I avoid, like the plague, discussions about systemd purely because of the FUD, emotional outbursts, and general scaremongering that I see in discussions about systemd. This time I'm not going to avoid it because of all the groups I take part in this one (and one other DE group) appear to be very level headed and talk facts.

Ok here's my take on systemd and Debian. Debian is the 2nd oldest of the still current distros. It has remained a cornerstone of the Linux community and a parent to 300+ distros for many years because of the good choices made by the Debian community. You don't get a position of strength like Debian has in the FOSS world with a track record of making poor choices. Because of this I, personally, am willing to wait and watch how the systemd and Debian thing pans out. I personally do not see any problems simply because of Debian's track record. There are much greater technical minds than me making the decisions, Linus Torvolds himself is letting systemd into Linux so there must be decent technical merit for it.

In my systems, I have more than 20 Debian installs and most are Testing and Sid, I have not had one issue with systemd. Even when it was pushed into Testing and install on my machines through a normal upgrade there were no problems. I remarked on LinuxQuestions only the other day in one of the few systemd threads that I have said anything in recently that with systemd my machines boot and shutdown quicker as well.

For Cobber, because I am the only one working on it, I am trusting Debian, its developers, and technical committees. I don't have the time to change Debian for Cobber in order to keep systemd out.

There are now 2 versions of the current LFS on with and one without systemd. I plan on building both to see how well things go.

I am, personally, yet to see anything convincing that says to me that systemd is the problem for Linux that many say it is. Most discussions turn into character assasinations and at one point someone said to me "Make your choice and make the right one or be judged by the community". That to me is excessive and plays on emotion rather than technical merit. I know some people won't agree with me, and that is their right and I respect that but the problem I have seen in most discussions is the anti-systemd group do not respect those who don't agree with them.

On 18 September 2014 11:14, Mike Bird mgb-trinity@yosemite.net wrote:

On Wed September 17 2014 17:24:57 David Hare wrote:

...

Do TDE users want freedom (as much as is possible) from systemd?

Since '82 we have used many unices, distros, and desktops but the last several years we have exclusively used Debian Stable and TDE and we have been very happy with them.

We are very cautious of systemd and intend to minimize contact with it even if that means switching OS, distro, or desktop.

It's not the spaghetti design or the mediocre code but rather the leverage systemd exerts to churn software projects and distros.

Volunteers of course choose for themselves where they wish to spend their time but to my mind there are far more useful things that can be done than trying to keep software in sync with systemd's whims.

Mike Bird

---

To unsubscribe, e-mail: trinity-users-unsubscribe@lists.pearsoncomputing.net For additional commands, e-mail: trinity-users-help@lists.pearsoncomputing.net Read list messages on the web archive: http://trinity-users.pearsoncomputing.net/ Please remember not to top-post: http://trinity.pearsoncomputing.net/mailing_lists/#top-posting

On Thu, Sep 18, 2014 at 12:15 AM, Felix Miata mrmazda@stanis.net wrote:

with the perspective reading http://ewontfix.com/14/ had upon me.

Thank you Felix. That is an excellent article.

Curt-

Michael . wrote on 2014-09-18 15:11 (GMT+1000):

Felix Miata wrote on 2014-09-18 00:15 (GMT-0400):

...

...

http://igurublog.wordpress.com/2014/04/03/tso-and-linus-and-the-impotent-rag...

...

...

http://ewontfix.com/14/

Felix thanks for the links, excellent reading and very informative but again personal opinion is rife in the first one. The second one is much better and does discuss some technical issues of systemd. The problem for me is if systemd is a problem why is Linus allowing it? Why are greater technical minds letting it take over if it is the problem people say it is?

IMO:

1-At the point when "Linus'" (solely I doubt) decision to proceed in fully supporting it was committed, need and desire for what was advertised/claimed it was and would become was valid.

2-It has been well crafted to induce resisting of support for alternative init system possibilities, in addition to making continued support for sysvinit virtually impossible.

