**** BEGIN LOGGING AT Thu Jun 02 02:59:58 2016
Jun 02 15:49:09 <DocScrutinizer05> pabs3: in https://lists.debian.org/debian-announce/2015/msg00001.html you write >> The sysvinit init system is still available in "Jessie".<<  I wonder how _feasible_ this option actually is, particularly regarding stuff like udev et al
Jun 02 16:34:17 <antrik> DocScrutinizer05: I'm running Debian testing with sysvinit just fine
Jun 02 16:34:59 <antrik> udev is build from the same source tree, but doesn't really depend on systemd
Jun 02 16:35:06 <DocScrutinizer05> it's rather a question if you're running it without systemd
Jun 02 16:35:14 <antrik> eh?
Jun 02 16:36:04 <DocScrutinizer05> I never looked into it but heard it's almost impossible to build e.g. udev without depending on systemd nowadays
Jun 02 16:37:35 <antrik> well, I do have the systemd libraries installed; but it's not a runtime dependency
Jun 02 18:27:05 <DocScrutinizer05> antrik: pabs3:  [ 0 -ne `ps aux|grep systemd|grep -v grep|wc -l` ] && echo "sorry, your system is not running without systemd. Even when you might have  sysv-init installed"
Jun 02 19:09:28 <DocScrutinizer05> for c in `echo $PATH|sed 's/:/ /g'`; do for cc in $c/*; do if (ldd $cc 2>/dev/null|grep -q systemd); then (echo "## $cc ###################"; ldd $cc) 2>/dev/null; fi; done; done|less +"/systemd"  ;-)
Jun 02 19:11:47 <DocScrutinizer05> best served by a root waiter
Jun 02 20:26:11 <antrik> DocScrutinizer05: well, I have a 'systemd-udevd' process, instead of just 'udevd'. that doesn't mean I'm running systemd. they basically just changed the name.
Jun 02 20:27:06 <DocScrutinizer05> hmm, I heard different, that's why people try hard to revive eudev or even invent vdev
Jun 02 20:28:34 <antrik> there might be some more or less heavy patching from Debian involved. I don't really know, and I don't really care.
Jun 02 20:32:05 <DocScrutinizer05> well, when all the misery around systemd actually resulted in debian doing heavy patching so you can get rid of systemd without breaking a bazillion packages, then devuan has basically already reached their goal
Jun 02 20:34:47 <DocScrutinizer05> I'm pretty happy when I know my system will NOT silently use 8.8.8.8 when I messed up the DNS resolver config
Jun 02 20:35:03 <DocScrutinizer05> plus a dozen more of similar PITAs
Jun 02 20:35:59 <DocScrutinizer05> cron suddenly resuming past scheduled events that were not executed in time since system been suspended, etc
Jun 02 20:38:35 <DocScrutinizer05> and the killer:  https://github.com/tmux/tmux/issues/428   https://github.com/systemd/systemd/pull/3005#issuecomment-208002146
Jun 02 20:39:30 <DocScrutinizer05> I simply don't need that kind of headache
Jun 02 20:40:21 <DocScrutinizer05> that shite worked great for 30 years and I don't want a Poettering to fix it into fubar now, just because they can
Jun 02 20:41:54 <DocScrutinizer05> first they create crap software that leaves oirphaned processes that should have terminated but were not (gnome), then they call for systemd session management to the rescue
Jun 02 20:44:08 <DocScrutinizer05> on a slightly related sidenote: did you know KDE akonadi "middleware" starts polling your email accounts even when you didn't start *any* email client?  And that's not perceived a massive security threat and violation of "do nothing unexpected" by the devels
Jun 02 20:44:59 <antrik> as much as systemd bothers me from an ecosystem point of view, I don't really mind it as a user. I'm only staying with sysvinit on this system because I'm afraid some of the config accumulated over 14 years might break in the transition and require manual intervention...
