. WP:COMMONNAME takes on less importance when it comes to simple acronyms that haven't fully supplanted the original name. Most reliable sources that discuss the GPL will use the full title on first usage anyway. Powers T 20:31, 24 August 2010 (UTC)
  • Oppose GPL is not so engained in culture to be commonly know by the general pubic. Where possible its almost always preferable not to employ acronyms. --Labattblueboy (talk) 01:05, 25 August 2010 (UTC)
The above discussion is preserved as an archive of the proposal. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.

GPL and sourceless programs

Hypothetically, would the creator of a program that has no source code (for example a program written entirely in handcoded machine code, or a program created by an evolutionary algorithm) be able to release their program under the GPL? Horselover Frost (talk · edits) 01:06, 1 January 2011 (UTC)

Of course. The source for an machine code program is the machine code. --Gwern (contribs) 01:18 12 January 2011 (GMT)

v3 minor fixes' proposals

The section Version 3 mumbles about Version 3:

  • it should (maybe in first para) claim that Version 3 is current and published in June 29, 2007,
  • it should make a summary of how the version is different from v2,
  • the comment (which is in fact a criticism from the Linux kernel developers) in the last para is in fact a to Draft 2, and it seems to me that the published Version 3 builds mainly on Draft 4.

Rursus dixit. (mbork3!) 09:53, 1 April 2011 (UTC)

I didn't read carefully. So I'll modify/merge the first-two proposals:
  • the entire first para should be about the final published Version 3 of June 29, 2007, and it should summarize how it differs from v2,
Rursus dixit. (mbork3!) 10:00, 1 April 2011 (UTC)

"Unfriendly" comment

...and a extremely unfriendly thing to say. I don't think that this contributes anything useful to the article. I'm not sure what WP:BLP has to do with anything in this situation, as removing that does not take the information out of context. Stallman's opinion that the metaphor of a virus is wrong, this is important to the article. That he thinks it unfriendly is not. His factual argument against the "viral" statement is important, his feelings on the matter are not encyclopedic, and as it's not a quote, I don't see any reason to keep it. - SudoGhost 07:17, 15 May 2011 (UTC)

In another source he called it "harsh". So IMHO it is OK to mention that he thinks it is a wrong use of negative connotations. There is another article on the subject, "Viral Code and Vaccination" by Robert J. Chassell. --AVRS (talk) 10:43, 15 May 2011 (UTC)
The usage of "harsh" or "unfriendly" is for me equivalent in this meaning so I dont have any preference. I picked unfriendly because it was used in the first cited source. On the question if its important for the article, for me what is important is to have a correct representation of the cited person opinion. Editing and removing parts of what a living person say, should be done very carefully, which WP:BLP is very clear on. Thus atm I op to keep the hole opinion as before (or using the word harsh instead of unfriendly). As an alternative, maybe it would be better to simply write a summery of the Viral license article as it relate to GPL, and have a link to the main article. This way we could skip the hole current setup of what Mundie said and then what Stallmans response was, and replace it with an objective text of the subject matter.Belorn (talk) 23:24, 15 May 2011 (UTC)
This would seem a good idea to me. Currently one would believe it's primarily Microsoft who has used the description yet AFAIK, and as noted by the viral licence article, BSD and similar licence proponents commonly describe the GPL in that way and I believe have been doing it longer then MS. For them it's an obvious annoyance since (2 clause) BSD code can be (and sometimes is) licenced under the GPL thereby preventing them from using it further. Nil Einne (talk) 12:16, 8 July 2011 (UTC)

v3 impacts?

I think would be nice to highlight some of the effects of the v3 changes. It seems as if Apple will not distribute any GPLv3 software. Being Apple, they won't comment of course, but it seems pretty clear. Major OS components like GCC, Samba, and Bash have not been updated in years. I only recently learned that this was due to these projects having switched to GPLv3. Apple has even gone so far as to write their own GCC-compatible compiler (Clang) and SMB networking layer. Clang is open-source (because it is based on the non-GPL LLVM), but Apple's SMB implementation is proprietary. Apparently Solaris doesn't include GCC or Samba either.

It seems like this should be a bigger issue that it is. With the release of MacOS X Lion, end users are now feeling the impact because their systems now work differently with Windows services or, in some cases, don't work at all. Is this only an issue with Mac users or are there other impacts as well? Thanks.

Jdaniel905 (talk) 02:01, 19 September 2011 (UTC)

This is a great initiative, but has Apple distribute any GPL software at all in the last 10 years? There is some stories about how Steve Jobs tried to make a proprietary add-on to GCC a 15-20 years ago, got legal resistance from FSF, and has from that point refrained from ever using GPL software in Apple products ever since. While I know apple has made several statement about their dislike of GPLv3, I dont know if they actually done any change to what software they will distribute.Belorn (talk) 03:59, 19 September 2011 (UTC)
Sure, here is the current list of open-source software that ships with MacOS X 10.7 Lion: http://www.opensource.apple.com/release/mac-os-x-1071/. As you can see, there are a fair number of GPL packages, but they are all GPLv2. If a package, such as Bash, for example, has switched to GPLv3 at some point, Apple stays on the last GPLv2 version. That proprietary add-on is actually Objective-C I think - the heart of MacOSX and iOS. After more research, this seems to be a a mess that has been brewing for some time. The GPLv3 and Lion's new SMB runtime seem to be just the culmination of a feud that has been running for 12 years. I honestly never knew about any of that. Perhaps it does merit an edit to the main page. Jdaniel905 (talk) 18:51, 19 September 2011 (UTC)
Thanks for the links. Looks like I only had a part of the full story. Definitely merit a addition to the article. Belorn (talk) 20:19, 19 September 2011 (UTC)