Mailing List Archive

On Sun, Sep 21, 2014 at 12:25 PM, Frank Peters <frank.peters@comcast.net> wrote:
> For any who need opposing information on systemd, check out the
> following web page:
>
> http://boycottsystemd.org
>
> This page was referenced on Slashdot recently and contains
> a good overview of the undesirability of systemd. Of course,
> Gentoo is mentioned as one of the few remaining distributions
> that still offer a choice.
>
> For me, point #7 is particularly odious:
>
> "7. systemd is viral by its very nature. Its scope in functionality
> and creeping in as a dependency to lots of packages means that distro
> maintainers will have to necessitate a conversion, or suffer a drift.
> As an example, the GNOME environment has adopted systemd as a hard
> dependency..."
>
> I do not oppose systemd. In fact, I'd rather not care about it in the least.
> But I want to be able to implement the boot process and system configuration
> in my own way and it seems that systemd will threaten that in the future.
>
> Take particular note of the end section "What You Can Do." There
> are plenty of alternatives and such an array of choices is what
> has always made Linux highly interesting, attractive, and useful.
> We all need to insist on keeping it this way.

This last part is important; if you don't like systemd, bitching about
it will do nothing: you have to use and contribute to the
alternatives. Linux (and Gentoo) are about choice, as long as there is
someone willing and able to provide that choice; no one will
(necessarily) provide that choice for you out of nothing.

Also, I would use better arguments than those stated in the posted
link: several of them are inaccurate, or even straight lies: in
particular to the mentioned point 7, it is false that GNOME has
adopted systemd as a hard dependency. That's just not true: GNOME 3
runs in {Open,Free}BSD just fine.

GNOME supports both systemd (logind, actually) *AND* ConsoleKit as
backends[2], so if you hear or read someone saying that GNOME depends
on systemd, that person is either spreading FUD, or showing her
ignorance.

Regards.

[1] http://undeadly.org/cgi?action=article&sid=20140219085851
[2] https://git.gnome.org/browse/gnome-session/tree/configure.ac#n139
--
Canek Peláez Valdés
Profesor de asignatura, Facultad de Ciencias
Universidad Nacional Autónoma de México

On Sun, 21 Sep 2014 12:37:58 -0500
Canek Peláez Valdés <caneko@gmail.com> wrote:

>
> This last part is important; if you don't like systemd, bitching about
> it will do nothing: you have to use and contribute to the
> alternatives. Linux (and Gentoo) are about choice, as long as there is
> someone willing and able to provide that choice; no one will
> (necessarily) provide that choice for you out of nothing.
>

The kind of choice I am speaking about is the choice of "rolling
your own." I want to be able to control and customize my system
in a way that I deem fit. The kernel, after it loads and does its
initialization thing, passes control onto an arbitrary program for
further configuration. This simple design allows extreme versatility
and customization for those who want it while also permitting more
complex schemes as well.

In this case, there is no contribution to be made. There can only
be a rant about leaving things the way they are.

How do you feel about the accuracy of the following statements which
are taken from a related web page at http://uselessd.darknedgy.net ?

"Most core Linux applications and even the kernel are developed by
a handful of companies, largely by Red Hat (who inherited much of the
work on GNU after acquiring Cygnus Solutions, thus also leading GNOME
and various other projects), who also support the opaque Freedesktop.org
standards.

"systemd is designed to be perpetually rolling software, not all that
different from a kernel in user space, as was elucidated in a 2014 GNOME
Asia talk. It has no clearly defined purpose beyond that other than the
vague 'basic building block to make an OS from' ...

"The end goal appears to be the creation of what we dub a Grand Unified
Linux Operating System (GULOS) and the destruction of the Linux distribution
altogether beyond cosmetic changes. GnomeOS, in particular. The latter is
actually a thing that GNOME aspire to accomplish."

IMO such planning and goals are slowly taking over the Linux ecosystem.
After all, RedHat cannot offer a fragmented and "hobbyist" OS to its paying
corporate clients. Only a "Grand Unified Linux OS," a la Microsoft Windows,
can compete in a professional market, and RedHat will thus lead the way in
destroying the simplicity of Linux.

These trends should be alarming to us all.

On Sun, Sep 21, 2014 at 1:30 PM, Frank Peters <frank.peters@comcast.net> wrote:
> On Sun, 21 Sep 2014 12:37:58 -0500
> Canek Peláez Valdés <caneko@gmail.com> wrote:
>
>>
>> This last part is important; if you don't like systemd, bitching about
>> it will do nothing: you have to use and contribute to the
>> alternatives. Linux (and Gentoo) are about choice, as long as there is
>> someone willing and able to provide that choice; no one will
>> (necessarily) provide that choice for you out of nothing.
>>
>
> The kind of choice I am speaking about is the choice of "rolling
> your own." I want to be able to control and customize my system
> in a way that I deem fit. The kernel, after it loads and does its
> initialization thing, passes control onto an arbitrary program for
> further configuration. This simple design allows extreme versatility
> and customization for those who want it while also permitting more
> complex schemes as well.

To "roll your own", somebody needs to provide the parts, and test that
the integration works. Nobody (necessarily) will do it for you; and
you can contribute by testing the parts you use and the integration
among them. You can use OpenRC + eudev + ubus + Xfce, help detect the
problems in them, and help reporting the issues when they don't work
correctly together.

If you don't do it, and nobody else does, then don't act surprised if
eventually everybody uses systemd, because there is people working on
it and testing it in different configurations.

> In this case, there is no contribution to be made. There can only
> be a rant about leaving things the way they are.

Wrong: see above.

> How do you feel about the accuracy of the following statements which
> are taken from a related web page at http://uselessd.darknedgy.net ?

It's a bunch of (very entertaining) FUD. To me, it losses all
credibility when it ascertains "Distro maintainers are lazy". Well
then, I expect that he maintains his own distro.

Also, I find it highly ironic that, after *years* of bashing systemd
and its design, when *no other* init system seems able to be a proper
competition, the next thing the systemd-haters try is to announce a
brand new init... by forking systemd. So, its design is not so bad
after all, right? Otherwise, they would have started from scratch.

> "Most core Linux applications and even the kernel are developed by
> a handful of companies, largely by Red Hat (who inherited much of the
> work on GNU after acquiring Cygnus Solutions, thus also leading GNOME
> and various other projects), who also support the opaque Freedesktop.org
> standards.

