On Tue, Oct 06, 2009 at 01:22:46PM +1000, Ruediger Landmann wrote:
On 10/06/2009 09:47 AM, Jeffrey Fearn wrote:
Mikhail Gusarov wrote:
IANAL, but this can be specified in single file, like Common_Content/common/README: "all the data in this directory is under GFDL", but better check with your legal department.
Rudi is talking with the legal people about this, we expect a separate update message shortly.
Thanks Jeff :)
Red Hat legal has identified a number of ambiguities around the licenses involved: specifically, the relationship between the license of the package against the license of the text in the Common Content files, against the license of books that users produce that incorporates that Common Content.
One particular problem is that as things stand right now, if the text in Common Content is licensed under the GFDL, this means that any book that anybody builds in Publican that incorporates that text must also be licensed under the GFDL (or a compatible license). This, in turn, creates an immediate incompatibility in any brand package that loads a legal notice with a different license...
Legal's solution is that we include a note that explicitly spells out that whatever license appears between the <legalnotice> tags in the Legal_Notice.xml file applies only to the books into which it is pre-loaded, and not the text of the Legal Notice file itself. Furthermore, they suggest pretty much exactly what you suggested, Mikhail -- we find as permissive a license as possible for the Common Content files, and license them under that, separately from the rest of the contents of the package.
So far we've looked at the WTFPL[1], CC0[2], and the so-called GNU All-Permissive License[3].
We had to regretfully reject the WTFPL on the basis that some people might find it offensive. :( This is a real shame, because it basically stands for everything that we need the license on the Common Content files to stand for...
When we read the GNU "All-Permissive" License, it turned out to be not what it claims, since rather than being "all permissive", it requires re-users to leave the license in place. Relicensing is therefore as difficult as it is now.
Although CC0 is cumbersome (check out the full legal code! [4]), it seems to do what we need it to do. It's therefore the current favourite as license of choice for the Common Content files, unless anyone on the list knows of a similarly broad license with less legalese?
Cheers
Ruediger
[1] http://en.wikipedia.org/wiki/WTFPL
[2] http://creativecommons.org/publicdomain/zero/1.0/
[3] http://www.gnu.org/prep/maintain/html_node/License-Notices-for-Other-Files.h...
[4] http://creativecommons.org/publicdomain/zero/1.0/legalcode
I'm forwarding a copy of this to the fedora-legal-list -- Spot may be able to suggest something appropriate to cover the publican Common Content.
We need this license to be compatible with inclusion in works produced by Fedora Docs, and in works that incorporate content from the Fedora wiki, right? If CC0 can coexist peacefully in that role with the new CC licensing used in both those cases, it does seem like the best contender.
On 10/06/2009 02:01 PM, Paul W. Frields wrote:
We need this license to be compatible with inclusion in works produced by Fedora Docs, and in works that incorporate content from the Fedora wiki, right? If CC0 can coexist peacefully in that role with the new CC licensing used in both those cases, it does seem like the best contender.
Assuming that the CC licensing is "CC-BY-SA" (Attribution Share-Alike), right?
I've asked Red Hat Legal here, just to make sure my instincts are right.
~spot
On Tue, Oct 06, 2009 at 02:14:12PM -0400, Tom spot Callaway wrote:
On 10/06/2009 02:01 PM, Paul W. Frields wrote:
We need this license to be compatible with inclusion in works produced by Fedora Docs, and in works that incorporate content from the Fedora wiki, right? If CC0 can coexist peacefully in that role with the new CC licensing used in both those cases, it does seem like the best contender.
Assuming that the CC licensing is "CC-BY-SA" (Attribution Share-Alike), right?
I've asked Red Hat Legal here, just to make sure my instincts are right.
Correct, the Docs project is switching to CC BY-SA 3.0 Unported: https://www.redhat.com/archives/fedora-announce-list/2009-October/msg00001.h...
