UNIX-like Thread (GNU, *BSD, ETC) (78)

1 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

Let us discuss and jerk one another off.

2 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

Why is Linus such a fucking faggot now? Everything he says is controversial and against everything he used to believe in. What went wrong?

3 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>2
Specific examples?

4 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

Today I wrote a bash script..
       V       ∧_∧     
      ∧_∧    (´<_`  )  < I bet you feel like a hacker now..
     ( ´_ゝ`)    /  ⌒i  
 ̄\  /   / ̄ ̄ ̄ ̄ ̄/.  |     
 ̄ ̄| /   ./   UNIX  / . | |
 ̄| |(__ニつ/_____/_| |____
田| | \___))\    (u ⊃
ノ||| |       ⌒ ̄

5 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>2
Fuck you systemd eater, he still knows a rat when he sees one, and pottering et al are the sewage eating antithesis of linux as it is.

6 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>2
let me tell you about the jews

7 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

what distros do you guys use?

8 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>7
i use mint. im a noob so its rather ok for now, except that i constantly run into problems. i consider changing to arch in the near future.

9 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

install gentoo

10 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>2 Because Android is the biggest Linux distro around. World domination at last, just not UNIX.

He doesn't know how the free software ecosystem works, and doesn't care. He doesn't know about static compiles because Drepper forbade it. (GCC needs to die.) Linus only cares about the kernel, not about managing userspace programs. He only initiates them: He no longer maintains git, and hasn't touched Subsurface since it went Qt.

He still insists on not breaking userspace, which is nice, and about reliable maintenance of kernel componentes, which, following the Reiser debacle, seems wise.

11 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>7 Devuan.

12 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>11
You have a QWERTY keyboard and meant Debian, right?

13 Name: a2b2 [Del]

gen2 gen2 g3nt00 gentoo jen two
jenny death jenny death 2jennydeath
2jen
jen2

14 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>13 I used to USE="gentoo", it is in many ways the best, but I spent more time re-compiling than getting any actual work done.

15 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>14
USE this! bends over

16 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>14
I have really been considering trying Gentoo. I see people talking about cutting size down of packages considerably, and improving performance considerably as well, are USE flags as effective as people say?

17 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>16

ROT13'd because I'm tripping some kind of spam filter.

Rssrpgvir ng jung?

Cresbeznapr? Frr uggc://jvxv.tragbb.bet/jvxv/TPP_bcgvzvmngvba#Bcgvzvmvat . Fher, gurer'f fbzr rssrpg, ohg abguvat lbh'yy abgvpr jvgubhg fcrpvsvpnyyl orapuznexvat sbe vg. Gurer hfrq gb or n terng fvgr qbphzragvat gur unez bs pregnva evpvat grpuavdhrf ng uggc://shaebyy-ybbcf.vasb , ohg vg'f qrnq abj. Vs lbh'er ehaavat fbzrguvat jvgu fhpu uvtu cresbeznapr erdhverzragf gung fjvgpuvat --znepu npghnyyl znggref, lbh'yy xabj.

Abg pyhggrevat hc lbhe flfgrz jvgu ohyyfuvg? Cerggl qnza tbbq. Pregnva guvatf yvxr pbafbyrxvg ner hfryrff gb zr naq npgviryl pnhfr ceboyrzf (r.t. pyhggrevat [u]gbc bhgchg), naq rirel bgure znwbe-vfu qvfgeb V gevrq fuvccrq jvgu ab jnl gb erzbir gubfr guvatf. Gung'f jul V hfr HFR syntf.

18 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>17 it's the weblinks that trigger the filter. I'll quote you for convenience.

Effective at what?

Performance? See AHHHHHHHHHH. Sure, there's some effect, but nothing you'll notice without specifically benchmarking for it. There used to be a great site documenting the harm of certain ricing techniques at AHHHHHHHHHH, but it's dead now. If you're running something with such high performance requirements that switching --march actually matters, you'll know.

Not cluttering up your system with bullshit? Pretty damn good. Certain things like consolekit are useless to me and actively cause problems (e.g. cluttering [h]top output), and every other major-ish distro I tried shipped with no way to remove those things. That's why I use USE flags.

20 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

Gentoo also doesn't use systemd, so it already has a leg up on most distros as openrc boots faster and isn't buttfuckingly retarded.

