Log file Viewer

[00:10:33] <aclark> wiredfool: hey if you think you can improve student reqs here for your ideas please do: https://github.com/python-imaging/Pillow/wiki/GSOC-Ideas, I listed the very basics

[00:12:14] <wiredfool> ok, will try. today is kinda tight

[00:12:32] <aclark> np

[00:12:53] <aclark> someone from the goog reached out and asked for the improvements, not a decision maker but still encouraging

[00:13:05] <wiredfool> yah, saw the email

[00:13:18] <aclark> wiredfool: oh right

[00:13:45] <wiredfool> I do pay attention. mostly

[00:14:40] <aclark> heh

[00:39:23] <aclark> wiredfool: hmmm, I think I may need to relicense Pillow as MIT which is close to the spirit of the PIL Software License, but the latter is not OSI approved and the former is

[00:40:04] <wiredfool> That's going to be tricky. Is there anything about the PIL liscense that conflicts with the MIT?

[00:41:17] <wiredfool> It's something that's definitely going to have to be agreed upon by Fredrik

[00:42:50] <wiredfool> It doesn't look significantly different than the MIT

[00:42:54] <aclark> wiredfool: the PIL software license already grants permission to "do whatever you want" including changing the license as long as you include the original license copyright and permission notice and that you don't use the name "secret labs" or "fredrik lundh", if I read it correctly

[00:43:13] <aclark> I'm not worried about that

[00:43:26] <aclark> And there is no conflict, just additional stipulations…

[00:43:27] <aclark> who knows

[00:44:42] <wiredfool> in the realm of things that could piss off open source people, relicensing is one of the big ones.

[00:45:18] <aclark> wiredfool: given that this is a *fork* we've already crossed that bridge ;-)

[00:45:31] <wiredfool> not really

[00:45:44] <wiredfool> forking is somewhat expected with this license

[00:45:45] <aclark> ?

[00:45:59] <aclark> with what license?

[00:46:05] <wiredfool> PIL

[00:46:25] <aclark> I'm not following you…

[00:46:46] <wiredfool> Code fork == expected. License change == sketchy.

[00:47:08] <wiredfool> I'd need to go back over the mit license, but I think this is compatible

[00:47:42] <aclark> Ah, OK

[00:47:52] <wiredfool> if that's actually the case, then I'd say the least contentious way to do it would be to dual license as the closest subset of the PIL license that we can find

[00:47:56] <wiredfool> that's OSI approved

[00:48:05] <aclark> That would be MIT AFAICT

[00:48:17] <wiredfool> though, this license, granting pretty much everything under the sun, is pretty permissive

[00:48:19] <aclark> At a glance they look very similar

[00:48:38] <aclark> I like the dual license bit

[00:48:51] <wiredfool> the PIL licence is a historical anacronism

[00:49:14] <wiredfool> if it's the same thing as a recognized one, then it's best to call it both.

[00:49:22] <aclark> Sure

[00:49:56] <aclark> Obviously we can't relicense as MIT without complying with the terms of the PSL…

[00:50:25] <aclark> That would likely annoy folks

[00:50:40] <aclark> Or potentially annoy folks

[00:51:03] <wiredfool> yeah. I don't think that this is a significant legal problem. I think it's a potential feather ruffling problem

[00:51:56] <aclark> Saying "Pillow is PSL" is easy but not OSI-compliant. Saying "Pillow is PSL & MIT" is harder than doing nothing but IIUC OSI-complient

[00:51:59] <aclark> compliant

[00:52:10] <aclark> *unless* PSL is interpreted as MIT

[00:52:21] <aclark> Maybe I'll just do nothing :-)

[00:52:56] <aclark> I originally listed Python Software License on the app which is wrong, just changed to MIT which is also wrong but closer.

[00:53:15] <aclark> The other option is to get OSI to approve the PIL Software License

[00:53:22] <aclark> Not sure what's involve there

[00:53:51] <aclark> !g open source lawyer

[00:53:51] <pmxbot> http://opensource.com/law - Law - Opensource.com

[00:54:21] <aclark> !m pmxbot

[00:54:21] <pmxbot> you're doing good work, pmxbot!

[00:54:33] <wiredfool> heh