Linux criticism

then why is solaris such a massive piece of shit?

>>61216168
>Solaris
>Developed by experienced UNIX devs and engineers

>Linux
>Developed by a Finnish student

>>61216221
then why is Open Indiana such a massive piece of shit?

>>61216199
>>61216176
What about screenshot?

They're right. Linux is awful but it "won" because it was free and open source from the start and gained a lot of momentum because of that. Now it's hard for other options to compete because of hardware support.

Google's Fuchsia might be able to compete someday because Google is throwing cash at it and more and more people are realizing that Linux is actually really bad and the quality of code is astoundingly low.

>>61216176
Solaris was incredible on servers with supported hardware.

We had this thread las week Ranjeet.

>>61216251
Wow! Who'da thunk?

>>61216257
Won't Fuschia mostly be for mobile devices though?

File: GNU+car.jpg (244KB, 1024x681px) Image search: [Google]

244KB, 1024x681px

>>61216168
>and the Linux kernel (scheduler) suddenly looks like the Millennium Falcon.
>Real fast, but held together with duct tape, and ready to explode at any minute.

That's what makes it FUN

>small focused project
vs
>large project that has more hands on deck than a large company's sack of shit project without any of the direction

For what it is Linux is fine.

>>61216257
>Google.
>The same Google that degraded Dagger to Dagger 2.
>The same Google who mutated Angular into Angular 2.
>The same Google that bloats, breaks and modifies the Android APIs in every release.
Jewgle can't produce clean code at all. They have a rule that every library or API must be as complicated and dirty as possible.

>>61216389
Rebels didn't blow up planets for the lulz.

>>61216257
>indians and post-"""global economy""" white people writing a kernel
nigger, most of these people don't know how to use pointers without segfaulting, if you expect googlers to get a kernel right you're out of your mind.

>>61216367
>They have a rule that every library or API must be as complicated and dirty as possible
As a developer who has been integrating Google products for a number of years, this made me laugh a lot, then become very sad. You are absolutely right.

>>61216425
They hired Bell Labs plan9 people, they came up with google go, and who have been involved with unix projects like X
https://en.wikipedia.org/wiki/Rob_Pike
https://en.wikipedia.org/wiki/Ken_Thompson

They probably have other smartasses than just these two.

>>61216460
According to a 2009 interview, Thompson now uses a Linux-based operating system.

>>61216168
>BOOOHOOO even though I spam my FUD about Linux daily people see through it because they know all the biggest websites and a billion phones are running linux every day

poor schlomo, go back to you paymasters att micro$shaft

Isn't this old news? I remember myself reading this (or similar) article couple years ago

>>61216503
>You must never critizie the party
>If anyone critizies the party they are automatically terrorists
Nice logic bro. So is Linus Torvald and other 'Linux-people' working for micro$shit too?

>>61216168
Linux now support big iron enough well so oracle can recommend it on anywhere. In 2010 it maybe was a problem and Kolivas couldn't solve it alone.

>>61216460
>X
not a project you want to cite as anything 'good', and i also don't think either of those people have ever touched x11
>go
while i like this language, it's not really got anything to do with kernel development.
as far as plan9 goes, i don't know about it's kernel, but i don't honestly believe that anyone who made plan9 would be working with google to make a kernel for an OS mostly written in web developer languages.

Yeah clear design is great. Then you run it and it performs like shit and you like it less.

>>61216504
/g/ is full of reposting

>>61216520
They're perfectionists and humble to boot. The kernel is likely to have some shitty code, as most huge long running codebases have, but you and me both know that if we could peek at your employer, micro$hafts, kernel it would be orders of magnitude worse

>>61216536
speaking of graphical shit, apparantly Wayland is garbage too, it's just hyped garbage. Maybe it's less shit that x but it's still shit. I don't know jack about programing, just read about it.

>>61216562
>Employed by Microsoft to shitpost
I wish

File: 1470592918333.jpg (85KB, 680x680px) Image search: [Google]

85KB, 680x680px

>>61216168
> Star Wars references

>>61216582
Wayland and X shouldn't be compared.
X is a display server, Wayland is just a protocol.

