How is GNU software worse

if it has more features? It's not like the old features go away.

Because it's written in C, and in C a program's shittiness grows exponentially with its complexity.

Have you ever seen RMS? He's fat and bloated, just like GNU code.

People like to claim GNU software is substandard but fail to specify anything instead referring to uncertainty and fallacies to make their point.

I agree OP. A larger binary doesn't mean slower execution.

>>2
that explains why git is so bad

>>1,3-5,7-1000 >>2

[lol@lol ~]$ valgrind cat
==2190== Memcheck, a memory error detector
==2190== Copyright (C) 2002-2011, and GNU GPL'd, by Julian Seward et al.
==2190== Using Valgrind-3.7.0 and LibVEX; rerun with -h for copyright info
==2190== Command: cat
==2190==
^C==2190==
==2190== HEAP SUMMARY:
==2190==     in use at exit: 36,863 bytes in 1 blocks
==2190==   total heap usage: 31 allocs, 30 frees, [b]40,544 bytes allocated[/b]
==2190==
==2190== LEAK SUMMARY:
==2190==    definitely lost: 0 bytes in 0 blocks
==2190==    indirectly lost: 0 bytes in 0 blocks
==2190==      possibly lost: 0 bytes in 0 blocks
==2190==    still reachable: 36,863 bytes in 1 blocks
==2190==         suppressed: 0 bytes in 0 blocks
==2190== Rerun with --leak-check=full to see details of leaked memory
==2190==
==2190== For counts of detected and suppressed errors, rerun with: -v
==2190== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 2 from 2)

GCC > Clang

Forte > Clang > GCC

Because it's a clusterfuck of complexity. Such as 4k+ LOCs to write a fucking "ls" command.

The GNU folk is simply batshit insane. They've completely missed the point of C, UNIX, programming and maybe life itself. They masturbate with maintaining ports of "ls" for 18-bit architectures or pre-standard compilers no one gives a fuck about anymore.

More features do not mean better usage, better user experience, more productivity. Specially when the word "feature" is used, which almost imply some degree of uselessness. Software must have "functionality". It must serve a purpose, a single one if possible, without adornments. GNU software is so substandard that they create new software to assess the problems other pieces of software they write create.

Look at >>8, a single `cat` allocating 40KB in the motherfucking heap. What the fuck?

This is why no one wants to maintain GNU software. There's no single reason a sane person would dedicate his time to decypher thousands of lines of batshit ill-indented insanery in order to make a fucking directory listing, or file copy, or whatever.

>>11
They aren't ``missing the point'' of C and UNIX any more than Microsoft are--they just choose to do something different entirely. They aren't going for ``the UNIX philosophy'' or lean code or tradition, they're going for large feature sets and wide market share with software that works on today's computers. I even emailed Stallman about this and he said he isn't big on the UNIX philosophy, he's fine with extensions (why wouldn't he be? GNU-only options encourage you to use GNU), and that the Emacs icon was once an overflowing kitchen sink ``for a reason''.

Complaining about GNU not meeting a philosophy it simply isn't trying to meet is dumb. UNIX isn't the measuring stick all operating systems are set by. GNU's Not Unix.

GNU is poo

C is such a crap language. All C programs are full of #define's and other shit.

Worse is better.

Having a program do what you want it to is terrible!

Everything that can be done using grep, sed and awk must be!

NO EXCEPTIONS

awk is the greatest programing language in the world!

>>15
I don't think so.

>>12
Yours is a good explanation, I didn't know beforehand Stallman was not such an UNIX fan.

Nonetheless, I've never seen GNU running under non-UNIX like systems. That wouldn't even make any sense, since core GNU software is UNIX-like userland. Being such, I can't see why they would do this if they were not bound to UNIX philosophy. If they know better than UNIX, why attempt to just "mock" it? I don't think that the existence of any pre-installed user base is an excuse.

My criticism is rather focused on that part of their software, the part which complicates UNIX. Not everything labeled GNU, or under some variant of GPL.

The fact is, if they're really "going for large feature sets and wide market share with software that works on today's computers", surely they got it wrong (again). Twenty-five command-line switches for ls or cat doesn't make these commands specially featureful. And, at least on the desktop market, GNU systems' share is rather ridiculous.

Finally, I totally agree: Microsoft is not doing any better at all. That has not been said anyway, and that is not the point being discussed.

>>17
I'm not that guy but what he's saying is common knowledge amongst anybody who knows the history of free software.

