ive come to think that if you open source your code, u should only use MIT. ie. make it a gift. because once your code is just out there in the open, you cant pretend you still control it. suddenly you're dependent upon lawyers. thats icky. i mean anyone could be using this guy's code to do something against the license, and dude would never know. well, if you give a shit, maybe don't leave your code sitting in the open on a microsoft server? either set it free, or don't.
That's a really bad take. GPL or AGPL License is not because someone doesn't want their code to be out, it's because we believe in fair contributions, ownership and building the projects in a community. Open Source with GPL is a Tit for Tat system (you either make things better for everybody, or you don't), not 'get altruism for free' system.
>Open Source with GPL is a Tit for Tat system (you either make things better for everybody, or you don't)
The GPL doesn't force you to make software better for everyone. People and communities are free to steal the project, make it better, and keep it all to themselves.
> The GPL doesn't force you to make software better for everyone.
It forces you to contribute. Read what GPLv3 states:
- Include a copy of the full license text
- State all significant changes made to the original software
- Make available the original source code when you distribute any binaries based on the licensed work
- Include a copy of the original copyright notice
Watch Linus' interview with the Intel CEO where he talks about what open source is. He clearly mentions that it's a system where personal reasons end up benefitting everyone involved. When someone contributes code for their own need, collectively everyone ends up benefitting from the individual changes. It's not an altruistic system, nobody is doing this to please corpos like MIT license usually is.
bit. The GPL allows you to make modifications to software, and then not publish the code of those modifications, as long as you don't distribute the software. If you only use the software in-house, you can make unpublished modifications as much as you like. That's what "keep it all to themselves" most likely referred to.
You can distribute the software and still not publish it. The source just has to be available to who you distribute the binaries to. It could be the case that your community is non technical and don't care about the source code. It could be the case that having the changes is "cool" and leaking them to be public wouldn't make it cool anymore and could get them kicked out or shamed.
>Make available the original source code when you distribute any binaries based on the licensed work
You only have to make it available to people you distribute the binaries to. If people with the binaries don't want it or if they want to keep it to themselves they are free to do so.
The reasons why corpos contribute back to upstream projects have nothing to do with the license. Usually it is just to shift the maintenance costs to the upstream developers.
but this doesnt really explain what makes gpl superior to mit. in my experience, gpl is expensive, and i have seen plenty of fair contribution, ownership, and community making things better with mit. what i was trying to say was that code published in the open will inevitably be abused by humans and machines alike. the difference between gpl and mit, in my view, is that more permissive licenses are less at odds with this reality. with copyleft there will always be lawsuits
Those lawsuits are necessary in order to protect free software.
They have also brought us much good, as without the GPL you wouldn't have OpenWRT or LineageOS, projects which typically depend on vendor-provided kernels or drivers in order to function.
so? the gist of the mit license is, "use without restriction". without the license the default is that you have no right to the work, so the license is what makes it a gift. you're welcome
no it very much isn't use without restriction - you must credit the the author and otherwise follow the licensee. I can't copy and republish code from the Nvidia driver leaks because those are copyrighted by Nvidia (unless I have permission from them). In the same way you cannot copy and republish code from a MIT project unless I have permission from them (via the MIT licensee).