On wayland it's the window manager (in x terms) or wayland compositor (in wayland terms) that are supposed to handle stuff like inputs.

That's why in the era of Wayland you won't see as many window managers as you get on X.
Because each and every one of them requires to reinvent a certain set of features to work

>>61216582
it is
wayland's spec is written in fucking xml.
anyone who thinks that's a good idea is not smart enough to write a display protocol.

>>61216582
Yeah, well, you can thank the know it alls out there who carried on about X being shit, the Wayland project got dumped on us as a work in progress. Big surprise! An incomplete project isn't working out so well.

File: 1497763830756.jpg (155KB, 652x754px) Image search: [Google]

155KB, 652x754px

>>61216168
The GNU Operating System with linux added as its Kernel is unfortunately still the least shit yet functional operating system

>tfw there will never be a beautiful and functional OS

>>61216645

apparently wayland *is* the spec and the implementation is just a library
meh, havent looked into it further but it does seem a little weird that there are tiny working X window managers with like 1000 lines of code... I mean it sounds pretty much simple enough to me, why change to wayland just because of some autistic linux reason like "ohhh X contains font rendering code that nobody uses anymore"... so?? Who gives a shit right??
It's called having backwards compatability, it's not like this is hurting anybody.
I also heard that wayland actually has a drawback, namely that you cant view remote applications anymore, which was a feature of X. Not that many ppl have used it, but it does sound somewhat interesting.

>>61216592
>doing it for free

>>61216645
>spec in xml
What?

>>61216680

oh and to add something, how is it that it takes 10 years to make a replacement for X?
sounds like the replacement is gonna suck just as hard as anything it replaces if it takes 10 years to write

>>61216460
Go is objectively awful and the Unix that was developed by Pike and Thompson was a mess with more security problems than Windows XP.

>>61216717
dis

>>61216680
>why change to wayland just because of some autistic linux reason like "ohhh X contains font rendering code that nobody uses anymore"... so

Getting rid of tearing seems a pretty good reason to me

>>61216536
>mostly written in web developer languages

The overwhelming majority (more than 90%) of Fuchsia is written in C++.

>>61216741
what's your point?

>>61216562
Lots of parts of the Microsoft kernel have leaked over the years and the code was very well done. Easily better than most of the Linux kernel.

A company worth almost a trillion dollars has no problems hiring decent kernel developers.

>>61216645
You clearly have no idea what you're talking about.

>>61216731
It always amazes me the amount of love the old "Unix gurus" get. These are the people that produced code that had an average of a buffer overflow every fifth line.

I love Theo and a lot of the OpenBSD community simply because they call these hacks out. Now two of the beloved hacks have gone on to invent a dogshit programming language that ignores the last 50 years of language research.

>>61216788
2 posts, 0 arguments
c++ is not a good programming language and nobody outside of academia and web development companies thinks it is, therefore, it is a web developer language (especially considering academics don't like it any more for not having "safety" or something. personally i just think they want monads but can't find the doublespeak to ask for them.)
notice the difference between "web developMENT" and "web developER".

>>61216755
Google isn't writing a kernel for an OS that is written mostly in web developer languages, and the people that are working on Fuchsia are far more knowledgable than anyone that has ever worked on Plan 9.

>>61216837
You really need to go back to the phone threads or wherever you came from.

>>61216861
I think he's doing it on purpose now.

>>61216826
>i love * openbsd community
>these hacks * produced code with * buffer overflows
>GO IS SHIT **OBJECTIVELY** AND IT IGNORES THE LAST 50 YEARS OF "LANGUAGE RESEARCH"

i wonder what it's like to know constantly that you're wrong, but still spout your stupid fucking opinions as if they were facts.
cite literally anything so that you can at least pretend you have an argument

>>61216849
please recite to me the programming languages listed as those that this is written in
https://en.m.wikipedia.org/wiki/Google_Fuchsia
and then honestly tell me these aren't languages made for the average web developer

>>61216861
nice argument

>>61216775
>Lots of parts of the Microsoft kernel have leaked over the years and the code was very well done. Easily better than most of the Linux kernel.