Jun 02 20:46:23 <DocScrutinizer05> that's one of the major points though. Not only your stuff needs such intervention, rather they are about to redefine linux at alrge and require patches to about $everything to match that new paradigm they try to enforce
Jun 02 20:47:18 <antrik> no, I didn't know, but it sounds useful. I think decoupling email notification from email readers is perfectly resonable. the whole idea of an "email program" as a single monolithic entity comes from Windows; it was never like in the traditional model on Unix
Jun 02 20:47:19 <DocScrutinizer05> based on really nothing more than a "we know better and that stuff is old anyway, so it must be bad and broken"
Jun 02 20:47:51 <antrik> DocScrutinizer05: that's the "ecosystem" bit
Jun 02 20:48:28 <antrik> I don't know whether anything actually requires intervention; it just *might* -- like with any major transition. I don't think systemd is very special in this regard
Jun 02 20:48:39 <DocScrutinizer05> I give a shit about what it been on which platform. I expect to be able to tell my machine to STOP email checking, and I am used to do so by not starting the email client
Jun 02 20:49:00 <DocScrutinizer05> otherwise I had configured postfix or sendmail or whatever already
Jun 02 20:49:12 <antrik> DocScrutinizer05: just because you are used to it, doesn't mean it's the one proper way :-P
Jun 02 20:49:52 <antrik> people with a different background have different expectations -- you can't demand *your* expectations to be the decisive ones
Jun 02 20:50:14 <DocScrutinizer05> it's a working valid way they must not sneakily modify completely, with breaking it since there is no option at all to stop akonadi from doing that, as soon as you start KDE
Jun 02 20:51:12 <DocScrutinizer05> they even agree that it's a flaw and not intentional
Jun 02 20:51:22 <DocScrutinizer05> but they shrug it away
Jun 02 20:52:43 <DocScrutinizer05> I wouldn't complain if kmail had a signal to akonadi which means "start email fetching" which I can control from kmail's gui
Jun 02 20:53:27 <DocScrutinizer05> then akonadi can happily continue acting as a MTA while kmail segfaulted or whatever
Jun 02 20:53:42 <antrik> uh... I agree that there should be a way to disable it.
Jun 02 20:53:46 <DocScrutinizer05> and when I tell kmail "offline" then it goes offline again
Jun 02 20:54:03 <antrik> shouldn't be tied to kmail though
Jun 02 20:54:26 <antrik> (it makes sense for kmail to have a control for it; it just shouldn't be strictly tied to kmail)
Jun 02 20:54:33 <DocScrutinizer05> but STARTING THAT before you even had a chance to open the kmail GUI, during KDE startup, that's plain rogue
Jun 02 20:55:04 <antrik> I don't think it is. but as I said, that's a matter of what background you are coming from
Jun 02 20:55:22 <DocScrutinizer05> it is rogue and a security threat
Jun 02 20:55:53 <DocScrutinizer05> I couldn't use such machine in a hisec corporate environment where I get fired for trying to do private email poll
Jun 02 20:56:09 <antrik> I don't agree. applications shouldn't "own" services. just because kmail displays mail, doesn't mean it's the only thing allowed to check the mailbox.
Jun 02 20:56:28 <DocScrutinizer05> NB akonadi also does general desktop file indexing and a lot of other shit
Jun 02 20:57:07 <antrik> yeah, Akonadi is, err, not universally appreciated ;-)
Jun 02 20:57:33 <DocScrutinizer05> HUH, nobody ever said "only kmail is allowed to check the mailbox" - I said "I NEVER ALLOWED AKONADI to check the mailbox during KDE start"
Jun 02 20:58:17 <DocScrutinizer05> I say *nobody* is allowed to check email without my explicit consent
Jun 02 20:58:24 <antrik> you shouldn't use a private machine (or a machine you use for private stuff) in a "high security environment" anyways
Jun 02 20:58:42 <DocScrutinizer05> please don't get me started on such "advice"
Jun 02 20:59:07 <DocScrutinizer05> that's not your call after all
Jun 02 20:59:23 <antrik> well, you are a control freak. that's fine. but in that case maybe KDE is not the right environment for you...
Jun 02 20:59:51 <DocScrutinizer05> BWAHAHAHA, because KDE is for losers only, thanks akonadi?
Jun 02 21:00:07 <DocScrutinizer05> then theyx need to write that on the label, on BOLD
Jun 02 21:01:24 <DocScrutinizer05> "ATTENTION! KDE is not fit for any secure environment, it's only made for the average john doe ignorant user who doesn't care about anything"
Jun 02 21:02:04 <DocScrutinizer05> I asked that in #KDE, they went "wtf? why?"
