@jalcine I really like this post, thanks for sharing -- reading through this, this is my first time reading the Parity license text.
Clause 1 I think fails the DFSG, via the "desert island test". If I make any modification, this license compels me to contribute that, which is not possible in e.g. an internet-disconnected environment. (For comparison, GPL differs in that provision only kicks in on distribution.)
@jalcine Clause 3 is also a little concerning mostly because it's so vague ("develop, deploy, monitor, or run" aren't defined) and far-reaching.
I think that would also violate DFSG #9 because this contaminates other software run by (how does that work? Like a bootloader or linker or ELF interpreter?) something Parity-licensed, which would make its inclusion in a distribution very difficult.
@ehashman @jalcine Parity 7.0 is in development, and this might be good feedback, if it's stopping major distros from being -able- to distribute software under it. Adjusting clause 1 to take into account the desert island test seems like a good suggestion to make. As far as clause 3 goes -- I don't understand why the AGPL might pass DSFG 9, but not Parity, on this front.
More broadly, the reason GPL software doesn't "contaminate" other software in the same way is because the virality clause only kicks in when making a combined/derivative work; simply using GPL software to run other software (e.g. using an Ansible playbook to run a remote command) does not require that other software to also be GPL licensed
@ehashman @jalcine The point of clause 3 is to apply copyleft to devtools users, which does include things like compilers and interpreters. By the way, if you want an older writeup of the parity license, there's a post that goes line by line and explains the implications and intent: https://blog.licensezero.com/2018/11/24/parity-5.0.0.html
@zkat @jalcine there's a long implied rule in FOSS licensing that restrictions invoked on "mere use" make something not FOSS. I don't necessarily buy into an absolute reading of that, but even the GPL requires the creation of a derivative work for the license to kick in to affect other software.
I see what Parity is trying to accomplish to advance the ideals of copyleft, but since clause 3 is broad enough to affect anything the software interacts with, this might in practice require writing an entire separate OS, kernel, programming language, etc. to properly comply with the license terms. FOSS concerns aside, I don't think that's particularly practical; it cuts off Parity devs from the rest of the ecosystem.
In the Clojure ecosystem (for example), Rich's choice to license the language EPLv1, which is GPL-incompatible, has limited the ability of devs to produce copyleft software written in Clojure. These systems are complex 😵
@zkat @jalcine (it's hard for me to discuss in rapid-fire form; these are the sorts of things I spend hours thinking about and often don't even come out of with any new insights... and then a single conversation with a very smart lawyer like Pam Chestek might wipe out any progress I figured I made 😅)
Licenses are a little like filesystems: They are foundational. One may find quite a lot of work crumbles when an unanticipated behavior manifests itself. Thus, many will be loathe to adopt them until their strengths and weaknesses are well established and understood.
Mentioning Pam Chestek is apt: If we're to avoid another bad breakout of engineer's disease, we probably need to bring them in soonest.
It's frustrating that the GPL is good but the organization is terrible - it feels like we need to fork the license and just replace G/GNU with some organization that doesn't suck.
It also makes it hard to continue some good projects that use the GPL. I know little of the GNOME Foundation, but to use them as an example their Outreachy project seems to indicate they're basically decent or at least have some level of the ethics the FSF lacks. And GNOME is GPL.
@jalcine If people are going to have "heros" how about picking better ones. Miguel de Icaza, for example...
@tfb didn't he leave f/loss?
@jalcine How has he left? He works at MS now ... And has been active in getting C#, .Net Core, Xamarin open sourced, and developed in the open. The he no longer is in the GNOME Foundation or .Net Foundation boards is a good thing. He's made space for others, unlike some Dicks.
I've been pointing to him as the sort of leadership we should collectively want since the mid-1990's. I stand by that.
@tfb I didn't know that - thank you for educating me. I need to read up on him - a lot of my knowledge is heavy Debian + KDE stuff
I thought you might've been throwing shade for him being at MS. From Midnight Commander, to being a big proponent of Linux in the Third World, to Gnome, to Evolution, to Mono, to Xamarin and F#, to OSS C# and .Net Core, he's had a really impressive FLOSS career (and career by any measure).
The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!