Forgot your password?
typodupeerror
Microsoft Linux

Author of Systemd Quits Microsoft To Prove Linux Can Be Trusted (theregister.com) 124

Lennart Poettering has left Microsoft to co-found Amutable, a new Berlin-based company aiming to bring cryptographically verifiable integrity and deterministic trust guarantees to Linux systems. He said in a post on Mastodon that his "role in upstream maintenance for the Linux kernel will continue as it always has." Poettering will also continue to remain deeply involved in the systemd ecosystem. The Register reports: Linux celeb Lennart Poettering has left Microsoft and co-founded a new company, Amutable, with Chris Kuhl and Christian Brauner. Poettering is best known for systemd. After a lengthy stint at Red Hat, he joined Microsoft in 2022. Kuhl was a Microsoft employee until last year, and Brauner, who also joined Microsoft in 2022, left this month. [...]

It is unclear why Poettering decided to leave Microsoft. We asked the company to comment but have not received a response. Other than the announcement of systemd 259 in December, Poettering's blog has been silent on the matter, aside from the announcement of Amutable this week. In its first post, the Amutable team wrote: "Over the coming months, we'll be pouring foundations for verification and building robust capabilities on top."

It will be interesting to see what form this takes. In addition to Poettering, the lead developer of systemd, Amutable's team includes contributors and maintainers for projects such as Linux, Kubernetes, and containerd. Its members are also very familiar with the likes of Debian, Fedora, SUSE, and Ubuntu.

This discussion has been archived. No new comments can be posted.

Author of Systemd Quits Microsoft To Prove Linux Can Be Trusted

Comments Filter:
  • by boudie2 ( 1134233 ) on Saturday January 31, 2026 @06:08AM (#65960658)
    I feel better already.
  • by Viol8 ( 599362 ) on Saturday January 31, 2026 @06:08AM (#65960660) Homepage

    Poettering can't.

    • Re: (Score:2, Troll)

      Poettering can't.

      Hahaha. Only you know what? All of this is part of why this - and every other year - isn't the year of Linux on the desktop.

      People who are getting fed up with Windows issues can't look at Linux as a clean, problem-free destination. Systemd, initd foopd hoopd sunnyd... the only people who care about this drama are embedded in the Linux world already. From the outside, all the stories about KDE this, Gnome that, arguments of distros and windowing managers and... if Linus is or isn't vibe coding this wee

      • by Viol8 ( 599362 )

        Android aside the vast majority of Linux installs are backend servers when stuff like how easy it is interact with and debug issues with the init startup and other fundamental subsystems of the OS *DOES* matter. Clearly you have no idea about any of this.

      • I think you got modded down for begging the question.

        What I mean here is, do people outside of the Linux world actually even hear about any of that stuff, let alone get scared off by it? What little I see about Linux from the mainstream tech press is generally cautiously optimistic and points users towards specific low-friction distributions where they won't be making a lot of choices. KDE vs GNOME is completely irrelevant to your average user, who wouldn't use advanced features if they had them. They don't

        • I think you got modded down for begging the question.

          What I mean here is, do people outside of the Linux world actually even hear about any of that stuff, let alone get scared off by it?

          People, yes. I'm one of them though I'd say turned-off rather than scared-off. I'm a three-decades-plus sysadmin. I've worked with SCO Unix, Irix, Linux, DOS, Novel Netware, OS/2 and every version of Windows from 3.0 to current though the vast, vast majority has been Windows. The value proposition of Linux over Windows on my personal machines is dubious since I can wrangle it to behave as I see fit. The spray of choices and politics in the Linux world is daunting.

          Random users and grandmas? No. Yo

    • by Arrogant-Bastard ( 141720 ) on Saturday January 31, 2026 @08:27AM (#65960696)
      It's pithy and funny, but it's true. Poettering is the enemy of Linux, open source, and security, which is why he was hired and well-paid by Microsoft. My guess is that they're still paying him to set up this sham company in order to continue sabotaging Linux.

      If the Linux community had any sense, they would blacklist this asshole -- from EVERYTHING -- for life.
    • by sinkskinkshrieks ( 6952954 ) on Saturday January 31, 2026 @08:51AM (#65960706)
      Hahaha. You read my mind.

      The problem is breakage-ignorant and make-everyone-change-everything people get way too much influence per the value that they deliver mixes with gullible project managers who then foist this half-baked, buggy, incompatible, fragmentation-producing shit onto everyone else. While we could've done better with FHS across NIXes, it would've been better to have Borne script caching infrastructure than take over almost every goddamn thing.

      New | different doesn't always deliver enough net value to justify change.