I never used Unix (not even for a minute) until after I decided to develop a free replacement for it (the GNU system). I chose that design to follow because it was portable and seemed fairly clean. I was never a fan of Unix; I had some criticisms of it too. But it was ok overall as a model.
http://stallman.org/stallman-computing.html

>>14
That's bad practice and the fault of the programmer. I hate to say it, but the cat-v.org/suckless/Plan 9 types have a point when they prove you can make software without fifty million preprocessor directives and macros, and it still works. That's what C was originally about.

>>19
C is crap. Back to the grave, Uriel!

>>19
C is only marginally less shitty with GNU m4. It's still primitive, barely-expressive shit though.

>>11,12
Bloating ls and grep with pretty colors and such pays dividends for interactive use, so I don't mind it. However, I do agree that the extensions to things like true, cat, or awk are actively harmful. Maybe if the GNU Project had started three years later or Larry Wall had released Perl three years earlier we wouldn't have this problem.

>>19-21
The most important thing about C, I think, is knowing when to quit. If you aren't writing system software you can probably do a lot better than to use C. The historic problem with this is that Unix shells, quite frankly, blow huge chunks for doing anything complicated. We have Perl and Python now, so this is less of an issue, but the old smell still lingers.

Would the world have been better off if everyone had been able to agree on a decent high level language? Possibly. Shell one liners are hard to give up, though.

Bloating ls and grep with pretty colors and such pays dividends for interactive use, so I don't mind it.
Of course this is only true if the default colours make sense. Dark blue on a black terminal screen? WTF were they thinking.

Since I can never remember what the colours mean anyway, I just alias it to add -l and disable any colouration. I'm also not fond of my directory listings looking like some kid's drawing.

>>23
Yes, the default colors suck.  xterm et al. ended up making their dark blue lighter in response to this.

alias ls='ls -F' is a nice compromise if you don't like the colors.  It's more compact, and you will never need to repeat the listing because you can't remember whether an entry is a file, directory, symlink etc.

GNU is worse?

Hah.

It might be bad, but it's still better than anything else.

I dont think GNU is bad...

maybe some misconception around...

>>11
40KB isn't a sign of anything odd because that could be its read buffer. 31 separate allocations is.

(Anonix cat has a 64K buffer.)

>>27
how is Anonix. is there still a repo for it somewhere?

>>28
At the REchan.

>>26
You must not have much in the way of critical thinking skills.

>>http://stallman.org/stallman-computing.html
The most powerful programming language is Lisp. If you don't know Lisp (or its variant, Scheme), you don't appreciate what a powerful language is. Once you learn Lisp you will see what is missing in most other languages.
When you start a Lisp system, it enters a read-eval-print loop. Most other languages have nothing comparable to read, nothing comparable to eval, and nothing comparable to print. What gaping deficiencies!
Lisp is no harder to understand than other languages. So if you have never learned to program, and you want to start, start with Lisp. If you learn to edit with Emacs, you can learn Lisp by writing editing commands for Emacs. You can use the Introduction to Programming in Emacs Lisp to learn with: it is free as in freedom, and you can order printed copies from the FSF.
To study Scheme, and programming through Scheme, I recommend Structure and Interpretation of Computer Programs, by Abelson and Sussman.
Please don't buy books from Amazon!
Even Stallman admits C is substandard.

Stallman is substandard.

>>30
Neither do you.

People blame GNU software for being something that it is not. GNU's foremost goal is to provide a complete and free system for all people who care for their freedom. At the time the project began, the systems that existed were transforming from being free to non-free. GNU happens to be an implementation of Unix simply because Unix is a good basis to begin as a general system; if there was any other popular system that was worth people's time, RMS would have modeled GNU after that system. It's obvious that GNU's implementation doesn't adhere to the Unix philosophy but some other philosophy. I interpret the design as something like: be useful to as many people as possible.

>>29
www.rechan.com

is this chinese? I don't speak it...

>>34
rechan.eu.org/prog/

if there was any other popular system that was worth people's time, RMS would have modeled GNU after that system
So why isn't it based on VMS?

>>36
I guess because RMS had access to Unix machines and not VAX machines when he started GNU.

access to Unix machines and not VAX machines
That makes about as much sense as saying ``access to Linux machines and not x86 machines''.

>>36
Because ``GNV" would be too confusing for the target audience for GNU.

>>39
target audience for GNU
People who don't hate their privacy and freedom?