The Systemd-Free Debian Fork Celebrates Its Second Birthday

[SilverMachine]

The way the systemd devs wanted to push KDBUS shows to me I need to stay away from systemd. Their devs think too much of themselves. They break an API and then expect others to fix it. This was Linu's comment on KDBUS:
Originally Posted by Linus Torvalds
Key, I'm f*cking tired of the fact that you don't fix problems in the
code *you* write, so that the kernel then has to work around the
problems you cause.

Greg - just for your information, I will *not* be merging any code
from Kay into the kernel until this constant pattern is fixed.

This has been going on for *years*, and doesn't seem to be getting any
better. This is relevant to you because I have seen you talk about the
kdbus patches, and this is a heads-up that you need to keep them
separate from other work. Let distributions merge it as they need to
and maybe we can merge it once it has been proven to be stable by
whatever distro that was willing to play games with the developers.

[finalzone]

The way the systemd devs wanted to push KDBUS shows to me I need to stay away from systemd. Their devs think too much of themselves. They break an API and then expect others to fix it. ...

That quote you posted was aimed to one developer neglecting the code inside Linux kernel, nothing to do with systemd. You are just taking out of context.

[ssokolow]

And that is a bad thing because? Mozilla devs said in the blog that their web-extensions will give even more control to Noscript and friends.
It isn't going to be exactly the same as Chrome, it's an extended version supporting all Chrome supports but adding stuff supported only on Firefox.

Because, at the same time as they announced that implausibly optimistic deprecation window, they silently axed the concept of a "native.js" that provides an escape hatch so you can maintain your own APIs before convincing Mozilla of their necessity.

Now, it's only "WebExtensions Experiments", where you have to build your extension, prove the API's desirability, write code of sufficient quality to include in Firefox trunk, and convince Mozilla to merge it, all entirely within the nightly builds.

Under this regime, it would have been prohibitively demoralizing and/or difficult to produce things like Firebug which other browsers hurried to copy off Firefox.

Examples:

Suppose someone sells out and starts bundling adware with a liberally-licensed extension. This refusal to allow new XUL extensions coupled with mandatory extension signing that can't just be turned off by a command-line argument means that forks like Beef TACO and AdBlock Plus can't happen anymore. (Now that I think about it, this is exactly the kind of thing the Tivolization clause of the GPLv3 was a response to. At least with Chrome, there's a thriving ecosystem of YouTube downloaders which instruct people on how to add --enable-easy-off-store-extension-install to their launcher shortcuts.)

As far as I was able to determine while experimenting with an extension of my own, Chrome/WebExtensions APIs lack the ability to specify an output path when writing to disk, so it's completely impossible to implement things like downThemAll! or Automatic Save Folder via that API. (The closest you can get is using the API for the browser's built-in downloads system, specifying a filename, and asking to skip the Save As dialog and go straight to the default download folder.)

(I've made plans to rely on an external downloader like jDownloader where the browser extension's only purpose is to scrape URLs and ensure that any necessary cookies and referrer headers get transferred over.)

Classic Theme Restorer is also a XUL extension so, when it gets sunsetted, I'll have to switch to either Pale Moon or Chrome to retain a proper menu (not a grid of toolbar buttons in a panel) and tabs which contract to icons rather than scrolling. (The latter should be attainable via userChrome.css in the near term, but not the former.)

I'm just glad that I never got hooked on Tree Style Tabs because Firefox was too sluggish. By the time this all hits, I'll have finished the unified bookmark/session/TODO storage tool I'm currently working on. (For the sake of rapidly prototyping a sufficiently responsive UI to import two decades worth of disorganized cruft, I'm starting it as a native application and I'll split the GUI out into a ReST client later.)

[arokh]

starshipeleven
Why would anyone put such effort in forcing this thing underneath every single distribution?
Why would anyone attack Torvalds that agressively, who is working hard observing the quality and granting security of the Kernel consistently, keeping our machines working?

There are "organizations" paying people to manipulate projects. How do you know that he is none of them after all?

Tinfoil hats on, chemtrails ahead!

I don't care if it's just a technical disability or an intended manipulation in the end and I don't care about anyone feeling rude rejecting this thing...
The only thing I know is that I do not tolerate any piece of software compromising my integrity. I would rather not use the computer instead or just without network connectivity.

I share the opinion that people who like to be spied on might choose another OS because they're the only people we don't need at all.

LMFAO

Never seen a sane systemd hater.

[Passso]

The only thing I know is that I do not tolerate any piece of software compromising my integrity. I would rather not use the computer instead or just without network connectivity.

I compromised my integrity last month. My left leg switched to SystemD, now it boots a little slower but all is fine.

[oooverclocker]

What happened was that people did the easy thing and only supported 1 system - which then meant that Debian/other devs had to do extra work to maintain the other systems themselves.

I absolutely understand the wish for better compatibility. But no matter which this one system would be it's clear to me that systemd is not acceptable at its current state. So if there is nothing that suits all requirements I don't see another chance than writing a new one. That's the reason for me I was looking for a systemd-free system with all comforts which I found in Devuan and I am sure that it will evolve as a new, modern and lightweight approach without heavy Gnome as default desktop environment. And it is really, really fast without startup or shutdown hangups or delays.

[trek]

probably Devuan is not "the most vibrant distribution" simply because you can install a systemd-free Debian without any special effort, so the project lost his main purpose

[oooverclocker]

It's not that simple. For example gconf2 depends on dbus-user-session and that depends on systemd. So many packages, e.g. terminator won't work anymore if you don't solve that in the long term. Not forking Debian would mean that your packages can always break because some developer decided to user systemd-depending packages.

[trek]

It's not that simple. For example gconf2 depends on dbus-user-session and that depends on systemd.

if someone needs gnome probably he can install systemd without additional fear
I speak for myself, using openbox and leaving out dbus, gtk3, systemd and all the latest "supercool" stuff is simple and makes me happy

[duby229]

Nobody did. You completely misunderstand the entire situation, here - nobody tried forcing it everywhere, because nobody cared about that.

What happened was that people did the easy thing and only supported 1 system - which then meant that Debian/other devs had to do extra work to maintain the other systems themselves. Nobody wanted to do that extra work, so the easy thing to do was just support 1 system. The only choice was to pick which one worked best, and SystemD just won that argument.

Forcing distros to use SystemD didn't require effort, that was the default minimum everyone naturally wanted to go to. Forcing distros to use OTHER systems is what required effort.

If all the people complaining about SystemD on forums were developers who maintained software to support other systems, this never would have been an issue, because they just would have done the work to support what they wanted. The problem was that none of them actually did anything other than complain and cause problems.

That's not entirely correct. Systemd was tried to force on every distribution. Gentoo simply refused and forked udev instead, and then later on managed to get upstream udev working behind upstreams back. Udev was a tool to force systemd on everyone.

I just thank god it didn't work.