31–40 of 235 posts
Sounds a bit like a modern, open approach to the problems that were solved by QNX. Excited to see how it evolves!
Earlier quoted context omitted.
Not really open. If you submit code, you grant a license to Google to use it in non-proprietary code.[1] So, at some point, once users and developers are locked in, Google can make later versions closed and proprietary enough to stop clones. Like they did to Android with Google Play Services. [1] https://cla.developers.google.com/about/google-individual
Disclosure: I work at Google (but not on Fuchsia), and I contribute quite a bit to open-source projects. Disclaimer: I'm not a lawyer, this is not legal advice, etc. I'm not sure what you mean by "you grant a license to Google to use it in non-proprietary code" — was that a typo and did you mean "proprietary"? In any case, Apache/BSD/MIT licensed code can already be used in proprietary code, the CLA does not change t…
> it's a copyright license, not a copyright assignment
>> So, at some point, once users and developers are locked in, Google can make later versions closed and proprietary enough to stop clones.
> you're welcome to fork it if you wish, or distribute proprietary, non-open-source versions of the code
It sounds like you're agreeing completely with everything the parent said.
Earlier quoted context omitted.
Google Play Services was never open. You can however still submit changes to the actual OS though. People said the same thing about Android in its early days. It was wrong then, still wrong now, and will likely be just as wrong in the future.
'Animats didn't claim they were ever open. His claim was more along the lines of: Google Play Services has over time subsumed more and more of core functionality, enough to stop the creation of clones. Which is true.
1. Code submitted by third parties to android 2. was moved into Google Play Services 3. And this was possible only due to the CLA.
This isn't true for a couple of reasons. 2 is unsubstantiated (it's possible this is true, but I don't think it is). Functionality certainly moved, but I don't think there's reason to believe that code was copied.
3 just isn't true. Android is Apache licensed, so code in android can be reused in proprietary code without a CLA. What you can't do is relicense Android without a CLA. Another user mentions that the Fuschia CLA doesn't allow relicensing anyway, but I'm not an expert and don't have knowledge of that.
This is a massive step back for open source. The fact that Linux doesn't have a binary-compatible driver interface is a good thing: It means hardware vendors have a very strong incentive to get their drivers upstream into the kernel. And indeed, on servers and laptops and desktop machines, this has largely happened. But on mobile platforms, with Android, it has not happened. For various reasons, but nothing fundamental.
We would all benefit if Android hardware drivers were upstreamed. This would allow for a much more competitive, and higher quality, software and hardware ecosystem on mobile platforms.
But Fuschia is going in the exact opposite direction: It makes it possible to build proprietary drivers. It's for this reason that I hope Fuschia is not successful. We already have a high quality kernel: Linux. If Fuschia is "not a science experiment", but instead intended to use proven ideas, then any improvement that could be added to Fuschia could be added to Linux. We don't need a new operating system whose only advantage is that it's easier to write proprietary drivers.
Earlier quoted context omitted.
Google Play Services was never open. You can however still submit changes to the actual OS though. People said the same thing about Android in its early days. It was wrong then, still wrong now, and will likely be just as wrong in the future.
'Animats didn't claim they were ever open. His claim was more along the lines of: Google Play Services has over time subsumed more and more of core functionality, enough to stop the creation of clones. Which is true.
Earlier quoted context omitted.
Not really open. If you submit code, you grant a license to Google to use it in non-proprietary code.[1] So, at some point, once users and developers are locked in, Google can make later versions closed and proprietary enough to stop clones. Like they did to Android with Google Play Services. [1] https://cla.developers.google.com/about/google-individual
Disclosure: I work at Google (but not on Fuchsia), and I contribute quite a bit to open-source projects. Disclaimer: I'm not a lawyer, this is not legal advice, etc. I'm not sure what you mean by "you grant a license to Google to use it in non-proprietary code" — was that a typo and did you mean "proprietary"? In any case, Apache/BSD/MIT licensed code can already be used in proprietary code, the CLA does not change t…
The sheer size of the response above, let alone content, is what creates the tone of “talking past the customer” which is what has alienated so many people from Google. The problem I’ve repeatedly experienced in Google open source and as a Google Cloud customer (contract with Google FDEs on-site) is that Googlers just don’t listen. You can’t trample the customer with your own narrative no matter how correct and elegant it is. You can disagree, but you can’t deny the feelings of others. It just doesn’t work that way.
Sounds a bit like a modern, open approach to the problems that were solved by QNX. Excited to see how it evolves!
Not really open. If you submit code, you grant a license to Google to use it in non-proprietary code.[1] So, at some point, once users and developers are locked in, Google can make later versions closed and proprietary enough to stop clones. Like they did to Android with Google Play Services. [1] https://cla.developers.google.com/about/google-individual
Not many people still consider non-copyleft open source licenses not "open" (even the FSF disagrees on the need for copyleft to be considered libre) and MIT, BSD, and Apache 2.0 in particular are widely considered proper open source, so I think you just mean "Not really copyleft", which, sure.
Another Google project that will end up being abandoned. Flutter, Dart, messaging, etc. They need to just fix Android.
Ah, the HN hivemind Markov chain is mired in another infinite cycle. https://hncynic.leod.org/
Then figure out which random sequence is a good fit for what's happening, as opposed to just looking for it in a random chain.