Jun 02 21:02:24 <DocScrutinizer05> I explained them ^^^^ and they shrugged it away
Jun 02 21:02:39 <DocScrutinizer05> though they admitted it was wrong
Jun 02 21:05:29 <DocScrutinizer05> NB you can avoid the problem when you forbid akonadi file indexing, so akonadi doesn't get started at all during KDE startup ;-P
Jun 02 21:05:43 <DocScrutinizer05> obviously that option is incorrectly labeled
Jun 02 21:09:10 <DocScrutinizer05> and I sense the generally identical mindset of "not *our* problem" in whole systemd cabal
Jun 02 21:09:48 <DocScrutinizer05> which causes a LOT of headache for many many people that are not JohnDoes
Jun 02 21:20:22 <DocScrutinizer05> just the dimension is like in another galaxy, since on a server you usually don't start KDE per default, but you nearly can't find any installation image for a server linux that doesn't come with systemd and friends
Jun 02 21:22:46 <DocScrutinizer05> and for example a 8.8.8.8 as nameserver fallback *hardcoded* into your libs is an absolute nogo for a server
Jun 02 21:23:24 <DocScrutinizer05> introduced courtesy systemd-resolverd or whatever bullshit
Jun 02 21:27:26 <antrik> well, the mindset of doing things that most people will appreciate while some will passionately hate is certainly polarising ;-)
Jun 02 21:28:02 <antrik> I'm not really sure where exactly I stand on that matter myself
Jun 02 21:52:04 <DocScrutinizer05> I beg to differ on the "most people appreciate" part. Most people have no clue - bejond the TV commercial blabla of the Cabal -of what systemd actually does. It ends at "even I could write system start config files now" and a nonsensical appreciation of faster boottimes which been achieved by making the boot process less deterministic and way less flexible
Jun 02 21:55:07 <DocScrutinizer05> I'm trying hard to not fall for one of the "historical reference" fuaxpas, but indeed not everything appreciated by a lot of people for what it claims it does or plans to do, is in the end actually something good and the result expected by those who initially cheered
Jun 02 21:57:34 <DocScrutinizer05> in the end the same old saying applies (in slightly augmented form): you're either one of those who wrote the business plan, or your're the customer, or you're the product
Jun 02 22:07:18 <DocScrutinizer05> linux evolved from unix, which was designed to be rock solid and educational, and simple in a positive way. Now there's a movement that denies all those historical roots and thinks linux should forget about compatibility to other unixes, should use "efficient" stuff instead of simple stuff, and all that matters is the "end user experience". In short: linux should morph from a unix into a mainstream OS like OS-X or Windows
Jun 02 22:09:07 <DocScrutinizer05> and Poettering is one of the loudest proponents of that movement
Jun 02 22:12:19 <DocScrutinizer05> funny enough the proponents of that movement think **they KNOW** what's the desirable end user experience. They don't even bother to ask the actual users, rather they throw one hype buzzword after the other into the public discourse - yet without any support why those are relevant
Jun 02 22:13:52 <DocScrutinizer05> prolly a disease of the times that everybody just falls for "It's FREEE"
Jun 02 22:13:56 <DocScrutinizer05> ~freee
Jun 02 22:13:57 <apt> i guess freee is http://www.youtube.com/watch?feature=player_detailpage&v=ZTdUmlGxVo0&t=3296
Jun 02 22:15:01 <DocScrutinizer05> anybody playing the "It's FREE!" card already won the discourse
Jun 02 22:15:38 <DocScrutinizer05> the irony is that it's always a lie, nothing is for free
Jun 02 22:17:24 <DocScrutinizer05> opting for the nleete new shite systemd promises today and prolly won't deliver tomorow just costs you your options to chose the optiomal linux system for your usecase tomorrow
Jun 02 22:18:08 <DocScrutinizer05> s/optio/opti/
Jun 02 22:18:08 <apt> DocScrutinizer05 meant: opting for the nleete new shite systemd promises today and prolly won't deliver tomorow just costs you your optins to chose the optiomal linux system for your usecase tomorrow
Jun 02 22:18:21 <DocScrutinizer05> s/optiom/optim/
Jun 02 22:18:21 <apt> DocScrutinizer05 meant: opting for the nleete new shite systemd promises today and prolly won't deliver tomorow just costs you your options to chose the optimal linux system for your usecase tomorrow
Jun 02 22:19:11 <antrik> well, the bit about "morphing into a mainstream OS" is actually a good description of the polarising mindset I mentioned
Jun 02 22:19:36 <antrik> as for them not knowing what people actually want, that just sounds like flamebait
Jun 02 22:20:29 <antrik> from what I've seen, most people tend to be rather happy with systemd etc.