FUD. In systemd (and GNOME, for that matter) work people from *many*
companies; RedHat is obviously among them, but it also has developers
from Mageia, ProFusion (recently acquired by Intel), Canonical, Suse,
Collabora, Sun, IBM, etc., etc., etc. Also, it has contributors from
basically every distribution out there (including Gentoo). You can get
a list of contributors from the git repository using:

git log --format='%aN'

and then you don't need to trust anyone, but the code itself.

> "systemd is designed to be perpetually rolling software, not all that
> different from a kernel in user space, as was elucidated in a 2014 GNOME
> Asia talk. It has no clearly defined purpose beyond that other than the
> vague 'basic building block to make an OS from' ...

I actually agree with systemd being perpetually rolling software, but
I think it's a good thing. Gentoo itself is a rolling released
distribution; systemd fits perfectly with our distro; I've been using
it since 2010 in servers, desktops, laptops and everything I can put
it on, like my media center.

I don't know right now, but there was a point when I was pretty sure
systemd worked better on Gentoo than on Fedora. It's possible that
it's the case now.

Lastly, if someone sees "basic building block to make an OS from" like
something "vague", then she should do her homework.

> "The end goal appears to be the creation of what we dub a Grand Unified
> Linux Operating System (GULOS) and the destruction of the Linux distribution
> altogether beyond cosmetic changes. GnomeOS, in particular. The latter is
> actually a thing that GNOME aspire to accomplish."

I think unification among distributions is an excellent goal, but it
doesn't mean that distros will lose its identity. They will just work
better between them.

Also, I think there will be always distributions that will work with
SysV, or OpenRC, or what have you. It's Free Software.

> IMO such planning and goals are slowly taking over the Linux ecosystem.
> After all, RedHat cannot offer a fragmented and "hobbyist" OS to its paying
> corporate clients. Only a "Grand Unified Linux OS," a la Microsoft Windows,
> can compete in a professional market, and RedHat will thus lead the way in
> destroying the simplicity of Linux.

Sorry, but I call it FUD. Truth is, everything in this discussion
(systemd, OpenRC, Linux, GNOME, even uselessd) is Free Software.
Therefore, nothing is stopping anyone to take the software and
stripping out the things they don't like about it... which, BTW, is
exactly what the guy in uselessd is doing.

> These trends should be alarming to us all.

Why? Because developers are writing software as best as they think
they can? You cannot stop any developer from writing whatever the hell
they want and releasing it as Free Software. You cannot stop users
from using said software. You cannot stop distro maintainers from
deciding that software X or Y is the best option for a distribution.

In the Free Software world, you cannot stop anyone from nothing. The
only thing you can do is providing more software, or helping someone
else to provide it.

Which brings me back to my original post. Don't like systemd? Help the
competition.

Otherwise you can of course rant, but in the end that will do nothing.

Regards.
--
Canek Peláez Valdés
Profesor de asignatura, Facultad de Ciencias
Universidad Nacional Autónoma de México

The words ‘Red Hat’ have put a chill down my spine for nearly 20
years.

On Sun, Sep 21, 2014 at 2:20 PM, Barry Schwartz
<chemoelectric@chemoelectric.org> wrote:
> The words ‘Red Hat’ have put a chill down my spine for nearly 20
> years.

I know, right? A company that actually pays money to developers so
they could work on Free Software.

How dare they!

Regards.
--
Canek Peláez Valdés
Profesor de asignatura, Facultad de Ciencias
Universidad Nacional Autónoma de México

Canek Peláez Valdés <caneko@gmail.com> skribis:
> On Sun, Sep 21, 2014 at 2:20 PM, Barry Schwartz
> <chemoelectric@chemoelectric.org> wrote:
> > The words ‘Red Hat’ have put a chill down my spine for nearly 20
> > years.
>
> I know, right? A company that actually pays money to developers so
> they could work on Free Software.

So does Apple, which puts an even colder chill down my spine.

On Sun, Sep 21, 2014 at 2:33 PM, Barry Schwartz
<chemoelectric@chemoelectric.org> wrote:
> Canek Peláez Valdés <caneko@gmail.com> skribis:
>> On Sun, Sep 21, 2014 at 2:20 PM, Barry Schwartz
>> <chemoelectric@chemoelectric.org> wrote:
>> > The words ‘Red Hat’ have put a chill down my spine for nearly 20
>> > years.
>>
>> I know, right? A company that actually pays money to developers so
>> they could work on Free Software.
>
> So does Apple, which puts an even colder chill down my spine.

Good for them. As long as the code is free, I don't care who pays for
it. Even Microsoft.

Regards.
--
Canek Peláez Valdés
Profesor de asignatura, Facultad de Ciencias
Universidad Nacional Autónoma de México

On Sun, Sep 21, 2014 at 2:45 PM, Canek Peláez Valdés <caneko@gmail.com> wrote:
[snip]
> Good for them. As long as the code is free, I don't care who pays for
> it. Even Microsoft.

Free as in libre, obviously.

Regards.
--
Canek Peláez Valdés
Profesor de asignatura, Facultad de Ciencias
Universidad Nacional Autónoma de México

On Sun, 21 Sep 2014 14:22:38 -0500
Canek Peláez Valdés <caneko@gmail.com> wrote:

> On Sun, Sep 21, 2014 at 2:20 PM, Barry Schwartz
> <chemoelectric@chemoelectric.org> wrote:
> > The words ‘Red Hat’ have put a chill down my spine for nearly 20
> > years.
>
> I know, right? A company that actually pays money to developers so
> they could work on Free Software.
>

Check out page 18 of the 2014 GNOME Asia talk:
http://0pointer.de/public/gnomeasia2014.pdf

"Our objectives:

Turing Linux from a bag of bits into a competitive General Purpose
Operating System.

Building the Internet's Next Generation OS.

Unifying pointless differences between distributions."

Can it be any clearer that the Gnome (RedHat) folks desire to
usurp total control of the Linux ecosystem to serve their own
ends? RedHat needs Linux to make a profit and it will mold
Linux to better attain this end.

Is Linux currently just a "bag of bits." A lot of people
would take serious issue with this inane comment, but according
to the Gnome (RedHat) folks they are here to save us all
from the terrible shortcomings of Linux (whether we want it or
not).

Notice the remark about the "pointless differences between
distributions." This is nothing more than a disguised condemnation
of the diversity, variety, and choice which has always been the
strongest feature of the Linux world.

Now check out page 5:

"What's systemd again? ... The glue between the applications and
the kernel."

IOW, the kernel and the applications, once sufficient in themselves,
will now require the product that they (RedHat/Gnome) make and control
in order to function at all. Don't like it? Tough. Try and find a
distribution without it, and good luck re-writing all this stuff from
scratch all by your lonesome.