21 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>12
No.

22 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

customizepkg is way better than USE flags. also, being able to just yaourt -Syua and be done updating everything in minutes instead of waiting hours for everything to compile is nice. but then I only ever used gentoo because I had some obscure hardware that wasn't supported on freebsd. actually, what I'd really like is a distribution like arch, but using the minix kernel instead of linux.

23 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

I remember customizepkg. It was neat if you just wanted to modify one thing, but as soon as you started playing with dependencies everything went to hell. USE flags are especially nice since the maintainers do a decent job of unifying them, so I don't have to hunt down whether this package requires me to --disable-foo or --without-foo or USE_FOO=no make or edit a custom makefile, etc. Anything more specific than that can easily be handled by epatch_user. Compiling is sometimes irritating, but it doesn't take hours unless you're doing something silly like compiling libreoffice or chromium with all flags turned on.

I'm not sure why you'd want MINIX though: it's notoriously academic - AST's desire to avoid cluttering it up with code that couldn't be used for teaching examples is partially what gave us Linux in the first place. Even GNU/HURD is more usable these days if you're running vanilla x86.

24 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

I eidnt realize thiw board wqs full of nerds......

25 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

Nerds, >>24-kun?

I think you mean experts.

This is theUNIX thread; you should expect discussions about UNIX and UNIX-related accessories.

If you are looking for nerds, try the anime thread.

26 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

How 'bout that mac huh?
UNIX certified
How much did you neck beards cry when you found out you don't use a UNIX machine

27 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

The only reason BSD and Linux are not UNIX-certified is because nobody could be bothered to pony up the money for the official certification. Functionally they are UNIX, which is what matters to developers and hackers.

That Mac is nice, too, but I would rather not run my server on it.
It is still vulnerable to the shellshock exploit.

28 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>27
I remember reading that I think one of the original UNIX guys (Thompson? Kernighan? Ritchie?) said that Linux was essentially UNIX.

Also, UNIX is Unix now.

29 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

Why the fuck do things like htop require lines in fstab on FreeBSD?

30 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>26
Why would anyone care? It's called 'GNU's Not Unix' for a reason.

32 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>5
systemd hate is a bunch of FUD started and fostered on tech websites with an obvious bone-to-pick such as zdnet.

33 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>32
Systemd hate is a bunch of rationally sound arguments started and fostered by people who don't like the kernel being fucked up just to fix bugs in a userspace program whose developers are too stuck up to fix their own bugs.

34 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>33
Really? Because all I see is a bunch of outsiders getting stuff like "Poettering and co. are unwilling to patch bugs that affect userspace" just from reading a tasteful selection of hand-picked e-mails from Phoronix.
The reality is so much milder. It's the same thing with the so-called drama with Linus saying stuff like retro-actively abort people. He's a mild person who posts over-the-top rants and people treat it like he's a dictator that's willing to crush every self-esteem on his mailing list.

35 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

systemd = nsa plot to backdoor lunix by making everything depend on it and then pulling some "goto fail" type shit

36 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>30 GNU really is something other than UNIX.

UNIX is composed of small, simple, specialized tools that are realiable, easy to verify, and easy to replace, and which nonetheless all work together, and from which arbitrary complex systems can be constructed, be it for maintenance, hacking, or rabid prototyping. Every part of the system is exposed, and UNIX gives you enough rope to shoot yourself in the foot.

GNU is a quasi monolithic bunch of interdependent software. That the GNU tools are small, specialized programs is because they copied east coast UNIX, but they have their own deliberately incompatible idiosyncrasies. You can see that best if you look at the GNU Compiler Collection, not just yet another C compiler, but one cross compiler for all languages, deliberatly hard to maintain, that absolutely requires its own libc, which is also choke full of non-standard extensions.
GNU embraced the architecture of UNIX, and extended it, no prize for guessing what the next step in their plan was.

Meanwhile, BSD has also replaced every part of UNIX™ with a free alternative, without obsctructing hackers from maintaing, contributing, or replacing parts of it. Including the kernel.
Linux is more successful than BSD because it was painted red, and therefore faster. It really was faster than 386BSD, and had better hardware support, but since then the BSDs have been catching up.
And Linux is not GNU either.