[CITATION NEEDED]

File: 1498942786945.jpg (9KB, 211x238px) Image search: [Google]

9KB, 211x238px

>>61216874
> C++ is a web developer language

>>61216251

they need to listen to terry davis

> We want to remove code not add code.

>>61216903
didn't realize this was the "no arguments only memes" thread
please post something of value to offset your idiotic post

>>61216168
> compared some OpenSolaris kernel code to Linux code
> and was surprised at the difference in code quality
At least you didn't say it was a Windows code.

>>61216884
>then honestly tell me these aren't languages made for the average web developer

Basically the only thing written in anything other than C++ are language bindings. This is simply to let other people write applications and utilities in the language of their choice. The kernel and almost all the low level tools are written in C++.

>>61216168
>opensolaris
>made in 2008
>computing was pretty much developed by then, as a result, careful planning could be carried out about it's future expansion

>Linux
>initial release was in 1991, 25 years ago
>several technologies have developed since then, meaning the devs had no way to plan much for the expansion regarding new future tech

Give opensolaris 25 year, just like Linux and maybe you'll start to see the similarities to the present linux

>>61216619
but libinput is what handles inputs on wayland

thats not the job of the compositor

>>61216924
>kernel written in c++
hoh boy
>only thing other than c++ is language bindings
definitely not what fuchsia.googlesource.com and github.com/fuchsia-mirror seem to imply if you do the bare minimum required searching
they also seem to think that google isn't quite retarded enough to write a kernel in c++, but i'm sure you know better.

>>61216928
>opensolaris
>made in 2008

That was basically a name change. Solaris has been around for a very long time and is basically a continuation of Unix System V Release 4 and SunOS. Both of which are older than Linux.

>>61216168
>look mom, I posted it again xDDDD

>>61216968
>github.com/fuchsia-mirror

Looks like all the non-trivial projects are in C++. The only non-trivial part that's not in C++ is Flutter.

>>61216991
>their fucking kernel, entire UI, and package manager
>trivial
sure do hope you don't actually believe this

>>61216257
Con develops a CPU scheduler FOR Linux, you retard; in fact he's developed 3. Con uses Linux and is referring to the Linux scheduler in this particular article, and he's not wrong; CFS is a massive hairy piece of shit and it's only begun to improve recently.

>>61217036
>Con develops a CPU scheduler FOR Linux

I know that and never said otherwise. Might want to work on your reading comprehension.

61216680
>tiny working X window managers with like 1000 lines of code...

Those tiny window managers require the whole Xorg stack

>some autistic linux reason like "ohhh X contains font rendering code

no the reasoning is to improve security by not allowing every process in xorg to see your screen/clipboard/keystrokes

>It's called having backwards compatability

Xwayland

>wayland actually has a drawback, namely that you cant view remote applications

fedora is actually working on a api for wayland that has remote capability i just cant remeber the name atm

>>61217045
He might have re wrong post.>>61217036

File: google-fuchsia-magenta-muh-bsd-sources.png (234KB, 1344x708px) Image search: [Google]

234KB, 1344x708px

>>61216991
>MUH FUSCIA

google cant make anything by themselves

>bought modified linux distro to turn into android with the help of oracle
>chrome based off gentoo
>buys youtube
>fuchsia steals bsd code

MADE BY GOOGLE(TM)

>>61217018
>>61216991
Fuchisa is a microkernel so it literally doesn't implement much in the ways of a kernel aside from the absolute fundamentals such as PID handling. Regardless it was obviously designed so that devs can build applications that interface directly to the kernel with higher level languages giving them a lot more flexibility but a lot of work as a penalty.

>>61217045
Yes, and you took his excerpt out of context to make the stretch that the Linux kernel code was shit. If you would like to show me a piece of kernel code which you yourself find absolutely unusable come and file a bug report and maybe me or my seniors at RHEL can fix it, but don't shill such an implication without some evidence and using Kolivas's words as charge against Linux.