But why stop here? All they need to do is get rid of Linus Torvalds
himself. After all, he's just a nuisance from a previous and obsolescent
generation. Let's have the truly progressive folks, like RedHat/Gnome,
assume command of it all.

On Sun, Sep 21, 2014 at 4:13 PM, Frank Peters <frank.peters@comcast.net> wrote:
[ snip ]
> Check out page 18 of the 2014 GNOME Asia talk:
> http://0pointer.de/public/gnomeasia2014.pdf
>
> "Our objectives:
>
> Turing Linux from a bag of bits into a competitive General Purpose
> Operating System.
>
> Building the Internet's Next Generation OS.
>
> Unifying pointless differences between distributions."
>
> Can it be any clearer that the Gnome (RedHat) folks desire to
> usurp total control of the Linux ecosystem to serve their own
> ends? RedHat needs Linux to make a profit and it will mold
> Linux to better attain this end.

Whoa. How did you jumped from "Turing Linux from a bag of bits into a
competitive General Purpose Operating System" to "usurp total control
of the Linux ecosystem to serve their own ends"? There is literally no
way you can start from the first and logically arrive to the second.

With Free Software you *cannot* usurp *anything*. The code is free and
is out there. Any large group of sufficiently talented developers can
take that code and do *anything* with it. Why it hasn't happened I
explain down below, but let me be very clear: that kind of talking is
nonsense.

> Is Linux currently just a "bag of bits." A lot of people
> would take serious issue with this inane comment, but according
> to the Gnome (RedHat) folks they are here to save us all
> from the terrible shortcomings of Linux (whether we want it or
> not).

Linux *is* a bag of bits, meaning a lot of loose coupled components;
that's why when a third party developer wants to build something for
Linux they end up creating a whole distribution (SteamOS), or bundling
everything and the kitchen sink (Google Chrome). It is not demeaning,
is a statement of fact.

> Notice the remark about the "pointless differences between
> distributions." This is nothing more than a disguised condemnation
> of the diversity, variety, and choice which has always been the
> strongest feature of the Linux world.

That diversity, variety, and choice is very well, but *someone* (in
fact, many "someones") needs to work maintaining that diversity,
variety, and choice. If there is a single tool that solves the
problems of many developers, they *will* rely on that tool, and stop
supporting any inferior/less featureful tool. You would like to keep
using the less featureful tool? Then help the developers of different
projects to keep using it.

> Now check out page 5:
>
> "What's systemd again? ... The glue between the applications and
> the kernel."
>
> IOW, the kernel and the applications, once sufficient in themselves,
> will now require the product that they (RedHat/Gnome) make and control
> in order to function at all. Don't like it? Tough. Try and find a
> distribution without it, and good luck re-writing all this stuff from
> scratch all by your lonesome.

As I stated in my previous mail to you, you are spreading FUD. GNOME,
systemd, *and* the kernel have developers from many companies and
projects. There is no Illuminati inside RedHat deciding the future of
no one but that company itself.

That's first of all; second of all, Gentoo doesn't require systemd.
You want to keep it that way? Help OpenRC, and eudev, and all the
alternative projects that don't want to rely on systemd. If you (and
all the others that don't want to use systemd) don't, then (I repeat)
don't act surprised when systemd is the only option in Linux.

> But why stop here? All they need to do is get rid of Linus Torvalds
> himself. After all, he's just a nuisance from a previous and obsolescent
> generation. Let's have the truly progressive folks, like RedHat/Gnome,
> assume command of it all.

Actually, Linus seems to be OK with systemd[1]. It's probably not his
favorite project, but in that interview it ends up giving many of the
best pro-systemd arguments I've heard.

If you want to believe (or fabricate) conspiracy theories, that's
fine; I (and most Linux users) don't care about that. We care about
Linux and technological sound solutions and arguments. And that's the
crux of the matter: as I have previously stated, *any* large group of
talented developers can take the free software in all the Linux stack
(from kernel to userspace), and do *whatever* the hell they want with
it, as long as they continue to return the modified code to the
community. That's how Free Software works; that's *exactly* what
Google has done with Android.

Then why the alternatives are not attracting *huge* amount of
developers? Why uselessd is one guy, and OpenRC three or four, and
udev has a handful of developers trying to keep up with systemd-udev?

Some people will tell you that it's because of RedHat's money. And
that is so obviously wrong that is even laughable. In the kernel,
systemd, and all the other parts of the stack (including GNOME) there
are *many* companies involved. And not only small companies like
Collabora and Igalia; but *HUGE* ones like IBM and Intel. Why would
those companies let another one (RedHat) take "control" of Linux?

They don't. They *support* the idea of systemd, because (pardon me for
raising my voice) IS TECHNOLOGICALLY BETTER.

And that's what most systemd-haters don't understand. They scream and
throw tantrums about systemd, while most developers (the people that
*actually* gives us Linux, the whole stack) quietly check out the
benefits and downsides of using systemd, and in a large majority
decide that the right thing to do is using it.

That's why Arch, Suse, Gentoo-based Sabayon, Debian and even *Ubuntu*
switched (or are about to switch) to systemd. Why would Canonical
start using systemd in its distribution if it would help its rival,
RedHat, to take "control"? They would not; they switched because a
large majority of developers agree that systemd is the superior
option.

Rich Freeman (Gentoo developer, member of the Council) said better than I[2]:

"The argument about whether systemd is better/worse than sysvinit was
a debate back in 2012-2013. Just about anybody actually contributing
to distros has moved on since then. That doesn't mean that there is
100% agreement on anything, just that at this point it seems unlikely
that things are going to change much either way on that front. A few
distros are likely to avoid systemd, and the vast majority are in the
process of adopting it.

"With Gentoo you can run whatever you want for PID 1, just as you can
use whatever bootloader, kernel, syslog, etc you want. Not all the
init options have equal support - upstart isn't even in the tree and
few packages supply scripts for runit. But, nobody is going to get in
anybody's way if they want to introduce upstart, etc.

"The fact is among those actually contributing to projects like
openrc, udev, eudev, and systemd everybody tends to get along just
fine. There is plenty of interest in finding common ground and
collaborating so that anybody switching from one to another can do so
easily, and so that these projects don't diverge where it isn't
intended. It seems like the heaviest fighting seems to involve folks
who don't contribute to any of these."

I will repeat the last sentence:

"It seems like the heaviest fighting seems to involve folks who don't
contribute to any of these."

