On Aug 5, 2009, at 16:59, Isaac Dupree wrote:
Simon Marlow wrote:
On 05/08/2009 14:58, Axel Simon wrote:
On Aug 5, 2009, at 15:44, Malcolm Wallace wrote:
I don't think it would be much of a problem to weaken the license of Gtk2Hs to a BSD license.
Don't forget you would need to obtain the consent of all contributors, whose patches are also under the LGPL.
True, but if I propose a discussion period on our mailing list during which people can object, then I think that would be sufficient.
I think strictly speaking you have to get explicit consent, rather than the absence of objection.
which GHC and the other BSD-components don't technically get, but it's strongly implied by submitting a patch. Similar for LGPL +exception (technically a contributor would be allowed to distribute a patch under just LGPL, or just GPL, or even GPL-2-only or GPL-3-only if they were silly). Socially, patches are generally assumed to be the same as the source license...
Can we get a list of all the Gtk2Hs code-contributors? (in which if a person only ever submitted less than a dozen lines of significant patches, it's probably not copyright significant) Also, does anyone here have an argument against trying to relicense? (and is it allowed, or is Gtk2Hs a "derivative work" of Gtk+ even when distributed separately?... I think it *ought* to be allowed in this particular circumstance, it wouldn't hardly be against the spirit of LGPL since Gtk2Hs is mainly simply a binding, but I'm not a lawyer :-)
When I re-implemented a binding for Gtk+, I chose LGPL because Manuel Chakravaty used LGPL for his Gtk+Hs binding. I did have a few requests to move to a BSD license from people who wanted to ship commercial Haskell applications. We once moved from version 2 of LGPL to version 3 at least for some or our code and we have tried to get consent from the relevant people before which was no big deal. Thus, I don't think it is a big problem unless people are in for an ideological license debate. Axel.