Jun 02 22:21:08 <DocScrutinizer05> sure they are, somebody gave it to them for free and told them it's the hottest shit around and they should be happy
Jun 02 22:21:24 <antrik> (though they *do* sometimes stumble over things that haven't been properly thought through... but I'd just call that sloppiness on the side of the designer rather than any particular mindset)
Jun 02 22:21:54 <DocScrutinizer05> that's not the point. systemd is Hotel California
Jun 02 22:22:52 <antrik> right. again, that's the ecosystem bit I'm concerned about, as opposed to acutal user-facing behaviour
Jun 02 22:23:06 <DocScrutinizer05> I don't care why - and if - some stuff in systemd doesn't work as good as they advertised it. I don't want the whole package, yet they forcefeed it and leave no choice
Jun 02 22:24:36 <antrik> indeed. again, ecosystem :-)
Jun 02 22:25:04 <DocScrutinizer05> it's basically a heart transplant for linux. You hope for it getting better but there's definitely no way back when it doesn't live up to your expectations
Jun 02 22:26:37 <DocScrutinizer05> once you opted in, all your stuff will sooner or later depend omn systemd and you can't get rid of it anymore
Jun 02 22:27:08 <DocScrutinizer05> ~nosystemd
Jun 02 22:27:09 <apt> i guess nosystemd is https://devuan.org http://without-systemd.org/wiki/index.php/Arguments_against_systemd http://blog.darknedgy.net/technology/2015/10/11/0/
Jun 02 22:27:39 <DocScrutinizer05> http://without-systemd.org/wiki/index.php/Arguments_against_systemd#Scope_creep
Jun 02 22:28:02 <DocScrutinizer05> "web server" REALLY NOW?!
Jun 02 22:28:07 <antrik> you are preaching to the choir :-P
Jun 02 22:28:33 <DocScrutinizer05> IP Forwarding, IP Masquerading & Basic Firewall Controls !!??!
Jun 02 22:31:28 <antrik> well, I don't mind them taking up issues that have not been handled well or integrated well so far... I "only" mind the fact that they do it in an all-or-nothing manner, with no room for alternatives
Jun 02 22:34:43 <DocScrutinizer05> yes
Jun 02 22:35:45 <DocScrutinizer05> they do in in an all-or-nothing manner and with a mindset that always implies all other devels were idiots ever since the punchcard got invented
Jun 02 22:35:51 <DocScrutinizer05> see >>Lead systemd developer doesn't understand su, expects it to do something else and then labels it a "broken concept" - su isn't supposed to inherit cgroups or audit, those concepts are relatively new and arrived well after the creation of su. TTYs were originally physical devices so of course su is supposed "inherit" the same device otherwise it would be truly broken. Pseudo TTYs emulate real TTYs so their behaviour is 
Jun 02 22:35:53 <DocScrutinizer05> obviously expected to be identical. su really is just a simple mechanism that calls setuid(2) in order to switch to another user. If he needs to write a new utility to handle scenarios that su was never designed to handle, no problem, but to label it as a "broken concept" demonstrates a lack of understanding of what su actually is.<<
Jun 02 22:37:30 <antrik> heh... nice point -- though IMHO a pretty orthogonal matter
Jun 02 22:38:51 <DocScrutinizer05> it shows the mindset
Jun 02 22:39:03 <antrik> (unfortunately, we see that a lot in and around Linux... I guess that's got something to do with a majority of developers being young and naive. and systemd is not even the worst offender on this score)
Jun 02 22:39:47 <DocScrutinizer05> in a more generic form the statement is "when your usecase doesn't accomodate systemd then something is wrong with your usecase. Systemd is great and there's no alternative to it"
Jun 02 22:41:58 <antrik> indeed
Jun 02 22:42:03 <DocScrutinizer05> ack regarding young and naive (and pretentious) programmers, and that there are other offenders too
Jun 02 22:45:09 <DocScrutinizer05> (offenders) see akonadi ;-)
Jun 02 22:49:01 <DocScrutinizer05> akonadi-virtuoso-nepomuk tracker-miner-fs eating 30 GB of storage when you'd offer as much to it? doesn't matter, please update to the (not yet available) new version of KDE!
Jun 02 22:49:26 <DocScrutinizer05> s/storage/MEMORY aka RAM/
Jun 02 22:49:27 <apt> DocScrutinizer05 meant: akonadi-virtuoso-nepomuk tracker-miner-fs eating 30 GB of MEMORY aka RAM when you'd offer as much to it? doesn't matter, please update to the (not yet available) new version of KDE!
Jun 02 22:52:21 <DocScrutinizer05> kmail 1.9 worked mostly great, then they moved to kmail_with_akonadi aka kmail2 and no way back from this broken crap. The all-or-nothing approach
**** ENDING LOGGING AT Fri Jun 03 02:59:58 2016