You don't *have* to use systemd; but if you *want* something
different, then you *should* contribute to the alternatives. Otherwise
people (starting with me, for what it matters) will start ignoring
you. "Oh, another one that critiques systemd without contributing to
any alternative. Most likely, he doesn't know what he's talking about.
Next."

Regards.

[1] http://www.itwire.com/business-it-news/open-source/65402-torvalds-says-he-has-no-strong-opinions-on-systemd
[2] http://article.gmane.org/gmane.linux.gentoo.user/277512
--
Canek Peláez Valdés
Profesor de asignatura, Facultad de Ciencias
Universidad Nacional Autónoma de México

On Sun, Sep 21, 2014 at 6:04 PM, Canek Peláez Valdés <caneko@gmail.com>
wrote:

> On Sun, Sep 21, 2014 at 4:13 PM, Frank Peters <frank.peters@comcast.net>
> wrote:
> [ snip ]
> > Check out page 18 of the 2014 GNOME Asia talk:
> > http://0pointer.de/public/gnomeasia2014.pdf
> >
> > "Our objectives:
> >
> > Turing Linux from a bag of bits into a competitive General Purpose
> > Operating System.
> >
> > Building the Internet's Next Generation OS.
> >
> > Unifying pointless differences between distributions."
> >
> > Can it be any clearer that the Gnome (RedHat) folks desire to
> > usurp total control of the Linux ecosystem to serve their own
> > ends? RedHat needs Linux to make a profit and it will mold
> > Linux to better attain this end.
>
> Whoa. How did you jumped from "Turing Linux from a bag of bits into a
> competitive General Purpose Operating System" to "usurp total control
> of the Linux ecosystem to serve their own ends"? There is literally no
> way you can start from the first and logically arrive to the second.
>

Actually, it seems like a pretty clear synonymous interpretation to me.
Also, I think you are using "literally" wrong in this context, as Frank
clearly "literally" just did so.


>
> With Free Software you *cannot* usurp *anything*. The code is free and
> is out there. Any large group of sufficiently talented developers can
> take that code and do *anything* with it. Why it hasn't happened I
> explain down below, but let me be very clear: that kind of talking is
> nonsense.
>
> > Is Linux currently just a "bag of bits." A lot of people
> > would take serious issue with this inane comment, but according
> > to the Gnome (RedHat) folks they are here to save us all
> > from the terrible shortcomings of Linux (whether we want it or
> > not).
>
> Linux *is* a bag of bits, meaning a lot of loose coupled components;
> that's why when a third party developer wants to build something for
> Linux they end up creating a whole distribution (SteamOS), or bundling
> everything and the kitchen sink (Google Chrome). It is not demeaning,
> is a statement of fact.
>

SteamOS and Google Chrome are both created by companies that want to have
THEIR pieces of top-down control over YOUR computer. They may have
legitimate (read: "Intellectual Property") reasons for doing so, but that
*is* nevertheless their goal, so if you're okay with ceding control to
these for-profit corporations, and paying in tangibles and intangibles to
do so, then fine. If not, do not use their products.


> > Notice the remark about the "pointless differences between
> > distributions." This is nothing more than a disguised condemnation
> > of the diversity, variety, and choice which has always been the
> > strongest feature of the Linux world.
>
> That diversity, variety, and choice is very well, but *someone* (in
> fact, many "someones") needs to work maintaining that diversity,
> variety, and choice. If there is a single tool that solves the
> problems of many developers, they *will* rely on that tool, and stop
> supporting any inferior/less featureful tool. You would like to keep
> using the less featureful tool? Then help the developers of different
> projects to keep using it.
>
> > Now check out page 5:
> >
> > "What's systemd again? ... The glue between the applications and
> > the kernel."
> >
> > IOW, the kernel and the applications, once sufficient in themselves,
> > will now require the product that they (RedHat/Gnome) make and control
> > in order to function at all. Don't like it? Tough. Try and find a
> > distribution without it, and good luck re-writing all this stuff from
> > scratch all by your lonesome.
>
> As I stated in my previous mail to you, you are spreading FUD. GNOME,
> systemd, *and* the kernel have developers from many companies and
> projects. There is no Illuminati inside RedHat deciding the future of
> no one but that company itself.
>
> That's first of all; second of all, Gentoo doesn't require systemd.
> You want to keep it that way? Help OpenRC, and eudev, and all the
> alternative projects that don't want to rely on systemd. If you (and
> all the others that don't want to use systemd) don't, then (I repeat)
> don't act surprised when systemd is the only option in Linux.
>
> > But why stop here? All they need to do is get rid of Linus Torvalds
> > himself. After all, he's just a nuisance from a previous and obsolescent
> > generation. Let's have the truly progressive folks, like RedHat/Gnome,
> > assume command of it all.
>
> Actually, Linus seems to be OK with systemd[1]. It's probably not his
> favorite project, but in that interview it ends up giving many of the
> best pro-systemd arguments I've heard.
>
> If you want to believe (or fabricate) conspiracy theories, that's
> fine; I (and most Linux users) don't care about that. We care about
> Linux and technological sound solutions and arguments. And that's the
> crux of the matter: as I have previously stated, *any* large group of
> talented developers can take the free software in all the Linux stack
> (from kernel to userspace), and do *whatever* the hell they want with
> it, as long as they continue to return the modified code to the
> community. That's how Free Software works; that's *exactly* what
> Google has done with Android.
>
> Then why the alternatives are not attracting *huge* amount of
> developers? Why uselessd is one guy, and OpenRC three or four, and
> udev has a handful of developers trying to keep up with systemd-udev?
>
> Some people will tell you that it's because of RedHat's money. And
> that is so obviously wrong that is even laughable. In the kernel,
> systemd, and all the other parts of the stack (including GNOME) there
> are *many* companies involved. And not only small companies like
> Collabora and Igalia; but *HUGE* ones like IBM and Intel. Why would
> those companies let another one (RedHat) take "control" of Linux?
>
> They don't. They *support* the idea of systemd, because (pardon me for
> raising my voice) IS TECHNOLOGICALLY BETTER.
>
> And that's what most systemd-haters don't understand. They scream and
> throw tantrums about systemd, while most developers (the people that
> *actually* gives us Linux, the whole stack) quietly check out the
> benefits and downsides of using systemd, and in a large majority
> decide that the right thing to do is using it.
>
> That's why Arch, Suse, Gentoo-based Sabayon, Debian and even *Ubuntu*
> switched (or are about to switch) to systemd. Why would Canonical
> start using systemd in its distribution if it would help its rival,
> RedHat, to take "control"? They would not; they switched because a
> large majority of developers agree that systemd is the superior
> option.
>
> Rich Freeman (Gentoo developer, member of the Council) said better than
> I[2]:
>
> "The argument about whether systemd is better/worse than sysvinit was
> a debate back in 2012-2013. Just about anybody actually contributing
> to distros has moved on since then. That doesn't mean that there is
> 100% agreement on anything, just that at this point it seems unlikely
> that things are going to change much either way on that front. A few
> distros are likely to avoid systemd, and the vast majority are in the
> process of adopting it.
>
> "With Gentoo you can run whatever you want for PID 1, just as you can
> use whatever bootloader, kernel, syslog, etc you want. Not all the
> init options have equal support - upstart isn't even in the tree and
> few packages supply scripts for runit. But, nobody is going to get in
> anybody's way if they want to introduce upstart, etc.
>
> "The fact is among those actually contributing to projects like
> openrc, udev, eudev, and systemd everybody tends to get along just
> fine. There is plenty of interest in finding common ground and
> collaborating so that anybody switching from one to another can do so
> easily, and so that these projects don't diverge where it isn't
> intended. It seems like the heaviest fighting seems to involve folks
> who don't contribute to any of these."
>
> I will repeat the last sentence:
>
> "It seems like the heaviest fighting seems to involve folks who don't
> contribute to any of these."
>
> You don't *have* to use systemd; but if you *want* something
> different, then you *should* contribute to the alternatives. Otherwise
> people (starting with me, for what it matters) will start ignoring
> you. "Oh, another one that critiques systemd without contributing to
> any alternative. Most likely, he doesn't know what he's talking about.
> Next."
>
> Regards.
>
> [1]
> http://www.itwire.com/business-it-news/open-source/65402-torvalds-says-he-has-no-strong-opinions-on-systemd
> [2] http://article.gmane.org/gmane.linux.gentoo.user/277512
> --
> Canek Peláez Valdés
> Profesor de asignatura, Facultad de Ciencias
> Universidad Nacional Autónoma de México
>
>

