Lennart Poettering on the state of open source communities

There is, and this reason is not always on the merits of the software.

I would like to add NetworkManager to the list of software that (used to) pain my existence. The case for including this were pretty much the same as PulseAudio, that other entities already had settled on it, in this case Gnome.

So the fact that some of the developers were important members of the Gnome community sort of gave their creation a direct path into Red Hat and Ubuntu, who based their desktops on Gnome.

Had they not been, the software would have had a much harder time getting included, and based on merits alone it would not have happened. That rubbed a lot of people the wrong way at the time, for some very good reasons.

Never forget chronology. The ideas and architecture in systemd are nothing new, in fact they're kind of obsolete. What has changed politically / strategically such that this specific instance can be rammed down everyones unwilling throats?

Lets run a thought experiment about if this hatred is a result of technical actions and behaviors or back room politics. Lets try... the upstart init system. A direct competitor. How many people have threatened to kill the authors of upstart? Oh, none you say? What a complete non-surprise.

This is the key weakness of his post. Hatred of authors of badly architected software is very rare in open source. Hatred of back room political dealings and extensive "embrace extend extinguish" product tying, well the feeling of hatred is completely unsurprising given that behavior being the opposite of the meritocracy straw dog he tries to tear down, but some claimed behavior does sound over the top (hire a hitman? really? a bit ridiculous if true). He could have made a decent post about misbehavior and over-reaction in politics and it would have been a correct and good post, but he can't admit in public the corrupt way an unwanted systemd is being shoved down our unwilling throats, so we end up with ridiculous "I am a technical guy" LOL yeah as if anyone believes the reported social problems are due to that.

My impression, at least with systemd, is:

1) systemd put in a whole bunch of work to be backward compatible. They make migration easier, at the cost of some complexity within the codebase. It supports some compatibility with init shell scripts and has inetd compatibility.

If you want to take a dim view of it, it's a little bit like Microsoft's "embrace and extend". systemd has "good" properties for adoption. They do seem to engineer aggressively for adoption, rather than just trying to make the best technical solution. Pottering does have a good understanding of these politics, which makes people angry.

2) Features. Other systems are more "modular", but don't cover as much ground. There are some supervision toolkits like daemontools and runit, but they are meant more for servers. These systems leave it up to end users to do more work. systemd you can kind of drop in and it mostly works, until you have to debug it.

Again, this is sort of like Microsoft. You can stuff in features to satisfy all the different camps, even if the combination of those features results in a poor architecture.

3) Genuinely good ideas/features. I guess this is why Pottering is so controversial. There is a bunch of crap thrown in with some good ideas. Socket activation is genuinely a good feature. It's not new -- Apple's launchd and I believe Sun's SMF use it too. But yeah I think we should give credit to the good ideas for some of the adoption.

There's no doubt they put in a lot of effort to systemd. The only thing that compared is Upstart. It does seem that people paid to work on open source by companies like Red Hat and Canonical are the ones who really influence the direction Linux is going on.

People had issues with Upstart too, though not to this level. So really the high adoption of systemd is not that much of a surprise, given that there were few realistic alternatives.

Yes, and that reason is that redhat has their hands in other software too, and added systemd dependencies to it whether people like it or not. So distros face the decision of "adopt systemd or your distro won't be able to run gnome".

>This outpouring of faux outrage

It is not faux outrage, it is real outrage. I find it odd how people can dismiss even someone like Linus' complaints as "faux outrage".

My point was that if Lennart worked with the community instead of against it then he would have had a more positive experience from the nice guys (who outnumber the dickheads). Or to use your "look what she's wearing" argument, you don't dress like a hooker for a date with a price. But that doesn't mean that a woman cannot wear what she wants and nor that she deserves sexual / physical abuse for her choice in outfits. But that if she wants to change the monarchy then she needs to play along with them and their silly dress codes. It's best to change a system by working with and inside the system.

In other words, the nice people in the open source community would be more vocal in support for Lennart and his vision if Lennart worked nicely with the community. Dickheads will always be around (which sucks), but at least their voices would be drowned out by the millions of Lennart supporters.

