[gentoo-user] Re: Removing pulseaudio

[Steven J. Long]

On Fri, Apr 26, 2013 at 04:50:43PM +0800, Mark David Dumlao wrote:
> On Fri, Apr 26, 2013 at 3:55 AM, Alan McKinnon <alan.mckinnon@×××××.com> wrote:
> > And you are vastly overstating the desirability of having pulseaudio
> > enforced on users without very good cause
> How much barefaced lying can you do in one sentence?
> 1) it's not enforced _on you_. USE=-pulse

Not enforced on Gentoo, no, which is why many of us use it. But we're discussing
pulseaudio in the wider ecosystem (you certainly are) which does affect us too.

> 2) bluetooth headset goes in, audio goes out is good cause.

Yeah and if you need it all power to you: look you can install it real simply
or it comes by default on some distros. What about the rest of us who either
don't give a damn about audio beyond the speakers on our computer, with hifi
TV et al separate, or are actually into quality audio, and use jack?

See: you cannot predict the use-cases. By definition, you will not be present
when the software is run by the end-user. So you have to learn humility, and
let the user decide. Hence what was said before about software not imposing
itself, especially when not in even use.

One True Way inturgrated idiot-box crap doesn't allow that. It's the antithesis
of Unix. And if you can't deal with the fact that Linux is a *nix, use something
else instead of imposing layers of crap on the rest of us. Especially your dud
spangly new ideas that are turds you want the rest of us to polish while you
sell your "enterprise" distro based on everyone else's work. It's poisoning
the software ecosystem.

> > and seem to have
> > underestimated how deep that rabbit hole goes.
> No I haven't. I have no idea how deep the complexity of pulseaudio is
> because I don't know how to use it. I don't know how to use it because
> it just works.
<snip>
> But if I compare
> how well I learned to use grub vs pulseaudio, two things that I use
> everyday, it's clear that one of them was more successful in hiding
> the complexity from me before I used it successfully. HINT: it wasn't
> grub.

Funny, I spent even less time learning to use the KDE artsd and it worked
too. I never had any problems with it at all, yet I've heard of a lot
of issues with pa, more worryingly to do with the mentality the "developer"
imposes as a condition of working with him. I still got rid of it, and am
much happier with my current, Lennartware-free, setup thanks.

Must be something about "what programs actually do, rather than just"
misleading analogies and invalid comparisons.
 
> If you actually talk like it matters what the programs do, rather than
> just making airy abstractions on what some ideal fetishized system
> should be like, you'll understand things better.
>
> > "It does no harm and might be useful for some" is simply not a valid
> > reason to enforce a package on all users, especially when said package
> > is the latest johnny-come-lately from a wunderkind with a proven
> > reputation for writing invasive code[1]
> Oh dear. I should've realized what this was really about. There aren't
> really any technical reasons behind this, are there? Just some good
> old fashioned Lennart hate boners.
>
> I have a perfect halloween campfire story for this group. The one
> where a malicious udev update gives a backdoor for He Who Must Not Be
> Named to install his LennartWare onto yor systems...

Newsflash: it's called "systemd" and you can't get udev without it.

Nor can you build udev separately, you must install all the requirements
and build the full systemd package: they deliberately broke that. Even
though systemd has nothing to do with udev: it's a complete layering
violation.

They have nfc about what "not breaking userspace" means. They tried to
push binary logfiles in the kernel; they broke module-loading and blamed
it on everyone else; and they designed a system with a race builtin, despite
claiming loud and wide that they are the "experts in the dynamic early
userspace domain". Oh and let's not forget the wonderful decision to use
XML in system space, plus the current nonsense about hw bus-ids being stable.

But sure, these amateurs are just who we want writing system-critical
code..

Smart businesses won't be so dumb. Nor will smart users. Good luck to the
rest of you, you have my sympathy: I see your pain on IRC every day.

-- 
#friendly-coders -- We're friendly, but we're not /that/ friendly ;-)