On Sun, Sep 21, 2014 at 5:15 PM, Harry Holt <harryholt@gmail.com> wrote:
[snip]
> Actually, it seems like a pretty clear synonymous interpretation to me.

I just happen to disagree.

> Also, I think you are using "literally" wrong in this context, as Frank
> clearly "literally" just did so.

Sorry; I'm not an English native speaker.

[snip]
> SteamOS and Google Chrome are both created by companies that want to have
> THEIR pieces of top-down control over YOUR computer. They may have
> legitimate (read: "Intellectual Property") reasons for doing so, but that
> *is* nevertheless their goal, so if you're okay with ceding control to these
> for-profit corporations, and paying in tangibles and intangibles to do so,
> then fine. If not, do not use their products.

That's your choice, and I respect that. But apart from the fact that I
would like to easily install whatever software I want in my computer,
is not only for-profit companies that want to do that; for any free
software program I wrote, if I want it available for all Linux users,
either I find a way to create packages/ebuilds for each distribution,
or I find someone that can do it for me.

Or even simpler than that: If I wrote a daemon, with SysV I could not
reliable write an script to starting it and stopping it in *all*
distributions. With systemd that actually works.

The old way doesn't scale.

Regards.
--
Canek Peláez Valdés
Profesor de asignatura, Facultad de Ciencias
Universidad Nacional Autónoma de México

On Sun, 21 Sep 2014 17:04:32 -0500
Canek Peláez Valdés <caneko@gmail.com> wrote:

>
> You don't *have* to use systemd; but if you *want* something
> different, then you *should* contribute to the alternatives. Otherwise
> people (starting with me, for what it matters) will start ignoring
> you. "Oh, another one that critiques systemd without contributing to
> any alternative. Most likely, he doesn't know what he's talking about.
> Next."
>

I appreciate your insights but let me just briefly once again
state my concerns as they may have been missed.

I do not use openrc, eudev, or anything similar, and I have no plans
to ever use systemd. All of these things are *unnecessary* at present.
I simply do not need them and do not foresee a time where I will
ever need them. In spite of any purported technical superiority they
still remain *optional*.

My system is booted and configured using my own custom scripts and
I doubt that anyone would be interested in those. They work very well
for me and as a consequence I have no interest in contributing to
alternatives that I'll never utilize. (In fact, I would encourage
everyone to develop his own set of boot/config routines. It is
not that difficult.)

The concern is that one day this will no longer be possible due to
the hegemony imposed by players such as those already mentioned.
I believe that this concern is a valid one. It will not happen
overnight but these changes will slowly creep into the Linux
universe.

My reasons are selfish. For me (and I'm sure for many, many others
who just are not aware) implementing these methods are way too much
work and will bring *no* improvements or benefits whatsoever.

If others need them then others will use them. But do not destroy
the ability to forge my own solutions.

On Sun, Sep 21, 2014 at 8:26 PM, Frank Peters <frank.peters@comcast.net> wrote:
>
> My system is booted and configured using my own custom scripts and
> I doubt that anyone would be interested in those. They work very well
> for me and as a consequence I have no interest in contributing to
> alternatives that I'll never utilize. (In fact, I would encourage
> everyone to develop his own set of boot/config routines. It is
> not that difficult.)
>
> The concern is that one day this will no longer be possible due to
> the hegemony imposed by players such as those already mentioned.

I think you need to relax a bit if that is really your worry.

You can still run a.out executables, and there is no roadmap for ever
disabling that. You can create device nodes using mknod, and I'd be
shocked if that ever went away. Just what is it that you actually
need the kernel to do for you that you don't think will still be
around in 20 years? Linus is VERY conservative about removing system
calls.

It isn't like the bits in sysvinit have an expiration date on them.

Sysvinit is only 2900 lines of code, and you could probably cut out
half of them without losing much. I doubt it will ever stop working,
but even if it did fixing whatever breaks will probably be trivial.

If the whole world moves to systemd the biggest problem you'll have is
that you'll have to write your own service startup scripts, but from
the sound of things you're doing that anyway. Most of the services
you probably run aren't linux-exclusive either, so while it seems
likely that many will start reporting their status to systemd it seems
unlikely that they will refuse to work without it.

--
Rich

On Sun, 21 Sep 2014 20:45:17 -0400
Rich Freeman <rich0@gentoo.org> wrote:

>
> You can create device nodes using mknod, and I'd be
> shocked if that ever went away.
>

But now certain static USB nodes, in particular those for
scanners, have been removed in favor of dynamic allocation
using udev or its equivalents. When this happened I was
certainly shocked, and it could be the beginning of a trend.

