There isn't much special secret sauce that Netflix has at the OS layer of things, so there's not much reason for them to keep the patches in-house (and then have to maintain as the public source rolls forward). Other vendors that give back are Dell EMC Isilon (keeping their OneFS code private), Juniper, Netgate, etc.
Sony is unique in that it was a one-time fork, and now that the product is out there's not much churn in things.
Most vendors have learned that keeping things in-house just causes pain down the road when you have to re-base with the latest FreeBSD release.
If you're fine with violating the license, then there is nothing stopping you from closing the source to GPL licensed software without a word of credit. I'm not sure what your point is.
The Nintendo Switch is based on FreeBSD as well. Also Juniper Networks and several other enterprise networking equipment manufacturers make extensive use of FreeBSD in their products.
With heavy use in Netflix's CDN resulting in something like more than 15% of internet traffic being delivered by FreeBSD, that's some kind of success I imagine.
Yeah I hear this statement a lot too but macos is based on nextstep much more than it was on FreeBSD.
Of course nextstep borrowed some userland from FreeBSD and I think this is where the confusion originates. The actual kernel is totally different though.
The fundamental services and primitives of the OS X kernel
are based on Mach 3.0. Apple has modified and extended
Mach to better meet OS X functional and performance goals.
It is correct to say some parts of OS-X/macOS are based on FreeBSD, but not to the point of making an unconditional assertion. As to user-land utilities, many are the same as found in a FreeBSD distribution.
A little more.. Apple hired Jordan K. Hubbard, the FreeBSD co-founder, back in 2001 to work in the Core OS Engineering Department. His role at that time was manager, BSD technology at Apple, overseeing the BSD Technology Base for Darwin, the UNIX-based core of Mac OS X.
I remember a very early release of Darwin, was interested at the time to see if it would eventually become a standalone product, but it never did. I assume there was a lack of corporate support, much like how Google is now trying to slowly close AOSP via removing core pieces like the dialer.
Can somebody elaborate on FreeBSD's "success"?