37 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>34
Phoronix? No, just reading bug reports should be enough, between that and the fact that systemd shouldn't need to have utilities like udev and logind depend on it which makes it bad practice it should be reasonably obvious that it was never designed to have scope or accountability. It could be a very good system at some point, but that would require people working on it who don't want it to be systemd for systemd's sake, it would require developing it within a scope and having it be modular instead of this giant mess that has arbitrary requirements and a lot of lazy code.
A lot of systemd hate may come from people who don't understand it, but it does deserve a lot of the hate it gets.

38 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>37
Basically systemd isn't a bad idea, it's a bad implementation by people who are to invested in the idea of it. It's certainly too immature to be implemented at this point.

39 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>38 I disagree.

I have used systemd ever since it entered SID (back when wheezy was testing). It has nice, consistent command line tools and usefully pretty-printed output. It replaces the System V runlevels with more intuitive targets. And it is an absolutely awful idea.

Where tools like insserv compute dependencies and adjust the order of start and parallelisation whenever a job is added, systemd does it during the boot. And it uses lazy evaluation, which is a good thing in principle, because it means that only jobs that need to be started are started. And systemd starts them only as soon as they are called for, which means that as soon as a connection is made to port 443, systemd launches the web server, which launches its modules, one if which is PHP, which requires the fastgci spawner, and also the MySQL database, which upon launching performs a self test and clean-ups and replays transactions as needed, whereupon it may or may not be ready to serve the request of the CMS written in PHP to refrech the page cache for the apache, so that the reqeust on port 443 can finally be served, long after it has timed out. That is called socket based activation, and it makes your boot so much faster, after backtracking through the dependencies as late as possible has made that moot.

But at least it is not sysvinit with its abominable inittab.

One nice feature about systemd is that it communicates via named pipes. When the pipes are out of sync, or otherwise break, it only requires restarting systemd to re-initialize them, and with systemd as PID 1, that means rebooting the entire system. Which you can do easily with systemd, as long as the pipes are in sync. Fortunately, the pipes breaking only happened on my system a couple of times. I really appreciate how systemd made the reset button shine, and well worth the investment after I hadn't used it for almost a decade. I don't know what I would have done had I tested systemd on a remote system.

40 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>39
Every system i've run that uses systemd boots in under a second (including one that runs about a dozen different daemons on a Raspberry Pi, which was a remote (about 200 miles away) system for 7 months (recently relocated to facilitate some GPIO experiments) with 0 issues.
I've never seen a single thing use that "socket based activation" nonsense.

41 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

Why don't any of you plebeians ever think of using Plan 9 from Bell Labs?

42 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>40 That's nice. My test system was a single core x86. My Raspian box never had systemd installed, and never even wanted to. What OS is your Raspberry running?

>>41 What hardware are you running your Plan9 on? What do you think of Inferno?

43 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>42
What OS is your Raspberry running?
arch linux.

>>41
you are using 9front, aren't you?

44 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>43

>arch

I tried Arch once. It could not connect to a WiFi with a whitespace in its SSID.
Probably for this reason, I prefer Gentoo to Arch for rolling release from source, but I'm sure it has improved substantially in the meantime.
For example, another user had this problem: ansuz.sooke.bc.ca/entry/254
This issue of Dr Skala has recently been resolved by Michael Biebl of Debian (and Ubuntu). That is progress.

45 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>44
I am in a similar position to you, but to be fair to Arch that sounds more like a wicd issue than anything else. I assume you were using Wicd because it seems to be the only major alternative to Gnome/KDE's offerings, and because it has a plethora of truly bizarre bugs like that.

46 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>45 I am indeed using wicd. It is more reliable than NetworkManager. The issue was, however, that there was no way to put a SSID with a whitespace in Arch's network config file. That was before I even installed a GUI. Servers should not need one anyway.

47 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>46
Servers should not need wifi, either. Just buy a Cat. 8.2 cable and use 40GBASE-T.

49 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>46
What exactly is unreliable about NetworkManager? It works perfectly well on gentoo without systemd and I have kept it running without interruption as long as my router has been on.

50 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>49 If I knew, I would do something about it. All I can say is that NM often wouldn't work for me, but I never had problems with wicd. It is also easier on dependencies and RAM. Maybe it's a Debian issue.