3-It evolved, expanded and subsumed ravenously, a proverbial snowballing multiplied, down a tall mountain with a steep slope, unstoppable without a bigger than herculean effort, from an organization more powerful than the Linux universe leviathan RedHat pushing it.

4-It does provide benfits.

5-Improvements continue to occur, and are expected to continue.

On Thursday 18 September 2014 06:42:49 TN Patriot wrote:

You might want to post up some proof Linus is allowing that abortion into his kernel. I haven't heard anything about that yet.

Oh dear! Here comes emotion instead of rational discussion on this list too.

"That abortion" isn't in the kernel, Linus's or anyone else's.

Lisi

The crew that makes up who says what and who says what stays and goes,

etc,

all voted and it came out a tie and had to be broken. That alone told me enough that half the Debian crew thought it was a bad idea. That's what I looked at and I'll take that to heart.

Emotions are involved here, please read the final paragraph to find out what I think about that. You are correct the vote was 4-4, 4 of the votes were employees of Canonical who created Upstart and upstart got 4 votes from those very same people. The other 4 votes had no ties at all to another distro or a pay packet.

You might want to post up some proof Linus is allowing that abortion into

his

kernel. I haven't heard anything about that yet.

Emotions again, please read the final paragraph to find out what I think about that. If you haven't heard it then why did Linus remove KS's upload privileges? If it wasn't because of his systemd uploads into the kernel, and the subsequent kernel crash issue and a coupe of bugs that everyone involved denied there coding was at fault, what was it for?

Now that is out of the way I'll comment on the emotional side of your reply., I am not interested in getting into an emotional slanging match. I don't want to go there and would appreciate it if you didn't go there with me. I'm interested in facts not emotions." Taking things to heart" and calling things "abortions" are akin to emotional blackmail and it really does lower the discussion to the lowest common denominator

On 18 September 2014 15:42, TN Patriot irgunii@gmail.com wrote:

On Thu, 18 Sep 2014 12:51:09 +1000 "Michael ." keltoiboy@gmail.com wrote:

...

I avoid, like the plague, discussions about systemd purely because of the FUD, emotional outbursts, and general scaremongering that I see in discussions about systemd. This time I'm not going to avoid it because of all the groups I take part in this one (and one other DE group) appear

to

...

be very level headed and talk facts.

Ok here's my take on systemd and Debian. Debian is the 2nd oldest of the still current distros. It has remained a cornerstone of the Linux

community

...

and a parent to 300+ distros for many years because of the good choices made by the Debian community. You don't get a position of strength like Debian has in the FOSS world with a track record of making poor choices.

The crew that makes up who says what and who says what stays and goes, etc, all voted and it came out a tie and had to be broken. That alone told me enough that half the Debian crew thought it was a bad idea. That's what I looked at and I'll take that to heart.

You might want to post up some proof Linus is allowing that abortion into his kernel. I haven't heard anything about that yet.

---

To unsubscribe, e-mail: trinity-users-unsubscribe@lists.pearsoncomputing.net For additional commands, e-mail: trinity-users-help@lists.pearsoncomputing.net Read list messages on the web archive: http://trinity-users.pearsoncomputing.net/ Please remember not to top-post: http://trinity.pearsoncomputing.net/mailing_lists/#top-posting

@ Felix, thanks again for your thoughts. your point, about "crafting" is well taken and I have to admit I do to a certain extent agree with you. Also RedHat, as you point out, does have immense power in the Linux ecosystem and, in my own not so humble personal opinion, this is not a good thing.

@ Lisi, I try to post what I know, if I am wrong I am happy to be pointed in the right direction.

My gut feeling with the Debian vote is Garbee voted systemd to limit Canonical's influence on Debian. I may be wrong and I hope I am wrong. I personally didn't mind upstart but in various things I have read over the years it seems to me that it was only ever going to be a stopgap measure as the code is apparently very untidy and it didn't improve the basic init system.

