4 ms·
> but I think BSD (permissive for developers, maybe restrictive for users) vs. GPL (restrictive for developers, permissive for users) Strawman. MIT/BSD is not
by cryptonector 7y ago
> but I think BSD (permissive for developers, maybe restrictive for users) vs. GPL (restrictive for developers, permissive for users)
Strawman. MIT/BSD is not restrictive for users.
When someone derives from my permissively-licensed code and doesn't share back with me? Not my problem, but theirs. They now have a legacy fork to refresh every so often at some non-trivial cost. If I'm maintaining a codebase for them to contribute, then they might be better off contributing. If my codebase is their secret sauce, well, good for them -- perhaps they should consider paying me to help them, and they are welcome to try, but it's not a requirement. It's all good.
Now, if I'm building a project to monetize, I might not open source the secret sauce. The GPL wouldn't help me there anymore than BSD/MIT. So, what good is the GPL to me?
- Digital-Citizen 7y agoNeither the MIT X11 license (one should be careful to identify which MIT license is being talked about as MIT has used many licenses for software) nor the modified (or 3-clause) BSD license look out for patent treachery. For all the user knows, perhaps your organization holds patents which read on the software you're distributing and this program is a means to give them something that tempts users to run, modify, and share software landing them in a patent infringement lawsuit. It's great that you write and distribute free software, but if you insist on using a non-copylefted free software license the Apache 2.0 license is a better choice for users of your software. But it is also in user's interests to look out for derivative works because that means users of those derivative programs get software that respects their software freedom. Proprietary derivatives of non-copylefted free software means software that doesn't respect a user's software freedom.