51 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>50
I've never had issues with NM on arch, but then I've never tried to use it on my server because it's plugged in with an Ethernet cable (only Cat. 6 and 100BASE-T, tho) so there's no need for wifi.

52 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>50
Probably a frontend issue, does it still happen with the tui interface?

53 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>52 It probably is a front-end issue, but in the end I found it easier to configure the network in /etc/network/interfaces than with any of the NM front-ends at the time. Since that time, I have never needed to touch NM ever again, because wicd just works.

54 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>26
Lawyers insisted on an uppercase UNIX trademark.

55 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>54
That's because UNIX™ certification has nothing to do with real Unix. They wanted to be able to collect bribes for certification without actually guaranteeing anything.

56 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>55
no I mean back in the 70s

57 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>56
Lawyers didn't know what computers were in the 70s. Most of them still don't.

58 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>57 The first spam on USENet originated from a law firm in the eighties.

>>54 Ken Thompson likes capital case.

59 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>58
oh, sure, inventing spam totally seems like someone who understands computers would want to be known for.

60 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>55
UNIX certification has plenty to do with real Unix (although, indeed, it has little pragmatic worth), and the reason Linux is not UNIX-certified has nothing to do with lack of money, but because it follows its own standard.

61 Post deleted by user.

62 Post deleted by user.

63 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

The ratio of discussing to jerking one another of in this thread is too damn high.

64 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>60 Linux is still POSIX-compliant, and I do not doubt that it would pass UNIX certification if someone would bother to pay for it. And then, a month later, a new version of Linux would be released.

65 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>64
Most of POSIX is about userspace stuff, so Linux doesn't really have much to do with whether a particular system is POSIX-compliant or not.

66 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>65 Most of the UNIX certification is also about user space programs. Obviously we are not talking about just the naked kernel.

67 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>64
No, it wouldn't. You can look up where Linux and UNIX deviate, actually.

68 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>67 Where?

69 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>68
On Google. I once found a document that details all the differences but I can't seem to find it again. But you can look up some noticeable differences if you look hard enough, for instance: wki.pe/STREAMS

For the record, the high cost involved in getting UNIX certified is not because of the certification per-se, but the significant reworking that Red Hat and the LSB would have to submit themselves into to actually be fully compliant. There's absolutely no point in doing it, since there's already a standard, it's the LSB.

70 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>69

>For the record, the high cost involved in getting UNIX certified is not because of the certification per-se

I was going to disagree, but you appear to be right. Seems the Open Group wanted Linux to be a certified UNIX, and the LSB working group wanted to comply, but nobody else cared.

And even the LSB is ignored by most distributions. It seems less to be a common standard for Linux distributions and more a measure of binary compatibility with RedHat. No wonder nobody cares about that.

72 Name: intelligent man of enlightenment [Del]

>>71
as much as I'd agree with you on how vital LSD is in order to provide hektik skids for automobiles, this isn't the thread about vehicles

This thread is about the UNIX family and other related UNIX-esque operating systems.

73 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>72
What does lysergic acid diethylamide have to do with vehicles?
The connection to RedHat developers seems pretty clear...

74 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

They say it is no coincidence that both LSD and BSD came from the same university…

75 Name: intelligent man of enlightenment [Del]

>>73
hello there, you seem to confused about the difference between 'Lysergic Acid Diethylamide' and 'Limited Slip Differentials', the latter of which is crucial for ripping fat skids in RWD cars like a Ford Mustang or a Mazda Roadster.

fuck Red Hat though

76 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>75 Hear, hear!

77 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

I once thought it would be nice to work for Red Hat, since I would have a chance to get paid for contributing to FOSS, and specifically to Linux. But when I looked into it, it turns out that most of the (publicly visible, anyway) stuff RH does is not really the sort of thing I want to do, and their goals seem to be not really the sort of goals I want to help forward.

Perhaps I have observed a vicious cycle. If so, I worry, because the average kernel dev isn't getting any younger, and Red Hat's support contracts will endow it with financial-alchemical youth for a good many years.

78 Name: キタ━━━━━━━━( ・∀・)━━━━━━━━!!!! [Del]

>>77 If it is any consolation, RedHat is not the only Linux company. RedHat is to Linux was ADA is to programming languages.