      I hope this venture goes for a long time, wastes lots of AI investment money, and dies a Transmeta-like death. (VLIW could've been cool but I think Itanium proved the compilers at the time didn't know how to optimize for it.)
    • Besides, Linux isn't - or at least doesn't have to be - synonymous w/ systemd

      Maybe Poettering can figure out a way for Emacs to run directly on systemd, w/o needing a separate kernel

      • I think the trope demands that emacs incorporate systemd. You access it via ctrl-e ctrl-s or some such. (Don't flame me for that; I use vi, and don't know any emacs ctrl- keys.)
        • by dskoll ( 99328 )

          I remember when people joked that Emacs stood for "Eighty Megs And Constantly Swapping"

          *sob* VSZ of a current Emacs at startup is about 740MB and RSS is over 90MB.

          But I still use Emacs. My muscle memory goes back 36 years and it's not going to change.

          • by tbuskey ( 135499 )

            I remember when people joked that Emacs stood for "Eighty Megs And Constantly Swapping"

            The trope is for 8 megs. My 486 motherboard couldn't go beyond 16MB and adding 8MB was > $2000.

            Linux would run, with X11, on 4MB. So double the RAM and emacs still caused swapping.
            Luckily, web browsers came out and made emacs look frugal with RAM

    • by gweihir ( 88907 ) on Saturday January 31, 2026 @03:38PM (#65961296)

      That essentially sums it up. And systemd already had a few nice catastrophes, none of which affected me because I do not use it.

  • No trust (Score:2, Flamebait)

    Well, Linux can already hardly be trusted, but CIA spy Poettering certainly never can.
    Let's hope systemd will die soon!

    • Poll: runit, s6, or OpenRC?
    • Let's go w/ BSD then? Or HURD, if one needs GPL?
    • Remind me again what was wrong with sysinit scripts?

      • Remind me again what was wrong with sysinit scripts?

        I think the idea of systemd, like most of the runtime systems, is fine, but I take issue with the expanse of systemd and it directly incorporating things, like date/time and resolver, it doesn't need to simply manage a system and its startup. The developers seem to have a not invented here attitude about things. Also, would it kill them to have it run as PID 2 instead of 1?

      • by thogard ( 43403 )

        Remind me again what was wrong with sysinit scripts?
        The sysv init was never finished. If you notice there are some similarities of inittab and a makefile.
        The idea was the runlevel and action fields could be used for dependency and concurrency. The run levels A,B & C are hints of that as is the current total lack of use of the id field. There are hints in the 5ess software as well. Init's early features were limited because it could never be paged out and multi-cpus weren't an option on the newer 3b line

        • The sysv init was never finished.

          Okay.

          If you notice there are some similarities of inittab and a makefile.

          What?

          The idea was the runlevel and action fields could be used for dependency and concurrency.

          But both of those things are actually used. So are the IDs, they actually have meaning. And the runlevels are part of dependency. You pass through runlevels on your way to other runlevels.

          The run levels A,B & C are hints of that as is the current total lack of use of the id field.

          Almost total. It's relevant to ttys :)

          Init's early features were limited because it could never be paged out

          init being limited is good. We want it to be simple. It doesn't often have to do anything but you want it to respond right away when it does. Today we can afford to burn more memory so we can afford a more complicated init from that perspective, but having more complexity in

          • by thogard ( 43403 )

            If you notice there are some similarities of inittab and a makefile.

            What?

            Look at the names and descriptions in the fields and the pre-tab makefile format for far more similarity.

            But both of those things are actually used. So are the IDs, they actually have meaning. And the runlevels are part of dependency. You pass through runlevels on your way to other runlevels.

            The ID's aren't used for dependency. The id in sun sysv is a hack based on things like telnetd and rlogin. Part of the plan was to have the action field reference the ids and a state which was not done. The idea was to have a make like dependency check based on the current state of what was running.
            Had inittab been expanded slight

            • In order to not need scripts you would need a very ugly inittab, even if it had been expanded.

              I didn't say the IDs were used for dependency.

  • uh (Score:5, Insightful)

    by drinkypoo ( 153816 ) <drink@hyperlogos.org> on Saturday January 31, 2026 @08:31AM (#65960698) Homepage Journal

    Poettering will also continue to remain deeply involved in the systemd ecosystem.

    I therefore trust that it will continue to be shit.

    • Re:uh (Score:5, Interesting)

      by sinkskinkshrieks ( 6952954 ) on Saturday January 31, 2026 @09:03AM (#65960720)
      When I was at Meta (briefly) and in SV startup land, there were way too many Poettering (and Musk) fanboys. While I grew up in SV, I was gentrified out and renounce asshole VCs, managers, engineers, and broligarchs, and now I'm elsewhere with saner housing prices.

      daemontools and runit at least were simple and dependable, could drop-in and stop services on-the-fly, and didn't have a weird-ass broken state where it refused to start because it failed X times before. Oh and GFL shipping systemd syslog to other systems. syslog-ng was even more performant and robust than rsyslog, even though rsyslog was nicer to configure.
      • now I'm elsewhere with saner housing prices.

