In my UNIX circles that I run around in, no one talks about FreeBSD. Honestly asking (yes, I guess I did not read the article) is most of its success on the commercial side? I think it is used in NAS boxes and routers/switches a lot, right?
You can infer others if you look at the historical sponsor pages: https://freebsdfoundation.org/our-donors/donors/
There are plenty of companies that may not use the full BSD stack but pull bits and pieces of code as needed. Microsoft is known to have built off BSD networking code for Windows.
30+ years ago, when everyone was basing their TCP stacks on the BSD implementation. That was one of the reasons that the infamous 64k+ "ping of death" affected more than just Windows in 1998.
The stack was re-written for Vista/Longhorn more than 20 years ago.
My point being that the GP question was about FreeBSD's success today and the Windows TCP stack is not a good candidate to illustrate that.
...though I guess you could also argue that people felt like they had more stake in GPL than BSD-licensed code.
With Raspbian, I would need more memory (1GB did the job).
On the negative side, there aren't as many packages for the BSDs on ARM.
Anyway, it got me interested enough to buy an x86 laptop just to run FreeBSD.
One thing that has amazed me is that despite the weight of development and investment in Linux far outstripping FreeBSD since around that time if not earlier, FreeBSD still managed to hold its own in performance, at least in some workloads and situations. Is there something about the people, community, methodologies, or technology that provide advantages to help achieve this? It would have been nice to know what could have been with a more even split of resources.
Although I will say in my experience by far the most capable people have been those with passion who would have been working on a project as a hobby anyway (if not so many hours), and often the best results come from those people attacking problems that interest or concern them as opposed to ones that go through committees and executives and project managers, etc. The latter often serve to crapify things in the long run by being driven by short-term and selfish thinking (not necessarily by any one developer or person, just the nature of the whole beast). Linux kernel has been doing as well as can be expected in avoiding that corporate crapification, but it certainly is not immune. I could easily believe the average developer-hour invested in FreeBSD has a far greater value to the project in general than that in Linux (yes I'm aware many FreeBSD developers do get paid to work on it, and many Linux ones don't).
(References)
https://marc.info/?l=freebsd-commits-all https://marc.info/?l=freebsd-hackers https://marc.info/?l=freebsd-current
PlayStation OS is basically the FreeBSD kernel and a custom userspace tailored to the console.
eg:
Specifically, companies needing redundancy
require more than one operating system, since any single operating system may fall
victim to a failure that could take out the
entire company’s infrastructure
Or does he mean BSD license vs GPL? In which case, I agree with his idea of diversified choices.
Yes, it keeps userland code honest if you follow the published API and not a specific behaviour. See for example the fsync() saga on Linux:
* https://lwn.net/Articles/752063/
* https://news.ycombinator.com/item?id=19238121
* https://news.ycombinator.com/item?id=19119991
It's the same reason why porting to obscure CPUs can be useful: DEC Alpha was never popular, but supporting it kind of forced Linux to be 64-bit clean in some ways, so when amd64 came along there was already a bunch of infrastructure in place.
And having 'external parties' that are not part of the same 'tribal structures' and same zeitgeist / group think can allow for experimenting of ideas.
Dave Cutler(Windows NT) too found the DEC support to be a safety test for cleanliness.
Nitpick (and you didn't necessarily imply otherwise), but x86-64 was a latecomer to 64-bit ports in Linux. SPARC I think was next after Alpha, and MIPS, PARISC, IA64, PowerPC at least all came before x86-64 was merged.
There was a story I heard that when working on the avionics for one of their planes, Boeing mandated three separate CPU architectures, with the code developed by three different, isolated, teams, using three different toolsets (languages and runtimes) to minimize the risk of some Unknown Unknown commonality taking them all out. Dunno what actually happened. Could be completely folklore.
>In my UNIX circles that I run around in, no one talks about FreeBSD.
There's a reason why the "BSD is dying" copypasta <https://everything2.com/title/BSD+is+dying> was born 25 years ago!
Juniper's boxes all used to run a FreeBSD derivative. But even they have introduced Linux based products.
At least the FreeBSD Junos still runs on the devices on top of KVM. I just can't understand why they didn't go for jails/bhyve or illumos zones instead.
This is in no way a downgrade. Juniper just read the writing on the wall. Arista boxes ran Linux from day one and have a much better reputation for stability than virtually any other vendor.
It still doesn't change the fact that putting all eggs to one penguin-shaped basket give me the heebie-jeebies. Monoculturization is never a good thing, yet everyone seems to run towards it with a glee on linux.
At least I can put money where my mouth is and continue budgeting for Juniper as long as they support FreeBSD.