New Group Calls For Boycotting Systemd

[TheBlackCat]

The phases of Venus and the moons of Jupiter could be used as an argument for a heliocentric solar system, but it wasn't evidence at all since it could be explained by other means, especially the Tychonic system that at the time was backed with the best data available.

Looking through a telescope wouldn't solve the dispute at all, and everybody, including Galileo, would have known that. The astronomers associated with the Vatican wasn't fools.

Whatever your interpretation of the implications might be, it is well-documented, including by Galileo himself, that prominent astronomers of the day refused to look through the telescope because they felt that what they were told they would see contradicted Catholic doctrine of the day, including the heliocentric model of the day.

[interested]

Although the events are correct, the people are not. Galileo did invite leaders at the time to look through a telescope, but it wasn't a Catholic tribunal, but rather Catholic (or rather Jesuit) astronomers. Things like the phases of Venus and the moons of Jupiter, which could been seen through a telescope, had a direct bearing on the idea of heliocentrism, and others, like mountains on the Moon, were very closely related ideas.

The phases of Venus and the moons of Jupiter could be used as an argument for a heliocentric solar system, but it wasn't evidence at all since it could be explained by other means, especially the Tychonic system that at the time was backed with the best data available.

Looking through a telescope wouldn't solve the dispute at all, and everybody, including Galileo, would have known that. The astronomers associated with the Vatican wasn't fools.
This isn't to say that the Catholic church wasn't abusive against Galileo, ultimately they didn't rely on scientific arguments against his ideas, but on brute death threats and imprisonment.

[interested]

This is what really pisses me off with both systemd and pulseaudio defenders. There is never any problem with either, the problem is always elsewhere, even if everything worked perfectly before either was installed.

That there may bugs in either PA and systemd isn't what was being discussed, but statements that Lennart Poettering is a mediocre programmer, and that systemd is badly programmed, and journald is "a pile of crap" etc.

Systemd haters aren't interested in a proper discussion about systemd, they only care about discrediting it. (A lost cause now that systemd has won).

Whatever you think of PA, the project really revitalized sound on Linux and began a systematic debugging of both drivers and ALSA. Sound on Linux before PA was bad and a low priority with developers.

"PA was a brilliantly executed software project" must be one of the biggest lies I've seen written anywhere, and only those that haven't suffered the first releases of pulseaudio (before LP left) can believe that absurdity. Pulseaudio wasn't adopted because it is the best or even a fully working solution - it was adopted because of politics.

I used the PA from the very beginning and I too had severe bugs and problems with it, but most of them was simply because my internal, slightly exotic sound chip had really bad drivers; when using a well supported PCI sound card, most of the problems disappeared (though non-native Linux programs like VLC and flash continued to be a problem for a while)

But that PA would have severe problems was to be expected, all other sound projects had similar problems, if not worse, and they weren't system wide either. As I said, this was a really an insanely hard problem that PA solved. Had it been trivial, you would have seen another system wide sound daemon many years before, and likely, many competing projects.

But that it was a hard problem that no one else had been able to solve, isn't the sole reason for why the PA projects was brilliantly engineered; Poettering also cared for the deployment of PA and that it required as little work as possible (beside bug fixing).

Pray tell me, how else could the problem have been solved? Making yet another sound system instead of ALSA? A complete rewrite of all the kernel drivers? A complete rewrite of every user space program?
Poettering avoided all these classic traps of redoing everything from scratch and thereby make a new backwards incompatible solution. He made a backwards compatible solution. Just the fact how easy you can disable or even remove PA shows how well engineered the project is. With most other solutions there would have been a flag day where everything was committed to the new way, or it didn't work.


Your claim that "politics" whatever that is, was the reason why it was adopted so quickly is simply wrong. Everybody had wanted a proper system wide sound solution for years, especially the DE designers, who rightly complained that they shouldn't have to deal with messy low level stuff like kernel drivers.
Yes, there were driver bugs and ALSA bugs and kernel bugs, and bugs in PA too, but it was still better than everything else before.

Besides, claiming that LP is some kind of genius and that pulseaudio works great (which it doesn't) thanks to him is to conveniently forget that pulseaudio just started working somewhat after LP dropped it.

Yet another lame attempt to discredit Poettering for all his hard work.

The second thing that really, really, pisses me off is LP's fanclub calling every critic a "hater". I never was a pulseaudio hater or a systemd hater until called that by some fanboi. While criticisms are accepted and handled properly, I even cooperate within my limited time and ability. But when some jerk labels criticism as hate, that is when I start opposing and stop debating, as it is easy to see that the fanboi won't even bother with facts.

You are missing the most important thing in the negative campaign against systemd. The systemd haters aren't interested in an honest discussion about systemd, they just want to discredit it. This is why they are called systemd haters.