        The trick is to live somewhere with sane housing prices while working for a Silicon Valley company, for SV pay -- or at least close to it. Yeah, they'll ding you a little for being remote, because they can, so you'll only be making 80% of what you would living in SV... but you'll still be making 3-5X what local companies will pay.

    • by gweihir ( 88907 ) on Saturday January 31, 2026 @04:23PM (#65961384)

      It really cannot be anything else, regardless of maintainer. It increases complexity while trying to hide it. That is the road to hell in all engineering. Sure, you can (and should) do it to a limited degree and very carefully when some technology is really mature and exceptionally reliable. But IT and software is not there yet and will not be there yet for a long time. And hence the idea behind systemd is a very bad one at this time.

  • Fade Away (Score:4, Funny)

    by SlashbotAgent ( 6477336 ) on Saturday January 31, 2026 @08:43AM (#65960704)

    Hopefully Poettering will fade away now.

    Systemd succeeded injecting itself because he had Red Hat's bulk behind him. As an independent, he'll have no traction. I guarantee he'll be looking for work ina year's time.

    Please, no one hire him.

    • I guarantee he'll be looking for work ina year's time.

      Or maybe he'll take up golf... Poettering, pottering, puttering, putting.

  • Trust? (Score:5, Insightful)

    by thePsychologist ( 1062886 ) on Saturday January 31, 2026 @09:27AM (#65960738) Journal

    Trust is more than just technical stuff. You actually have to have the right attitude. Pottering's pretentious and overconfident attitude already casts doubt that he can be trusted.

  • Can someone ELI5 what they are trying to build? Is it something similar to Android's verified boot + attestation or Play Integrity?
    • by laktech ( 998064 ) on Saturday January 31, 2026 @12:55PM (#65961026)
      DRM
    • by Wyzard ( 110714 )

      Likely something similar, yes. People have valid concerns about remote attestation being used for DRM, but that's mostly from the standpoint of an individual who uses services provided by big companies. For businesses deploying Linux across an enterprise, however, there's a lot of value in being able to cryptographically verify that your own machines are running exactly the software they're supposed to be.

      Poettering wrote a lengthy blog post [0pointer.net] a few years ago which describes his overall vision for a signed,

  • Let's hope this is about countering attacks and not about implementing Android Play Protect certification on Linux.

  • by Z80a ( 971949 ) on Saturday January 31, 2026 @10:46AM (#65960812)

    He was probably being told daily to shove copilot in system-d somehow.
    Not ruin linux on purpose or any sort of "make windows win" kind of deal, just get copilot inside linux because the copilot SOMEONE has to use the thing.

    • He was probably being told daily to shove copilot in system-d somehow. Not ruin linux on purpose or any sort of "make windows win" kind of deal, just get copilot inside linux because the copilot SOMEONE has to use the thing.

      Given what he's decided to do, it's more likely he wanted to implement his trust verification ideas in Windows and got shut down, so he decided to go do it with Linux.

      As someone who spent much of the last decade thinking and working on topics related to system integrity and remote proof thereof, I'm interested to see what his ideas are and if they're actually novel, or at least have some innovative twist.

  • by dnaumov ( 453672 ) on Saturday January 31, 2026 @01:49PM (#65961084)

    "Author of Systemd Quits Microsoft To Prove Linux Can Be Trusted"

    does not mesh with

    "It is unclear why Poettering decided to leave Microsoft."

  • He is a known incompetent with an outsized ego. Nothing he does "proves" anything in the direction of trustworthiness. But that he was willing to join Microsoft in the first place shows a lot.

    • Linus Torvalds had demonstrated how fed up he was with Pottering's wide-ranging and stable system breaking schemes. The unannounced and disastrous "kill all background jobs" misfeature was typical, and took time and work to clean up from.

  • "Linux celeb Lennart Poettering has left Microsoft and co-founded a new company... It is unclear why Poettering decided to leave Microsoft. We asked the company to comment but have not received a response."

    The writer might have early onset of dementia, or perhaps early onset of AI syndrom.

  • Sounds like a sleep medication.

  • I threw shade on systemd a few weeks ago because he still worked for MS. I'm going to have to find some new material when I bitch about systemd.

  • Get back here and fix PulseAudio.

    • We no longer need it. We have pipewire and wireplumber. They do that job and some other jobs too. The combo replaces both pulseaudio and JACK, and also connects ALSA clients so they get their own volume controls as well.

  • Cryptographic verification of the system components and all the other stuff is important, it's a way to detect and limit the damage once a system is compromised and in the process of being infected by malware. That, though, happens far too late. For it to work you have to assume that the parts of the system that enforce verification don't have exploitable bugs in them, and we've already seen that's never the case. Especially when a single key held by a single entity is the trust root for a large number of s

Some people manage by the book, even though they don't know who wrote the book or even what book.

Working...