I think it is worth noting that Google does this with Chromium so that they can make the proprietary Chrome without any concern about relicensing.
The way they've been trending, I think it's very likely that they'll close all the source they can at some point.
Conversation
Replying to
I think you're wrong about that and from my perspective Chromium dropping support for proprietary services would be major progress. It's used by a bunch of downstream browsers, not just Chrome, and it would be better off as a much more vendor neutral project closer to LLVM.
2
Replying to
Except it's never been run that way, and the upstream treats any use of the chromium source by anyone that isn't Google with extreme derision.
The moment that any of those chromium based browsers begin to seriously compete, I think Google decides they've had enough with open.
1
Replying to
They really don't treat it that way. Rather, they treat Linux distributions making Chrome-like builds of Chromium using their proprietary services that way. They happily work with Microsoft and others upstream including giving them commit access and optionally Chromium emails.
1
They don't want people to make browsers using proprietary Google services. Most major browser projects also don't want to use them. I don't think you can compare how they treat Edge with how they treat say, Arch Linux making a browser using what they see as their Chrome services.
2
Replying to
The only remaining point I think I want to add here is that Google gave the distros that support in the first place.
But this is trending very far away from my beliefs on CLAs.
1
Replying to
Well, you shifted the topic to this. I was just mentioning that cla.developers.google.com/about/google-i is an example of a CLA that I regular contribute under and I'd only consider it a problem if they were using something like GPL where they get to do stuff with my code I can't with their code.
2
As long as they use permissive licensing, I'm fine with contributing under permissive licensing including letting them do whatever they want with the code. MIT license is my default for anything and I find it benefits me the most.
1
I won't contribute to a project if they use restrictive licensing but expect me to let them use my code permissively. That is why I brought up the example of a CLA that I am fine with for a permissively licensed project. I wouldn't contribute to a Google project that was GPL.
1
I'll happily contribute to a GPL project, but I expect them to use my code as GPL if they expect me to use their code that way. That's just how I see things. Their CLA is not that much different from just giving them Apache 2 licensed code. Their attributions are generic anyway.
1
It would be a bit unfair if they preferred say, MIT licensing, but expected people to give them a patent grant via the CLA. They generally prefer Apache 2 though, so that seems fair enough. The CLA is largely just explicitly stating what giving them Apache 2 code already does.
They could word it as you agreeing to give them your code under the Apache 2 license with the generic attribution used by that project (The Chromium Authors, etc.) and the semantics wouldn't be that different. Not very fair if they don't give you their code as Apache 2 though.
1
So, similarly, if they used the MIT license, they wouldn't be giving you that explicit patent grant, but their CLA has you give them one, so that wouldn't be very fair either.
I just expect they let me use their code in a comparable way as they get to use mine, CLA or not.