Martin Bishop5d ago
FreeBSD Foundation

At SCALE 23x, we asked a simple question:
Why do you love FreeBSD?

From performance and security to documentation, ZFS, and the community itself, the answers filled the board quickly.

FreeBSD means different things to different people:
Now we’re asking you.

Why do you love FreeBSD?

Drop your answer in the comments.
#FreeBSD #OpenSource #BSD #SCALE23x #Community

Mar 20 at 8:46pmBuffer
Show threadF4GRX Sébastien5d ago

@FreeBSDFoundation did you take a stance against usage of AI to write freebsd code?

will you fight the OS age verification laws?

Thats the only things we care about. There could be a migration of linux users to BSD, are you ready for it?

Show threaddch  5d ago

@f4grx I can’t speak on behalf of the foundation but I have been involved in drafting some of the AI options the Project is looking at.

Within our community are Hard No AI people, through to Hell Yeah people.

A strong position at either end will almost certainly cause some people to leave the project.

Threading this needle in a way that is both respectful to individuals and acknowledging the history of our licence is not simple.

@FreeBSDFoundation @EnigmaRotor

Show threaddch  5d ago

@f4grx for age notification, personally I do not expect govts to actually enforce this for the very thin slice of open source OS, but I could see it targeting Windows iOS macOS Android.

When we see what those companies are doing, whether legally or technically, we have a more clear signal.

I don’t think the FLOSS community has a spare 50 million to sue federal and state govts on enacted legislation, our best best is if orgs like ACLU or FSF engage

@FreeBSDFoundation @EnigmaRotor

Show threadEnigmaRotor5d ago
@dch @f4grx @FreeBSDFoundation What I don’t get is the concept of deporting this task to the operating system level. Ok there are already information entered when adding a user so an extra data field added won’t change the face of earth. Yet, the overall thing about age verification seams to induce concerns about the “what’s next?” thing. We already had proofs of misconduct by certification actors (see discord). There again, is a difficult question on a societal level. Gosh, talking technical is much more simple…
Show threadEnigmaRotor5d ago
@dch @f4grx @FreeBSDFoundation We should rely on people experience and clearly state the level of exigence the OS deserve. I would understand using AI as a review assistant, or a drafting coder. And I would be ok as long as a human commits the code with the same level of exigence FreeBSD shown in its history. What I would fear a lot is the stupid copy/paste or even worse a direct commit by Claude Code. And another questioning is about the difference we can expect between the operating systems and the ports/apps. While the OS can and should (in my opinion) be preserved from laziness AI-coding, I would expect the inevitable for ports being based on “tainted” upstream repositories. Fingers crossed to still count on a solid FreeBSD OS, and then it will be up to users to trust packages or not, or review their quarterly/latest update policies to avoid a big mess.
AI became a societal question and I would not expect anyone or any project to have a clear and easy positioning on it. Preserving what is important to get solid infrastructures is vital for the years to come. But we already now the rest will likely be a maintenance/cybersecurity nightmare. (Think: the xz affair scaled at another level, by malicious actor or less-than-junior-skill contributors to codebases). Coding used to be an art of some kind, not just a long list of generated tokens. What a world we are living in…
Show threaddch  5d ago
@EnigmaRotor we already have AI tooling in ports as well as LLM-contributed content in ports too. I feel that’s a reasonable compromise so far. But there’s no way to label 36_000 ports as SLOPFREE that I can think of tha scales with humans and is still trustable as an end user.
Show threadEnigmaRotor4d ago
@dch That’s a concern but to me core/project can only act on a limited but important OS level, so that merges are carefully reviewed. The current (if I understood well) way of limiting participants (commit bit with serious dev/validators) is maybe enough to secure things at FreeBSD level, in my opinion. As for the ports, do we need review-bots that would audit code (meaning: glorified AIs judging proposed PR as “safe” or not by some defined rules/standard and escalating to human). Well that’s very complicated. In any case, I think we can expect a huge workload for project maintainers/integrators. The risk is : will they be held accountable for all, when the virtual developer won’t be because of its nature of dev bot… so to me everyone is under pressure. Project lead, maintainer, user… I’m feeling old. I already wasn’t very _agile_ but now, I’m depressed, I think
Show threadkens5d ago
@FreeBSDFoundation definitely
the documentation because its very informative
Show threadPranay S. Yadav4d ago
@FreeBSDFoundation I love FreeBSD for its elegant simplicity and cohesiveness, in the way it is designed and in the way it functions. Plus the docs+manual are just fabulous!
Show threadgmc5d ago
@FreeBSDFoundation Jails and ZFS!