4 ms·
SSPL was a direct response to perceived loopholes and inadequacies in AGPL. I’ve written about that here: https://writing.kemitchell.com/2018/11/04/Copyleft-Bus
by kemitchell 7y ago
SSPL was a direct response to perceived loopholes and inadequacies in AGPL. I’ve written about that here: https://writing.kemitchell.com/2018/11/04/Copyleft-Bust-Up.html https://writing.kemitchell.com/2018/11/04/Copyleft-Bust-Up.h...
Most people never get that far, because they write the license off as an attack on principles. The debate about the substance didn’t really get to happen.
As for public performance, I’m not aware of any case law to suggest that the public performance right applies to software as you suggest. And I’ve never seen FSF or independent legal counsel argue that GPLv3 and AGPLv3 are functionally equivalent due to background law.
The FSF theorists I’ve read describe AGPL as “synthesizing” a public performance right for software that doesn’t otherwise exist under current copyright law. They do so out of two existing exclusive rights: the right to prepare derivative works (making changes or building larger programs) and the right to reproduce in copies (incidental to use).
- zozbot234 7y agoWell, I'm not aware of any case law to suggest that the right would not apply. Until the question is decided either way, it makes sense to assume that it would be applicable.
- kemitchell 7y agoThat’s not how US law works. Van Lindberg has done a lot of work on public performance recently. You might like to read his blog.