>
> Just what is it that you actually
> need the kernel to do for you that you don't think will still be
> around in 20 years? Linus is VERY conservative about removing system
> calls.
>

There are things which are not system calls that could easily be
changed. It is not too far fetched to consider a time if and when
systemd became so popular and entrenched that the kernel would be
hard-coded to pass control only to systemd and nothing else.

>
> If the whole world moves to systemd the biggest problem you'll have is
> that you'll have to write your own service startup scripts, but from
> the sound of things you're doing that anyway. Most of the services
> you probably run aren't linux-exclusive either, so while it seems
> likely that many will start reporting their status to systemd it seems
> unlikely that they will refuse to work without it.
>

There are a growing number of applications that will no longer compile
without either dbus or udev. In fact, even though I don't use them,
I had to install both eudev and dbus in order to be able to use certain
applications (I just substituted a symlink to /bin/true in place of
dbus-launch to keep that unnecessary daemon from starting).

I am not that familiar with systemd components, but it is not too
unrealistic to consider many more applications in the future making
at least some components mandatory.

It is obvious that the Linux of 10 years ago is no longer appealing
to many people and there will be mounting pressure to introduce changes
just for the sake of having changes.

If I have to adapt then I will certainly adapt, but it would be better
to keep current options.

On Sun, Sep 21, 2014 at 10:02 PM, Frank Peters <frank.peters@comcast.net> wrote:
> On Sun, 21 Sep 2014 20:45:17 -0400
> Rich Freeman <rich0@gentoo.org> wrote:
>> You can create device nodes using mknod, and I'd be
>> shocked if that ever went away.
>>
>
> But now certain static USB nodes, in particular those for
> scanners, have been removed in favor of dynamic allocation
> using udev or its equivalents. When this happened I was
> certainly shocked, and it could be the beginning of a trend.

Fair point, although to some extent this reflects the nature of how
modern devices work. Back in the day you had a few serial/parallel
ports and you could tell which one was which because they all used
different IO ports or IRQs that were hard-coded into the designs. Now
you just have one USB host controller which is really the only actual
true hardware device on the system and everything that is hooked up to
it is virtualized. Plug-and-play really did away with the way device
nodes tended to work, and systems like udev are probably the cleanest
solution. I for one am happy that I haven't had to configure an IRQ
since the 90s.

>
>>
>> Just what is it that you actually
>> need the kernel to do for you that you don't think will still be
>> around in 20 years? Linus is VERY conservative about removing system
>> calls.
>>
>
> There are things which are not system calls that could easily be
> changed. It is not too far fetched to consider a time if and when
> systemd became so popular and entrenched that the kernel would be
> hard-coded to pass control only to systemd and nothing else.

That seems extremely unlikely. How many people ran anything other
than sysvinit as their init for the 15 years or so before upstart came
along? Making the kernel dependent on systemd would defeat the whole
purpose of having a separation between userspace and kernelspace.

>
>>
>> If the whole world moves to systemd the biggest problem you'll have is
>> that you'll have to write your own service startup scripts, but from
>> the sound of things you're doing that anyway. Most of the services
>> you probably run aren't linux-exclusive either, so while it seems
>> likely that many will start reporting their status to systemd it seems
>> unlikely that they will refuse to work without it.
>>
>
> There are a growing number of applications that will no longer compile
> without either dbus or udev. In fact, even though I don't use them,
> I had to install both eudev and dbus in order to be able to use certain
> applications (I just substituted a symlink to /bin/true in place of
> dbus-launch to keep that unnecessary daemon from starting).

Well, it seems likely that dbus will be a kernel module before long,
so it will be readily available. I'm sure there are plenty of
programs that don't work if you don't have any number of kernel
options disabled. Kdbus is viewed as the future standard mechanism
for linux inter-process communication, so programs relying on it
should be as surprising as programs that rely on ptys.

Much of the issue boils down to the linux world becoming more
complex/functional. Back when you could assume that your printer was
attached to a parallel port and spoke postscript things were simpler.
Today people want to plug in their USB headset and have the computer
know to use the USB headset for their teleconference and put the
output in the speakers when the phone rings. That just isn't going to
work with a world where you output a sound by directing a .au file to
a device node.

--
Rich

Canek Peláez Valdés posted on Sun, 21 Sep 2014 17:28:48 -0500 as
excerpted:

> Or even simpler than that: If I wrote a daemon, with SysV I could not
> reliable write an script to starting it and stopping it in *all*
> distributions. With systemd that actually works.

IMO this the the big reason why upstreams are supporting systemd; it's a
small, often trivial, file, with a lot of bang for the buck. Ship any
other initscript and you cover a single distro. Ship a systemd unit file
or two and you cover a half-dozen rather major unrelated distros and
growing, along with most of their derivatives. Sure it's an upstream
reference file that individual distros can and reasonably often do
modify, but it's a reference file virtually guaranteed to work as-is on
nearly all those distributions, and you just can't get that elsewhere,
full-stop.


Meanwhile, for me as a gentoo user one of the biggest benefits of systemd
is that once it's the general standard pretty much everywhere I won't
have the problem of having to learn something different to maintain for
instance my openwrt-based router, as it'll be the same general init-
system on my main systems and on my router. My biggest problem with the
router right now is that it's not using an init system I'm familiar with,
and having once gone thru everything and understood how it worked well
enough to be comfortable working with the configuration, and then having
configured it, I promptly forgot all that stuff once I got it working and
had no need to screw with it any longer. Now it's seriously outdated,
but I don't want to deal with updating it and having to go thru all that
stuff to learn its special-purpose init setup once again, just to get it
working and be able to forget about it again.

I'm *REALLY* looking forward to the day when it's all standardized on
systemd and I can put the same systemd knowledge I use while maintaining
my general systems to work when I update openrc as well, and other than
the few unique unit-files, I'll "just understand it" and not have to
worry about relearning all that every time I decide it's time to upgrade
the router again.

Of course the same thing applies if I decide to make a job out of my
currently and long-term hobby of Linux. Gentoo's openrc is certainly
rather niche knowledge and won't help me much with the statistically more
likely chance that my potential employer has standardized on
centos/sle[ds]/ubuntu-server/debian/whatever. But my gentoo systemd
knowledge will "just transfer", being as useful on any of them once
everybody's switched, as on gentoo.