While it's true that systemd does have it's supporters; equally not everybody "who actually knows about this stuff" likes systemd either.

Here's a theory. This is pure conjecture on my part—I'm not close to the systemd project or the various distros—so please take it with a grain of salt.

Programmers prefer to do less work. Simple tools get the job done and are easy to understand, but often require more effort to use effectively. Using a hand saw, it is hard work to cut a solid piece of wood, but the only servicing it needs is to sharpen the blade. On the other hand, with a chainsaw it is much easier to cut through wood, but there are many more things that you need to care about to keep it running smoothly. When put this way, the hand saw seems superior in its simplicity, but remember that it's not immediately obvious that chainsaws can be difficult to start, or that the chains get stuck sometimes. Standing in the hardware store, the chainsaw might seem like the way to go, but one might regret not choosing the simpler option when complications arise later.

It seems to me (conjecture, remember) that systemd offers a lot to the distros. It gives them hooks to manage difficult things and, after all, the systemd people are doing all the work. That explains why the distro maintainers might choose it, but my worry is that the convenience comes at a cost.

Bully or not, please be friendlier. I will start and hope you're following.

I've seen a few interviews with Mr. Pöttering and I admire his cleverness and his ambitious fight for a better GNU/Linux. I am not a fan of his person, nor his speech, but this is not the time, nor place to call him an egocentric elitist. There is a reason for everything, I too would become such a person to protect myself from the everyday threatening and hatred. Things like that break a person!! Keep this in mind, we as Engineers are especially very susceptible and sensible to attacks, because we have sharpened and trained our minds to try to find a meaning behind every action.

Remember Aaron Schwartz! He couldn't bear the weight on his shoulders anymore. He was not a soldier, he was not a criminal, he was like you and me a person that deserves to be respected. Souls break apart and suicide isn't rare among our people. You shall keep this in mind, when you speak up without a filter.

Whenever I've been complaining about something about Linux, I was told to improve it, or shut up. That was a culture that I thought as very unfriendly, but it was nonetheless logical. I could pass that along to you, but you have an opinion about Systemd and Pulseaudio that I simply cannot fathom as authentic or fact based.

ALSA did not allow controlling the volume of different applications. Period. PulseAudio made it happen!! Yes it's not perfect, but it's opensource and maybe it'll be as good as the software you want to use on your own computer.

With Systemd systems are starting up much much faster than tweaks to other init systems can provide. You're comparing Apples to Oranges, when you say, but if you disable everything then and only then my favorite init system is faster than systemd. Not everything in systemd is smart or good, but it fixes a problem that people have neglected for too many years. Systemd changed that!

Let's not discuss what is better or not, you have to be a Software-Architect and not an untrained Software-Developer to discuss such things in detail without being a hypocrite, desinformative or misleading. My impression was that even though the ideas of Mr. Pöttering were brilliant, his Software-Architecture skills are his Achilles heel. That's what he needs more training in.

Thank you for taking the time reading this.

I don't know. I wouldn't expect being harassed due to software I wrote.

This is, by the way, an offense you can end in jail for, for good reasons.

This is what systemd essentially is, an attempt to have a standardised minimal base OS beyond the Linux kernel (again like the BSD's), I don't see how it's 'highly political', if anything it's practical, which in turn is why it has been widely accepted by distros and software which makes use of it's features, like DE's.

I am likely forgetting something here, but systemd is way more than a "mere" init replacement at this time.

On one hand, personal and death threats are way out of line. On the other hand, he kinda rip what he sow. He could have done the same post without mentioning Linus and Gentoo for instance. When you put so much "energy" to push for a key component, don't be surprise by the push back. Especially if you're not willing to compromise. Yes, it can get out of hand, but that doesn't mean the push back has no ground.

Based upon what anecdotal evidence is that the 'general perception' ?

And furthermore how on earth would he be able to 'push this software' just because he works at Red Hat ?

He can complain about global warming all he wants, but if he sends a mail-bomb or threatens any ExxonMobile higher up manager, or bombs an oil-rig, that shit will not be tolerated. Altough you may argue otherwise.