I am perfectly fine with people who doesn't want to use systemd or PA for whatever reason they feel like. But they should spend their energy on improving what they like instead of negative campaigning against something they don't use or like. E.g. GNU SysVinit doesn't even have a simple build test system, so the only way to test if a new version works is by booting a system with it. People who think SysVinit is the right solution for their Linux distro, should help the developers instead of attacking systemd. But they don't, just like they don't seem to care about ConsoleKit and anything else they may need on their non-systemd distros; they seems to prefer to attack open source developers and projects instead. This is why they are haters.

You criticise ad hominem attacks, but you are the ones labelling every critic as hater. Systemd and pulseaudio critics are becoming extremists, that is true. But the fault lies squarely on the fanboy camp, that refuses all arguments and keeps pushing ahead without any consideration for any use case different from theirs.

Criticism is only useful when it is honest and with an agenda to improve the criticised. But this isn't the case with systemd haters, they want to destroy systemd if possible, not improve it.

I do distinguish between systemd haters and systemd opponents. There is also a third group of those who doesn't care about systemd and prefer something else, but they don't attack systemd either, they just work on what they like instead.
AFAIK, Slackware still doesn't believe in package management, but slackers aren't wasting everybody's time by relentless trolling every rpm/deb thread and attacking the software for being badly programmed, the concept flawed, and call for boycotting distros that dares to use package management. They just do the things they believe in.

I think those who doesn't want to use systemd should stop caring about it, and start caring for what they want to use instead. This is much more productive.

[TheBlackCat]

You criticise ad hominem attacks, but you are the ones labelling every critic as hater. Systemd and pulseaudio critics are becoming extremists, that is true. But the fault lies squarely on the fanboy camp, that refuses all arguments and keeps pushing ahead without any consideration for any use case different from theirs.

First, no software will ever satisfy all use-cases. At a certain point, software developers have to decide that certain use-cases are too niche to be worth hurting other, more common use-cases. The burden is on the opponents of systemd to show that their use-cases are important enough to be worth the taking time and energy away from other, more common uses-cases, and in many cases outright hurting other uses-cases.

Second, for all the complaints about systemd, people criticizing it have been very short on actual valid use-cases. Pretty much the only valid one I have seen is that it is not well suited to very low-end embedded systems, which is a valid complaint, but systemd doesn't really interfere in this area anyway since the software that depends on systemd also isn't suited to such systems. People complain about not liking the design, but not liking the design is not a "use0case". People have complained about it having features they don't need, but having features you don't need isn't a "use-case". "I just don't want to use it" isn't a "use-case". Not liking the person or (what you think is) the company behind something certainly isn't a "use-case".

So besides ultra low-power embedded systems, what are the use-cases that systemd supporters aren't considering?

[TheBlackCat]

Galileo Galilei was accused for promoting a heliocentric view, a problem that can't be decided by looking into a telescope, so that story is probably yet another fabricated "feel good" story. Famous quotes and anecdotes without precise source, are usually made up or heavily distorted from reality.

Although the events are correct, the people are not. Galileo did invite leaders at the time to look through a telescope, but it wasn't a Catholic tribunal, but rather Catholic (or rather Jesuit) astronomers. Things like the phases of Venus and the moons of Jupiter, which could been seen through a telescope, had a direct bearing on the idea of heliocentrism, and others, like mountains on the Moon, were very closely related ideas.

[jbernardo]

When Galileo Galilei was accused of heresy and put before a tribunal of the catholic church, he invited his judges to take a look through his telescope. They refused and said: ``Don't bother us with facts. We've already formed an opinion.''

I would never have thought of comparing the systemd fanclub with the inquisition, but it is indeed an apt comparison.

[interested]

When Galileo Galilei was accused of heresy and put before a tribunal of the catholic church, he invited his judges to take a look through his telescope. They refused and said: ``Don't bother us with facts. We've already formed an opinion.''

Galileo Galilei was accused for promoting a heliocentric view, a problem that can't be decided by looking into a telescope, so that story is probably yet another fabricated "feel good" story. Famous quotes and anecdotes without precise source, are usually made up or heavily distorted from reality.

Also, remember that the hard data and facts spoke against a Copernican solar system; no star parallax could be observed, the predictions of the planetary movements didn't correspond with the observed positions, unlike the Tychonic and geocentric systems that could be arbitrarily precise at the cost of complexity.

Anyway, Lennart Poettering's major software creations like PulseAudio, Avahi and systemd, have all been extremely successful projects, used on all major distros. Especially PA and systemd are extremely impressive projects because they solved really hard problems, and solved them while still being backwards compatible. No "flag day" where everything old stopped working, and everything had to support a brand new way of doing things by doing rewrites. This shows that Lennart Poettering is an excellent software engineer.

It is quite laughable that people that have made no success software, or no known contribution to open source projects, and with no other credentials to their anonymous internet handles to why they should be able to judge the systemd code quality, just spout out that "Lennart is a mediocre programmer" and that systemd is badly programmed etc.

