Hacker News Reader

Comment by drewg123 | Hacker News Reader

Netflix currently employs about 8 FreeBSD committers (possibly more, I'm counting on my fingers before coffee, so I might be forgetting people). Most of us contribute our changes back to the project. Some of these are major things, like async sendfile. Our goal is to upstream everything we can.

We also do other things, like sponsor conferences, and work behind the scenes, advocating for FreeBSD support from a wide variety of enterprise hardware vendors.

View on HN

But why FreeBSD over Linux? Honestly it all seems like a waste of time. Linux is already the dominant free OS. Maybe it would be better to put resources into improving Linux (like Microsoft and Google do) instead? You can advocate for it all you want but the idea that FreeBSD is going to ever become as relevant as something like Linux is extremely unrealistic.

Not to mention Linux has a much better license for end-users in that it makes some effort to guarantee the code stays free, and that corporations can't close and modify the code and redistribute it for profit.

I feel like "advocating for FreeBSD" really does more harm to the open source ecosystem than good.

>You can advocate for it all you want but the idea that FreeBSD is going to ever become as relevant as something like Linux is extremely unrealistic.

Do you have any idea what you're talking about?

BSD literally invented the networking stack that was copied (poorly) by Linux, and most other Unix variants including Darwin (macOS). It is strictly superior for networked applications. Read here for more info (https://news.ycombinator.com/item?id=8167126)

Would also add that, overall, BSDs are much more lean and well designed than Linux. There are thousands of Linux developers submitting patches all the time, in every direction. That's part of what makes Linux great, but it also makes it a huge mess of code duplication and design bloat. Dbus is a mess, /proc is a mess, netlink is a mess, systemd is a mess, hell eBPF is an insane mess. Linux has a huge legacy of bad APIs, tools and design choices, that were integrated in a rush. That makes Linux trendy, but definitely not what you want when you're just interested in a stable, understandable kernel to run your core infrastructure ou embedded equipments. Not to mention even tooling and network daemons on BSDS are much easier to work with and configure IMHO.

I've been doing kernel development for Linux since a while now, and I'm always amazed at how much time it takes me to understand a new subsystem I didn't use before, because everything is SO damn over engineered it's a farm fest for geek. Whereas when I hack OpenBSD or FreeBSD on my free time I feel like I can be productive making changes in an unknown subsystem in just a few days of reading code and playing with it.

They way I've seen it described is that BSD is designed, Linux is grown.

Every part of BSD is from BSD. The kernel, network stack, init system, userland, sshd, etc are all made and released together. Ideas are driven by teams and committees and then implemented.

Every part of a functioning Linux system is from a different vendor. "Linux" makes the kernel and network stack, the init system comes from the FreeDesktop project, the userland comes from GNU, sshd comes from BSD. Things are driven in a variety of different ways, by different people, with different goals and thoughts on how Linux should look. Eventually it all gets glued together and we see what we've got.

>Not to mention Linux has a much better license for end-users in that it makes some effort to guarantee the code stays free, and that corporations can't close and modify the code and redistribute it for profit.

I disagree very strongly with this statement. The best license for the end user is the one that enables the developer to thrive financially. If that license is open source it's a bonus for the end user. I also believe that the software is only truly free if you're contributing because you care, not because of a viral contractual clause.

You gotta be joking here right?! Because something is more popular doesn't mean it has quality over everything else. And with licensing we could argue about your statement. I will end it here with BSD's networking stack, kqueue, jails, dtrace and for god's sake try something different, you may be surprised (illumos based operating systems for example)

> But why FreeBSD over Linux? Honestly it all seems like a waste of time. Linux is already the dominant free OS.

Linux isn't an OS, it's a kernel. There are hundreds of distributions that package Linux and make an OS out of it. You don't think that all that duplication of effort is an even larger waste of time?

Who cares if FreeBSD isn't as relevant. Monocultures suck. There's lots to like and lots to dislike in Linux, and I'm happy that we have other working systems from which to draw lessons and inspiration.

Microsoft has done quite a bit of FreeBSD development in the past 3 years, and is starting to do general subsystem work (VNET, jails)

Reply on news.ycombinator.com