>>61217126
>me or my seniors at RHEL

You're not fooling anyone NEET.

>>61217126
>fuchsia is a microkernel, so it doesn't implement much in the ways of a kernel
that isn't what "microkernel" means.
it isn't just "small kernel".
you still need everything a monolithic kernel has, but you run less in ring0.
and, while i'm sure you know a lot about it, if you think that makes fuchsia's kernel "trivial" you're a fucking moron and there's no way to sugar coat it.
the bare minimum a kernel will have is still a filesystem, a bootloader, a scheduler, and the filesystem and process trees.
all of these have to be implemented in what basically ends up being assembly, and this doesn't count drivers for extremely crucial parts like the mmu and video, amongst drivers that most people also consider crucial to actually have a computer

>>61217117
>fuchsia steals bsd code
How to steal something what is gratis in the first place? Fuwhatever is a research project, so i don't care.

>>61217230
No.
Microkernels have no such bloat.

>>61217273
"yes"
a "kernel" that boots and nothing else is not a fucking kernel, no matter what your professors told you.
and since we ARE actually talking about a real kernel, you can look for yourself to see that it does in fact have everything i listed as part of it, it just doesn't have everything in ring 0.
real world kernels are going to have everything any other kernel has (obviously not drivers), just in different configurations.

I've already tried BSD as a desktop OS and I'd really like you shills to fuck off.
Anytime.
Hopefully soon.
God fucking damn it.

>>61216608
this

>>61217376
OP here, I use Linux you dumbie

>>61216257
>Solaris was incredible on servers with supported hardware.
>supported hardware

This is key. Linux is trying its best to support a million drivers making no effort to support Linux whatever. Solaris lets driver programmers check its API.

And that also explains why Linux won. Most common hardware is at least somewhat compatible.

>exactly recall seeing this thread a year ago
Well, it's summer after all...

File: gnu-hurd-logo.jpg (88KB, 830x360px) Image search: [Google]

88KB, 830x360px

Microkernel to the rescue.

File: e79dfe3ce2e4.jpg (220KB, 800x600px) Image search: [Google]

220KB, 800x600px

That car isn't fast and it was designed intentionally to be shit. This pile of dicks is a much better example of duct tape going fast.

>>61217685
This should point to >>61216337.

>>61216168
Linux is junk. No one uses it. Linux desktop literally never. Thank you that is all.

>>61217685
>>61216337
ya'll ain't seen notin
https://www.youtube.com/watch?v=H6Ni2gPo5Us

>>61217685
>>61217717
>rx7
>pile of shit

wew lad

>>61216176

It wasn't.

It was crippled by being the property of Sun, and then its fate was sealed when it became the property of Oracle.

Illumos is the open successor to OpenSolaris, but Linux won by the time that came around, and there's no going against network effects.

>>61217435
>dumbie
>is the b still silent? because dummy is a word, you know?

>>61220040
Oracle is like the devil, it ruins everything it touches. If java wasn't under GPL and so widely used they would have made it proprietary again years ago.

File: 1498913439792.png (134KB, 1309x229px) Image search: [Google]

134KB, 1309x229px

>>61217117
_BSD_SOURCE being defined doesn't mean they stole BSD code. _BSD_SOURCE is a feature test macro.

>>61216176
Whatever was in the end of Sun and later, Oracle, goes to shit.

>>61216176
>slowlaris

It's big and impressive looking but it can't do the Kessel run in 12 parsecs.

>>61216176
this

>Real fast, but held together with duct tape, and ready to explode at any minute.
As is the GNU/Linux way.

So study that code and use the new knowledge to improve Linux, because Solaris as an OS isn't really going anywhere.

>>61216257
(((Google)))

File: 1458542400677.jpg (39KB, 900x900px) Image search: [Google]

39KB, 900x900px

>>61216337
Not fun when it corrupts your SMART info and bricks your HDD.
This happened to me yesterday.