On Thursday 18 of September 2014 02:24:57 David Hare wrote:

I have today installed Debian Sid with TDE R14 on a Dell Inspiron 6400. I am very pleased to find everything (except suspend) seems to work fine with no systemd stuff whatsoever!

I have at the moment an older version of dbus (1.5.8-1) .. later versions bring libsystemd-login0 .. and libpulse0 from stable, later versions bring libsystemd0. TDE needs dbus, kaffeine needs libpulse0. These libs may not be a problem anyway.

Shutdown works. Usb notification/mounts work, which do not in other DE's without (or even with) systemd.

Do TDE users want freedom (as much as is possible) from systemd?

Does the TDE team have any particular policy towards systemd dependency?

David

TDE R14 currently provides basic support for systemd:

+ tdm - sets the class to greeter

+ kdesktop - uses SetIdleHint on / off on activate / deactivate screensaver

+ tdepowersave - uses systemd to monitor the active session and sets Inhibit function to ensure their own power event handlers

+ ksmserver / kickoff / tdepowersave - through tdehw-lib uses systemd to suspend, hibernate, reboot, and poweroff

All this support is handled entirely via dbus calls - without linking systemd libraries. Thanks to this integration with systemd is used, only if systemd is installed. The user will choose whether to install systemd or not.

We plans to use the same method to solve multi-seat support with systemd. Overall, the intention is to be able to use systemd, where this is appropriate, but not to be dependent on it.

Many are still unaware of the true nature of systemd and believe it is

merely a replacement init system. This is direct from the "horse's mouth":

I believe many who support or at least are not anti-systemd do know it is more than an init system. They, like myself, are yet to see anything that convinces them that it is the bad thing many who have become emotional about this issue believe it is.

And yet you find it's okay to sling back and yet not provide any source or

proof, just as Michael does in his reply. Hypocrites.

Lisi did not sling back and neither did I. We both commented on your use of emotion, which you continue with by name calling. This is not a school yard, it is supposed to be a technical discussion, and this is the general reason I ignore discussions like this.

As for proof I answered you with a question, a question you have ignored and chosen not to answer. You believe there is nothing of systemd in the kernel, Lisi supports you, I don't and I support my claim with the fact that Linus banned KS from uploading anything to the kernel because of the systemd bug (which highlighted a bug in the kernel). If systemd is totally divorced from the kernel Linus, and other kernel developers, had absolutely no right whatsoever to demand KS fix what they consider to be a bug in something that is not under their control or in their kernel. They could have, and probably should have, let the bug remain and this would have forced all distros to revert back to an older init system. There is no justification at all for the heated argument in April over systemd calling the kernel log if systemd is totally divorced from the kernel.

On 18 September 2014 23:36, David Hare davidahare@gmail.com wrote:

On 18/09/14 07:56, Slávek Banko wrote:

All this support is handled entirely via dbus calls - without linking

...

systemd libraries. Thanks to this integration with systemd is used, only if systemd is installed. The user will choose whether to install systemd or not.

We plans to use the same method to solve multi-seat support with systemd. Overall, the intention is to be able to use systemd, where this is appropriate, but not to be dependent on it.

Thanks Slávek.. that gives some hope (although we know TDE cannot influence upstream and mainstream)

Many are still unaware of the true nature of systemd and believe it is merely a replacement init system. This is direct from the "horse's mouth":

http://0pointer.net/blog/revisiting-how-we-put-together-linux-systems.html

Make up your own minds whether you want to support and use that without choice.. many good people, who have no inclination to emotional rant, already have.

http://igurublog.wordpress.com/

Others either accept it or (with increasing difficulty) try to find workarounds.

David

---

To unsubscribe, e-mail: trinity-users-unsubscribe@ lists.pearsoncomputing.net For additional commands, e-mail: trinity-users-help@lists. pearsoncomputing.net Read list messages on the web archive: http://trinity-users. pearsoncomputing.net/ Please remember not to top-post: http://trinity. pearsoncomputing.net/mailing_lists/#top-posting