@dos @dpk However, there's another joke in here about the *mutability* of this license choice: what happens if a license is *removed* from the list? By saying "currently" as opposed to "which appeared at any time", it's creating a challenge: it's not an append-only set, it could end up in strange places if something got removed
@cwebber I recently worked with @n0toose and @gedankenstuecke on reformulating Codeberg’s licence policy to avoid dependency on the FSF or OSI lists with exactly this problem
in mind (considering especially the OSI’s recent bizarro moves in the slop machine space, and I bet when Stallman kicks the bucket the FSF will also start doing some kind of weird shit too)
@cwebber @dpk Courts operate not just on the license's letter, but also its spirit, so taking over FSF to publish a permissive GPLv4 wouldn't necessarily be as effective as it may seem at first glance - even if it would still cause plenty of chaos. In contrast, the spirit of this joke license is pretty much "an overly complex way to say it's MIT-0/0BSD" - that is, unless a court decides otherwise, judging from the whole context around a particular case and particular people involved ;)
@dos @dpk And if you don't know, here's from memory what happened:
Wikipedia was licensed under the GFDL, and that was before CC BY-SA was available as the world's most popular copyleft license for cultural works. How to relicense with so many contributors?
So... Creative Commons, Wikimedia, and the Free Software Foundation collaborated on adding a new version of the GFDL that allowed for relicensing to CC BY-SA if it were done within a short time window to allow Wikipedia to do it
@dos @dpk Source: from memory.
But a fun fact about me: before I worked on decentralized network tech stuff (and partly during it), I used to work on the tech team (at one point tech lead) at Creative Commons, and went deep in the weeds on many FOSS / free cultural licensing things
I have much too cursed knowledge about these things because of it
@cwebber ahaha that's amazing
As someone who was present at the drafting of the GFDL, and someone who thinks CC-BY-SA is not actually a #copyleft license, I think Wikipedia picked the correct evil of two lessers.
GFDL was a peace treaty between RMS & Tim O'Reilly written in the form of a license.
CC-BY-SA is copyleft designed by libertarians.
True copyleft must allow reproducibility from first principles. CC-BY-SA doesn't.
@bkuhn @dos @dpk I think copyleft with source-requirement is good, though I'm not convinced it's viable for all places where CC BY-SA is useful for. For example, distributing raw film footage and editing files for some videos isn't really necessary and may be an unnecessary burden if you think about much of the video content out there today, so having a weaker version of "copyleft" is sensible to me. I would be fine with a different term for it though.
In the case of #Wikipedia, though, printed copies should always offer &/or be accompanied with electronic copies. Wikipedia should really be under copyleft and it's not.
After all, paper printout is but a rudimentary form of DRM.
I'm not saying CC-BY-SA is useless and shouldn't exist, I'm saying there *is* no #copyleft in the CC license group. CC even used to encourage photographers license scaled versions CC-BY-SA & keep high quality images proprietary.
Cc: @cwebber @wikipedia @dos @dpk
Remember, @lanodan, that an exclusive rights holder (e.g., a copyright holder) always has the permission to issue any material under any license (or many different licenses) or not.
This is what allows “proprietary relicensing” to happen at all.
That sad, the other part of your analysis **is** correct: CC-BY-SA permits me to take a high-res image you give me, modify it, reduce it to low res, and refuse to share the modified high-res image with anyone.
I wrote:
> “After all, paper printout is but a rudimentary form of DRM.”
I also admit that 23 years ago, I tried to argue otherwise re: the above statement. @novalis was the one who called me on that. Thanks again for that.
@bkuhn
Isn't "derivatives of free works remain free" an important copyleft principle that CC-BY-SA protects?
Source-available for non-software works would be good, obviously, but I think the justification is different than for software. Most of the time, if you have a copy of an image and a CC-BY-SA license, you can make whatever you want - very false for software.
IMO the more compelling case for "source for published works" is archival, not freedom of use.
@cwebber @wikipedia @dos @dpk
Good point, @danielittlewood. At first, I was about to reply that #CC-BY-SA is a “weak #copyleft” but that's not quite right (as it's nothing like traditional weak copylefts like #LGPL).
I think I'm going to coin a new term right here, right now: “CC-BY-SA is an *incomplete* copyleft”.
I agree it;s got the “share & share alike” clause as it says on the tin. However, any license that says a paper-format book need not share editable e-copies ain't no copyleft!
Cc: @cwebber @wikipedia @dos @dpk
I deeply hope you're right that Courts care about the spirit of the license. I have yet to see it in practice, and curious where you saw a Court do this.
This is an important issue we're exploring in the #Vizio case. I hope you'll follow it. https://sfc.ngo/vizio
@cwebber @dos @dpk section 14 contains:
"Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns."
"Later license versions may give you additional or different permissions. However, no additional obligations are imposed on any author or copyright holder as a result of your choosing to follow a later version."
That's not much wiggle room without risking legal uncertainty. 🤷♀️ I'm ok with that.
@kpcyrd @cwebber @dpk Even without these clauses the outcome would likely be similar in western jurisdictions if a hostile takeover was attempted, but there's of course no way to know for sure without actually having a court decide (at which point it's already costing you time, effort and money), so every bit of reassurance/deterrence helps anyway 😁
@dos @dpk well the joke is that it's as permissive/lax of a license as the most permissive/lax license in effect, but most lax licenses aren't lax because they can be relicensed, they still ordinarily operate under the terms of the original license, unless there's a specific clause for relicensing
But GPLv3 "or later" has the challenge where you have to trust the FSF, and that does have that challenge (and is meant to evoke thoughts about that)
License upgrade stewardship is a tough problem.