--
Duncan - List replies preferred. No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master." Richard Stallman

Rich Freeman posted on Sun, 21 Sep 2014 22:34:23 -0400 as excerpted:

> On Sun, Sep 21, 2014 at 10:02 PM, Frank Peters
> <frank.peters@comcast.net> wrote:

>> There are things which are not system calls that could easily be
>> changed. It is not too far fetched to consider a time if and when
>> systemd became so popular and entrenched that the kernel would be
>> hard-coded to pass control only to systemd and nothing else.
>
> That seems extremely unlikely. How many people ran anything other than
> sysvinit as their init for the 15 years or so before upstart came along?
> Making the kernel dependent on systemd would defeat the whole purpose
> of having a separation between userspace and kernelspace.

Agreed. There's far too many and too broad usages of the Linux kernel
for that sort of hard-coding, at least without at least a kconfig option
for it.

Is android suddenly going to switch to systemd? Unlikely, and it's
generally acknowledged to be the biggest usage of the Linux kernel out
there these days, so hard-coding android breakage isn't going to happen.

Plus even if it did, we're dealing with open source here and Google would
simply patch that out and their own solution in as they do with a bunch
of other stuff. And if google could do that, so could anyone else.

Then there's the tivos and the embedded medical devices and the multiple
automotive systems likely running their own little embedded Linux
kernels. Hard-coding systemd for all of that? Not going to happen.

As for the loss of the usb static device nodes, did you (Frank) file a
bug about it breaking your userspace? That's one of Linus' most firm
kernel rules -- you do *NOT* change the userspace/kernelspace API/ABI and
break userspace. However, there's a known exception. Rather like the
old philosophical question as to whether if a tree falls in the forest
and nobody hears/sees it, did it actually fall at all, if nobody notices
the userspace/kernelspace ABI breaking, did it really break at all?

Unfortunately, for support for stuff like the big databases, etc, the big
users all tend to be on enterprise distros with years-old kernels and
sometimes the changes that break that don't get noticed for years simply
because nobody running those apps is running anything close to current
kernels, or if they do, they aren't reporting the problem. Bu the time
the breakage is actually noticed and reported two years later, other
userspace may depend on the new behavior and it can become a choice of
which userspace to break, the newer stuff depending on the new behavior
or the older stuff that was broken but that nobody noticed or reported
for years. That can cause problems, particularly when those old and now
broken userspace programs are big-dollar enterprise users, but sometimes
it happens.

And Linus and the other kernel devs are constantly pointing out that if
they break userspace, report it as soon as possible so it can be fixed.
Those who fail to do so, unfortunately very occasionally have to live
with the resulting breakage, at least to some extent, tho they still go
to rather extreme lengths to finesse things if and when they can.

So if your userspace breaks due to a kernel change, report it as soon as
you detect it and ask that it be fixed. Linus is very likely to make
sure it happens. If you didn't do that, well...

--
Duncan - List replies preferred. No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master." Richard Stallman

On Mon, Sep 22, 2014 at 2:00 AM, Duncan <1i5t5.duncan@cox.net> wrote:

> Rich Freeman posted on Sun, 21 Sep 2014 22:34:23 -0400 as excerpted:
>
>
> As for the loss of the usb static device nodes, did you (Frank) file a
> bug about it breaking your userspace? That's one of Linus' most firm
> kernel rules -- you do *NOT* change the userspace/kernelspace API/ABI and
> break userspace. However, there's a known exception. Rather like the
> old philosophical question as to whether if a tree falls in the forest
> and nobody hears/sees it, did it actually fall at all, if nobody notices
> the userspace/kernelspace ABI breaking, did it really break at all?
>
> [snip]
>


> And Linus and the other kernel devs are constantly pointing out that if
> they break userspace, report it as soon as possible so it can be fixed.
> Those who fail to do so, unfortunately very occasionally have to live
> with the resulting breakage, at least to some extent, tho they still go
> to rather extreme lengths to finesse things if and when they can.
>
> So if your userspace breaks due to a kernel change, report it as soon as
> you detect it and ask that it be fixed. Linus is very likely to make
> sure it happens. If you didn't do that, well...
>
> --
> Duncan - List replies preferred. No HTML msgs.
> "Every nonfree program has a lord, a master --
> and if you use the program, he is your master." Richard Stallman
>
>
>
There are, in fact, a number of things that systemd breaks, and that the
devs refuse to fix, that even Linus has complained about. To quote:

"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.

But I'm not willing to merge something where the maintainer is known to not
care about bugs and regressions and then forces people in other projects to
fix their project. Because I am *not* willing to take patches from people
who don't clean up after their problems, and don't admit that it's their
problem to fix.

Kay - one more time: you caused the problem, you need to fix it. None of
this "I can do whatever I want, others have to clean up after me" crap.

Linus
"

And it's not just Linus. Something so pervasive, so entrenched into the
base of the system, AND that is causing problems for kernel devs to the
point that they have to implement work-arounds really needs to be reigned
in and forced to be more responsive to the needs of the OS / Linux
community as a whole, rather than the all-too-often response of "We don't
care that we've broken things you used to do in the past - this won't be
fixed and it's YOUR problem." That is the pervasive attitude of Kay
Sievers, Red Hat, and others involved in systemd development.

Here's another take from Christopher Barry, in a mailing list post from
just last month:

systemd is a coup. It is a subversive interloper designed to destroy
Linux as we know it, foisted upon us by the snarky
we-know-better-than-you CamelCase crowd. They just don't get it down
deep where it matters. systemd is not pointing in a direction that we
should be going. It does not encourage freedom. It does not encourage
choice. It does not display transparency. It does not embrace
simplicity. It seizes control and forces you to cede it. It makes
applications and major system components depend on it, and they cannot
function without it. It's gaining speed by luring naive or lazy or just
plain clueless developers into the fold with the promise of making
their lives easier. Buying into this way of thinking ignores the
greater dangers that systemd represents.

https://lkml.org/lkml/2014/8/12/459

When someone wants to take away my freedom, I get concerned.

On Mon, Sep 22, 2014 at 8:47 AM, Harry Holt <harryholt@gmail.com> wrote:
>
> On Mon, Sep 22, 2014 at 2:00 AM, Duncan <1i5t5.duncan@cox.net> wrote:
>>
>> Rich Freeman posted on Sun, 21 Sep 2014 22:34:23 -0400 as excerpted:
>>
>> And Linus and the other kernel devs are constantly pointing out that if
>> they break userspace, report it as soon as possible so it can be fixed.
>
> There are, in fact, a number of things that systemd breaks, and that the
> devs refuse to fix, that even Linus has complained about. To quote:

Duncan was talking about linux, you're talking about systemd. If Kay
broke the kernel Linus wouldn't be complaining about it, he would be
doing something about it.