It is seems more likely that those statements are deliberate lies in order to discredit systemd, since the systemd haters have lost all the technical arguments against systemd, and have been routed from every Linux distro of importance.

[jbernardo]

With PA, Linux finally gained a system wide daemon; PA was a brilliantly executed software project; it had no flag day where every program suddenly had to support a new audio system, it didn't made yet another sound system, but relied on the previous ALSA, there where no major rewrites, just defining a safe subset of ALSA, and then starting the hard work of debugging both ALSA and all the faulty sound chip drivers in Linux at that time. PA became a developer and debugging nexus for all this, which is also why lesser informed people blame PA for ALSA and driver bugs.

All desktop distros started to use PA from early on, and they continue to do so, despite the PA haters absurd claim, that "PA is broken and doesn't work".

This is what really pisses me off with both systemd and pulseaudio defenders. There is never any problem with either, the problem is always elsewhere, even if everything worked perfectly before either was installed.
If you don't have sound the fault is due to buggy alsa drivers - however, in real world, the fix for 90% of the sound issues I've found is always "apt-get purge pulseaudio" (or the equivalent in your distro). Just yesterday I had to remove it from a PC to be able to have audio over HDMI again (which had worked perfectly until skype dragged in pulseaudio). "PA was a brilliantly executed software project" must be one of the biggest lies I've seen written anywhere, and only those that haven't suffered the first releases of pulseaudio (before LP left) can believe that absurdity. Pulseaudio wasn't adopted because it is the best or even a fully working solution - it was adopted because of politics.
Besides, claiming that LP is some kind of genius and that pulseaudio works great (which it doesn't) thanks to him is to conveniently forget that pulseaudio just started working somewhat after LP dropped it.

The second thing that really, really, pisses me off is LP's fanclub calling every critic a "hater". I never was a pulseaudio hater or a systemd hater until called that by some fanboi. While criticisms are accepted and handled properly, I even cooperate within my limited time and ability. But when some jerk labels criticism as hate, that is when I start opposing and stop debating, as it is easy to see that the fanboi won't even bother with facts.

You criticise ad hominem attacks, but you are the ones labelling every critic as hater. Systemd and pulseaudio critics are becoming extremists, that is true. But the fault lies squarely on the fanboy camp, that refuses all arguments and keeps pushing ahead without any consideration for any use case different from theirs.

[ceage]

<Facts>

When Galileo Galilei was accused of heresy and put before a tribunal of the catholic church, he invited his judges to take a look through his telescope. They refused and said: ``Don't bother us with facts. We've already formed an opinion.''

[interested]

Spot on. People are mistaking Lennart's manic hyperactivity for brilliance and his obsession with writing crapware for competence. The guy is a mediocre software designer and programmer at best. He has A LOT of energy but this doesn't necessarily mean he would be able to use it in the best interests of the community even if he wanted to.

Yet another lame ad hominem attack on an open source developer. Rather predictable that systemd haters continue with such personal attacks since they have no technical arguments left.

But lets look at Lennart Poetterings achievements so far:

He was the major force behind getting sound right on Linux: Before his PulseAudio project, sound was in a sorry state on Linux distros; there where no system wide sound daemon, but several DE specific ones like artsd, that everybody complained about, including the DE developers.
Making such sound daemons was an extremely hard problem on Linux, which is why no one really attempted to even try to make a system wide one, but just restricted themselves to a limited set of DE programs.

With PA, Linux finally gained a system wide daemon; PA was a brilliantly executed software project; it had no flag day where every program suddenly had to support a new audio system, it didn't made yet another sound system, but relied on the previous ALSA, there where no major rewrites, just defining a safe subset of ALSA, and then starting the hard work of debugging both ALSA and all the faulty sound chip drivers in Linux at that time. PA became a developer and debugging nexus for all this, which is also why lesser informed people blame PA for ALSA and driver bugs.

All desktop distros started to use PA from early on, and they continue to do so, despite the PA haters absurd claim, that "PA is broken and doesn't work".

There were no alternatives to PA before it was invented, and no serious alternatives afterwards, simply because the task is huge, difficult and ungrateful. Even those who smugly announce that they disable PA, benefit from the hard work of debugging ALSA and the sound chip drivers that PA developers like Lennart Poettering did.

systemd:
That project is now one of the largest open source projects in existence, with a dozen people with GIT commit access and +600 contributors. The project has attracted some really eminent kernel developers like Greg Kroah-Hartman:
http://en.wikipedia.org/wiki/Greg_Kroah-Hartman, and experienced coders from Debian, Gentoo, Red Hat, etc. Hardly a sign of "crapware".

Besides that, I can inform you that systemd actually works really well and is extremely stable, and has vibrant developer community unlike most of its competing projects.

All the doom and gloom predictions that systemd was badly programmed and therefore wouldn't work, turned out to blatantly wrong accusations, usually made by some anonymous person without any actually credentials or knowledge about the systemd or low level system programming at all.