On 10/06/2009 02:46 PM, Paul W. Frields wrote:
On Tue, Oct 06, 2009 at 02:14:12PM -0400, Tom spot Callaway wrote:
On 10/06/2009 02:01 PM, Paul W. Frields wrote:
We need this license to be compatible with inclusion in works produced by Fedora Docs, and in works that incorporate content from the Fedora wiki, right? If CC0 can coexist peacefully in that role with the new CC licensing used in both those cases, it does seem like the best contender.
Assuming that the CC licensing is "CC-BY-SA" (Attribution Share-Alike), right?
I've asked Red Hat Legal here, just to make sure my instincts are right.
Correct, the Docs project is switching to CC BY-SA 3.0 Unported: https://www.redhat.com/archives/fedora-announce-list/2009-October/msg00001.h...
Yeah, neither I nor Red Hat Legal sees any problem with using the CC-Zero license in conjunction with CC-BY-SA, as described previously in this thread.
~spot
On Tue, 6 Oct 2009 14:01:52 -0400 "Paul W. Frields" stickster@gmail.com wrote:
On Tue, Oct 06, 2009 at 01:22:46PM +1000, Ruediger Landmann wrote:
So far we've looked at the WTFPL[1], CC0[2], and the so-called GNU All-Permissive License[3].
We had to regretfully reject the WTFPL on the basis that some people might find it offensive. :( This is a real shame, because it basically stands for everything that we need the license on the Common Content files to stand for...
Agreed, this is unfortunate. :)
When we read the GNU "All-Permissive" License, it turned out to be not what it claims, since rather than being "all permissive", it requires re-users to leave the license in place. Relicensing is therefore as difficult as it is now.
I think this is not a correct interpretation, as the mere fact that a license requires preservation of a licensing notice doesn't mean that it has a copyleft effect; this is well established in FOSS tradition as evidenced by BSD and MIT and Apache (etc.) licensing. Nevertheless, it is true that CC-0 requires no preservation of the CC-0 text; indeed it logically couldn't because in CC-0 the copyright holder is at least attempting to abandon all ability to enforce copyright on the work.
No objection to CC-0 though, which in the end is probably no worse than and probably better than traditional simple public domain dedications.
- RF
"RF" == Richard Fontana rfontana@redhat.com writes:
[Offensiveness of WTFPL text] RF> Agreed, this is unfortunate. :)
Might I suggest simply modifying the offensive language? I know license proliferation is bad, but if the result is legally equivalent and serves the necessary purpose then I don't see any reason not to just do it.
- J<
Oh, and don't please don't ever CC a closed list. I got a nice auto-reject message for my trouble.
- J<
On Wed, 07 Oct 2009 19:06:17 -0500 Jason L Tibbitts III tibbs@math.uh.edu wrote:
Oh, and don't please don't ever CC a closed list. I got a nice auto-reject message for my trouble.
As did I.
[removed publican-list from cc]
On Wed, 07 Oct 2009 18:44:54 -0500 Jason L Tibbitts III tibbs@math.uh.edu wrote:
[re: WTFPL]
Might I suggest simply modifying the offensive language? I know license proliferation is bad, but if the result is legally equivalent and serves the necessary purpose then I don't see any reason not to just do it.
Certainly Fedora should be prepared to accept licenses that are equivalent to the WTFPL just as it accepts the WTFPL. Also there's no question that the WTFPL has earned an important place in FOSS culture despite being rarely used. However, this is a situation where Red Hat is acting as copyright holder and/or outbound licensor. For *Red Hat code* we generally avoid licenses that are not commonly used (i.e., that aren't associated with substantial project communities), and we generally avoid licenses that are modified versions of other licenses, unless the modified version is itself a commonly-used license. Of course the mere fact that a license is popular doesn't mean it's better, but we see a lot of value in promoting license standardization.
For those reasons (and not any sense of primness), we wouldn't encourage our developers to apply the WTFPL to Red Hat-copyrighted code, and we'd certainly oppose applying some sanitized WTFPL derivative to Red Hat-copyrighted code. (On the other hand we encourage our developers to make licensing decisions that are informed by concerns about their users.)
- RF
-
Jason L Tibbitts III -
Paul W. Frields -
Richard Fontana -
Tom Callaway