You should probably re-read what was said, instead of spewing nonsense.

It's probably because he specifically works in and identifies with the larger open source community, and because people who (pretend to?) collect money to get him killed specifically identify with the open source community too.

But productive assholes are not destructive assholes, nor is being an asshole a universal character trait. A fan of sports team X can be a faithful friend to other fans of sports team X while being an asshole to all fans of rival sports team Y. A team member can be verbally abusive towards those who don't use the One True Brace Style, while being the primary architect for the system. That person is a productive asshole.

That said, these people are usually called "difficult to work with", not "assholes". Calling a coworker an asshole usually makes it more difficult to work with them, not less.

This carries the danger of people shortening it into "I have the loudest voice" in their mind, which begs the question how other communities unite the role of "last instance for contentious questions" (aka benevolent dictator) with the ideal of being nice to everyone.

But this a problem which turns up in every community where there are politicized issues. See: anything to do with women talking about video games.

We can take this problem even wider and start looking at say, American domestic politics and it's infatuation with "second amendment solutions". It's apparently socially acceptable to drop that phrase when discussing elections, policy and the president when you're a state or federal representative. Locally in Australia we've got an entire political party that carefully fans the flames of almost but not quite encouraging violence and intolerance.

You can certainly improve things at a community-by-community level, and OSS definitely should try to do so (and there should be support to do so in sensible, constructive ways). But this is a problem which goes a lot further then software development.

For some people, BSD init works just fine and they're happy with it and they abhor the additional complexity. Without any doubt, Systemd also has its flaws - it's not compatible with *BSDs, or generally non-Linux systems, it's a departure from the "everything is accessible using a simple text editor" principle that has brought Linux where we're now, and there's a non-zero switching cost for each and every package.

Many people know polipo-audio (later renamed to pulseaudio) and the crashing propensity of its earlier versions, and who are kind of intimidated by the whole DBus/ConsoleKit tangle that introduces many moving components that are hard to debug when they fail. A system that is "perfect in theory" but crashes often is not that great.

So people are kind of apprehensive when the same guy who brought them crashing sound demons a couple years ago now comes over happily with a solution to replace the most central component for their system. Thus far, I've been pleasantly surprised by Systemd silently doing its thing and working as advertised. And I wouldn't swap it out for upstart if someone gave me the choice.

The morale? Yes, people grow up, complex systems become more manageable with time when people write debugging aids for them. (Including things such as VirtualBox which make debugging central system components much easier - imagine living in a world where your systemd crashes and bugs simply get closed with a WORKSFORME tag and ignored).

And in cloud computing the admins are not kings, but peons.

Peons tilling the cattle (server) farm of 100s of servers, each running any number of VM instances. If one or more of those instances go down, new ones are spun back up.

And you see this attitude within systemd. If a daemons crashes you don't leave it down and try to figure out why it crashes. Instead systemd will simply restart any daemon it finds not to be running when it should.

Basically this is not uptime by way of applying carefully measured and maintained administration. This is uptime by machinegun.

It is what allows anyone with some grasp of php to rent server time on a amorphous blob like the Amazon EC2 and spin up the next Twitter or Facebook.

No need to optimize for or maintain the hardware. If the current load is too much, wave your credit card and have 1000 fresh instances behind the the load balancer, courtesy of Amazon or RH.

Datenwolf is demonstrating the kind of exasperation that have in the past driven people to look for alternatives to Windows. They thought they had found the promised land in Linux, but now the blight is coming over the walls and taking up residence even here.

The exact kind of behavior that has driven many from Windows to Linux.

That's not an issue that has two sides. That's an issue about his technical work, and then some people doing something abusive. There's two sides to that technical debate, but there's only one side to the other debate, and that's "stop being awful." There would only be two sides if the technical issues surrounding his work somehow justified the abuse. And they don't, because basically nothing justifies that kind of abuse in the tech world.

The comment "Linux is an awful community" is not technical, but it came after the abuse. Hopefully you can see that there's a difference between that and threatening someone.