Last week, we hosted an event to mark the inaugural #SovereignTechFellowship cohort. It was an opportunity to reflect on the program’s impact, share insights from its evaluation report, and celebrate the vital contributions of these maintainers of critical digital infrastructure.

In our latest blog post, we take you inside the event with a recap of the highlights ➡️ 

https://www.sovereign.tech/news/who-will-maintain-our-digital-future

@icing @hugovk @matk

thinking about hcreate(3) again and I'm kinda curious how in the hell this made it into POSIX being like.. objectively a bad interface, even for programming standards back then.

So I'm kinda curious why they thought adding it is a good idea...

@karolherbst We need to know! 😂 😅

There are a few spots left for the event I'm hosting tomorrow evening in Berlin about the #SovereignTechFellowship. If you're interested in learning about the often unrecognized work of maintainers like @hugovk, @icing, @matk, Sarah, Jan and the other #FOSS folks that goes into maintaining our digital infrastructure and future -- and what we at the @sovtechfund are doing about it -- register here: https://events.sovereign.tech/fellowship/

@hugovk You would hope they would rate-limit the spammers instead, and give established maintainers higher limits... Damn.

@neal @pid_eins @pemensik Depends on what time a low-level tool would need it, and if user interaction is possible and if D-Bus is available...

libsystemd directly linking against something like libappstream is a bit scary (as the latter in turn links against a lot of data parsing libraries written in C, and GLib too).

Maybe this is an issue we just leave to system builders and distros to work out - it's very different from missing codecs IMHO.

@neal @pid_eins @pemensik That's what AppStream was made for originally 🙂 (to allow the system to dynamically fetch missing pieces from whatever source it has).

Let me be cheeky and preempt @pid_eins's systemd 259 posts:

In systemd 259, I'm making it possible to run commands that need privileges as your current user instead of as root. With "run0 --empower", you'll get a session as your current user in which you can do anything that root would be able to do, without actually being root.

This is very useful when you need to run something with privileges but still want all created files and directories to be owned by your current user.

#systemd259

@Lyude@queer.party With the connectivity, size and PC hardware, if it's priced like a comparable PC it will have a ton of additional applications besides being a game console. Truly exciting!

@josh @pid_eins @pwithnall Please talk this out with the people working on the FHS instead of me. Bug reports are accepted.

But you do raise the interesting question: Why *didn't* people just extend the FHS, and created the LFSH instead?

We are both Debian developers, rest assured, the question "do you intent to bring this in line with what distros do today" was the first thing I asked. Not to gatekeep (I am not supposed to do that), but out of curiosity.

@pwithnall @pid_eins From my POV, the existing situation sucked, with the FHS being outdated and conflicting with both the implementation reality of nearly all Linux distros and newer attempts like the LFSH. With people wanting to move the FHS into the future, I hope the specs can be reconciled, which would be a massive win for absolutely everyone (even if the LFSH "only" ends up a stricter subset of the FSH, that alone would be better than the status quo).

At least that's my hope for it...

@pwithnall @pid_eins I would much prefer that. 😃
Thanks, @pwithnall

It feels to me like I almost have been tricked into fighting a war that I didn't know was even happening, against people who I consider allies and who share the same goal.

I have made the "no turf war" position very clear to the FHS people, more so than I feel comfortable, because they haven't actually done anything yet since joining fd.o. And now I would like to give them a chance, like every other fd.o project.

@pid_eins I agree it can be read that way (I also did so), and I have spoken to Neal in response to it.

I do believe I have made it clear that an "us vs them" mentality is not appreciated.

@karolherbst @cas @neal ...and I will hold the people who submitted the FHS to that statement (which should not be too hard to keep, given that the FHS has a built-in section to document differences between various UNIXes).

@pid_eins If this becomes a "People's Front of Judea" vs "Judean People's Front" situation, rest assured, I won't be having that.

I spoke to the people behind it and I was assured that this wasn't the competition, so for now I believe them and see what ultimately happens.

@pid_eins Likewise 😉 - Systemd or the UAPI project could have just expanded the FHS, but since that didn't happen, we had two places saying different things.

Modernizing the old-but-still-referenced thing seemed like a no-brainer, and I didn't expect systemd to care much. The FHS hasn't been very useful for a while, but is still referenced. So modernizing it while keeping it generic enough seemed like a positive thing.

@karolherbst @cas @neal It was not presented to me as "the competition", but rather as an opportunity to align different specs to say the same thing eventually or at least to not disagree anymore.

@razze Thank you :-) - I don't intend to work on this directly, I'm more like a librarian than author here. I'm (trying to) not take on any more projects at least, I definitely have enough stuff to maintain 😄

@pid_eins Ahhhh, there is a Phoronix article! Well, that explains a lot...

@pid_eins See https://mastodon.social/@matk/115503440064987554

On a related note, I've been meaning to chat with you for a year now about fd.o+systemd things, but have been stretched rather thin (and missed the opportunity at FOSDEM). I'm sorry that this is how that issue came up, I vastly underestimated how people care about the FHS 😅

I'll reach out 🙂