From: Joe <email@example.com>
Sent: Thursday, 9 August 2018 7:10 PM
Subject: Re: Using Sid (was: New `no sound' problems)
On Thu, 09 Aug 2018 08:14:44 +0200
deloptes <firstname.lastname@example.org> wrote:
> Joe wrote:
> > Having said that, I don't think I've had more sound problems with my
> > sid workstations than with my stable server. Sound is generally a
> > pig on Linux, as the software base seems to change every few years,
> > and until recently, multiple sound cards had the same problem as
> > multiple NICs in that the OS couldn't seem to identify them
> > reliably. I've solved most of my sound problems by getting brutal
> > and actually ripping out and blacklisting drivers for the sound
> > devices I'm not using. Nothing less seemed to permanently solve the
> > identity crisis.
> for my workstation (I want to turn it on and just work), I use
> stable. For my server(s) the same. IMO Sid belongs in a VM for
> playing arround. If you want to be one step ahead of time, try
> testing it is usually stable.
Outside the release freeze, testing is only a little more stable than
unstable, and gets fixes later. In the long term, there's not a lot to
> If you don't read/write code, I don't see why someone would use
> unstable. As I mentioned ubuntu is much better to take in such a case
> (Not a developer, but want to be ahead of debian time)
Because there are a few applications still under development, they are
seriously buggy and continuously increasing in features. Even a few
weeks can make a big difference in functionality. I'm looking at you,
libreoffice, libreCAD, geda PCB, etc...
And since I'm not a professional developer, unstable is the practical
way to donate to Debian, in the form of bug reports. All the work has
already been done in stable.
> Regarding the sound - I never had a problem in the past 12+ years.
You are fortunate. I went though a period where the assignments for
sound card 0 and 1 would randomly flip, every few weeks or months. I
didn't find whatever magical incantation would prevent this, if it
If you look up sound problems in conjunction with Linux, the wealth of
results you get will tell you how it has been. Because it has happened
over such a long period of time, almost all of what you find will be
obsolete and completely worthless, which makes fixing the problems so
> Why? Because I did configure the system properly and I use stable. So
> instead of "getting brutal" you could setup your system properly and
> forget about the issues.
"Properly", eh? You mean spending a few days messing around with those
intuitive udev naming rules? Why should that be necessary? Surely,
running a *sound* utility *once*, and telling it which sound card I want
to use should be sufficient? Why should I need to mess around with
system stuff in order to choose my sound card and prevent it toggling
my choice now and then? That kind of stuff should happen automatically
at installation time, once and for all. Possibly it does, now.
> One bad thing that people do is the install things on the production
> system just to try them out. Take a second system - or a second drive
> - or a second installation on the same driver. Test there and move to
> the working environment, when you are sure it works.
> With other works make backups before doing something on your
> production system.
Yes, it would be nice to have batches of identical computers, and
nothing to do all day but mess about with them... this isn't a
commercial system, and I have neither the time nor the money to treat
it as one. I'm a computer *user*.