--
Rich

Harry Holt <harryholt@gmail.com> skribis:
> When someone wants to take away my freedom, I get concerned.

It isn’t even necessary to be concerned about nefarious plans to be
concerned. The kind of ‘vertical integration’ systemd represents is
plainly bad software design, on its face; it is a violation of
modularity and reusability. It takes away so-called freedom simply by
being bad software that ignores decades of programming experience,
with results that are predictable -- including failure or breakage if
you try to remove or replace the poorly separated parts.

This is Programming 101, but usually it is impossible to argue with
someone on such grounds, because <the usual excuses>.

The main reason the term ‘Red Hat’ has put a chill down my spine for
20 years is not that I figured they were some evil plot to steal
‘freedom’, but that they have never been good at simplicity,
modularity, or stability of interface.

On 22 September 2014 07:13, Frank Peters <frank.peters@comcast.net> wrote:
> On Sun, 21 Sep 2014 14:22:38 -0500
> Canek Peláez Valdés <caneko@gmail.com> wrote:
>
>> On Sun, Sep 21, 2014 at 2:20 PM, Barry Schwartz
>> <chemoelectric@chemoelectric.org> wrote:
>> > The words ‘Red Hat’ have put a chill down my spine for nearly 20
>> > years.
>>
>> I know, right? A company that actually pays money to developers so
>> they could work on Free Software.
>>
>
> Check out page 18 of the 2014 GNOME Asia talk:
> http://0pointer.de/public/gnomeasia2014.pdf
>
> "Our objectives:
>
> Turing Linux from a bag of bits into a competitive General Purpose
> Operating System.
>
> Building the Internet's Next Generation OS.
>
> Unifying pointless differences between distributions."
>
> Can it be any clearer that the Gnome (RedHat) folks desire to
> usurp total control of the Linux ecosystem to serve their own
> ends? RedHat needs Linux to make a profit and it will mold
> Linux to better attain this end.
>
> Is Linux currently just a "bag of bits." A lot of people
> would take serious issue with this inane comment, but according
> to the Gnome (RedHat) folks they are here to save us all
> from the terrible shortcomings of Linux (whether we want it or
> not).
>
> Notice the remark about the "pointless differences between
> distributions." This is nothing more than a disguised condemnation
> of the diversity, variety, and choice which has always been the
> strongest feature of the Linux world.

Let's make ten different electric sockets, twenty different way to
calculate version number for softwares, thirty software licenses, and
don't forget to make at least five mutually-incompatible APIs for
every browser features that all do roughly the same thing differently.
Oh, and everyone had to write their dates in Month-Year-Day, period.
Is your life any better from having this kind of "diversity"?

Encouraging pointless differences is not supporting diversity. In
contrary, encouraging pointless differences *kills* diversity.
Diversity is a mean to the end of producing better software system, it
isn't an end of itself. If having less diversity means that I can take
my software, bring it to another totally different system, and it
works just as well as it was, then so be it.

Rich Freeman posted on Mon, 22 Sep 2014 08:53:44 -0400 as excerpted:

> On Mon, Sep 22, 2014 at 8:47 AM, Harry Holt <harryholt@gmail.com> wrote:
>>
>> On Mon, Sep 22, 2014 at 2:00 AM, Duncan <1i5t5.duncan@cox.net> wrote:
>>>
>>> Rich Freeman posted on Sun, 21 Sep 2014 22:34:23 -0400 as excerpted:
>>>
>>> And Linus and the other kernel devs are constantly pointing out that
>>> if they break userspace, report it as soon as possible so it can be
>>> fixed.
>>
>> There are, in fact, a number of things that systemd breaks, and that
>> the devs refuse to fix, that even Linus has complained about. To
>> quote:
>
> Duncan was talking about linux, you're talking about systemd. If Kay
> broke the kernel Linus wouldn't be complaining about it, he would be
> doing something about it.

Exactly. If a kernel change broke userspace, by Linus' definition, that
kernel change is broken, full-stop.

If they find out about it in the same kernel cycle, it's reverted, and
that's about as hard and fast a rule as it gets. (The only exception
would be if there's a break of userspace either way and no way to finesse
it, in context, if that break was fixing a previous break, then Linus
gets to call which break to fix.)

But of course it can only be found out about in the same kernel cycle if
someone affected is testing kernel rcs and reporting breakage.

If the breakage is found later, it's still breakage and still subject to
revert. Only by then some other userspace may be depending on the new
behavior, in which case there's a problem. Obviously this is more likely
the longer the "broken" behavior has remained in the kernel. They'll try
to finesse this case and it really is amazing sometimes the extents
they'll go to do it (one case was a special-casing of the behavior to the
specific usage in question, they were able to detect that specific usage
and special-case the specific otherwise broken behavior around it), but
if that's not possible and it has only been a kernel cycle (people only
tested the release, not the rcs, so only the single release kernel has
that behavior), they'll probably still revert it, in part because there's
relatively little released userspace that will depend on it that quickly
and very likely it'll not have made a major distro release yet.

But if the broken behavior isn't reported for several kernel cycles, say
a year (about five kernel cycles), then it really is a tough call,
particularly when there's established and widely used software already
depending on the new behavior.

Again, bottom line, report kernel breakage of userspace, the same kernel
cycle that breakage happens if at all possible, which means testing an
early enough kernel rc (rc3 is good), and it'll normally either be fixed
or the commit introducing the change reverted. The longer you wait
beyond the kernel cycle it was introduced, the more likely other
userspace depends on the new behavior, with a revert becoming
correspondingly more problematic.

And again, if it's not reported, was it a break in the first place? Just
make sure it's reported!

--
Duncan - List replies preferred. No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master." Richard Stallman

On Mon, 22 Sep 2014 06:00:20 +0000 (UTC)
Duncan <1i5t5.duncan@cox.net> wrote:

>
> As for the loss of the usb static device nodes, did you (Frank) file a
> bug about it breaking your userspace? That's one of Linus' most firm
> kernel rules -- you do *NOT* change the userspace/kernelspace API/ABI and
> break userspace. However, there's a known exception. Rather like the
> old philosophical question as to whether if a tree falls in the forest
> and nobody hears/sees it, did it actually fall at all, if nobody notices
> the userspace/kernelspace ABI breaking, did it really break at all?
>

Bug reports won't be considered. The removal of the kernel scanner module
was well-planned and deliberate. The new way is to use libusb to access
the scanner from user space. If it affects me then it affects countless others
(and there are many forum posts about this issue) but these changes will
not be reversed.

One must move ahead along with the others or be crushed and forgotten.