FSF Issues GNU/Linux Name FAQ

jdavidb writes "The FSF has issued a FAQ about why they believe you should say "GNU/Linux." Surprisingly long." They're certainly... thorough.

Non-GNU Linux

[sfraggle]

I have a small linux distribution for the Psion Revo [alkali.org]. Interestingly, I can name this just Linux (not GNU/Linux) because it contains no GNU software. All the normal GNU base utilities (glibc, gnu text/shellutils, bash) have been replaced with small embedded replacements (uclibc, busybox). So I can leave off "GNU/" and I am still correct.

Re:Why are they so desperate ?

[Anonymous Coward]

Its actually the other way round: GNU was around long before the Linux kernel was written. Essentially people have taken the GNU system and renamed it to Linux.

Pronounceability is cultural

[yerricde]

"Linux" surely is more pronouncable than "GNU"

Are you sure? In some human spoken languages, the "gn" cluster is considered "more pronounceable" than the "ks" cluster. What's pronounceable is what you've been brought up with. Yes, speakers of English are at an advantage vs. French speakers at learning the consonant clusters of Russian because English speakers are used to clusters, but it's hard for anybody who didn't grow up in southern Africa to learn to make the hundreds of click sounds that typically start a word.

Re:Why are they so desperate ?

[gowen]

Linus makes a better posterboy (my words) than RMS does. He's not a raving gun nut

No, no, no. It's ESR who is a gun nut not RMS. Someone less likely to dig firearms than RMS it would be hard to imagine. ESR, however, is bald as a coot which explains his need for penis proxies...

Re:Why are they so desperate ?

[gorilla]

He's not a raving gun nut

I think you're thinking of ESR who is the self described [tuxedo.org] gun nut. RMS has certainly been known to shoot guns on occasion [crynwr.com], but he's not known as an advocate.

Re:Non-GNU Linux

[quigonn]

You don't need the glibc to compile and link the kernel. The Linux kernel comes with all the functions it needs.

Political views

[vlad_petric]

We strongly disapprove of his political views, but we deal with that disagreement honorably and openly, rather than by trying to cut him out of the credit for his contribution to the system.
...

If you free that Perl simply cries out for mention, and you want to write GNU/Linux/Perl, go ahead.
...

Should we say "GNU/BSD" too?
BSD systems today use some GNU packages, just as the GNU system and its variants use some BSD programs; however, taken as wholes, they are two different systems that evolved separately.

jeez ...

The Raven

Re:Hmm, not so sure about the BSD bit

[gorilla]

The first BSD release was in early 1977 [oreilly.com]. This wasn't a full OS, it was really a set of patches & a Pascal system. The first bootable system would be in late 1979, or late 1980, depending on what you consider already having a Unix license. From 1980 to 1992, there was a continual upgrade of BSD code, which meant that the original Bell Labs code was almost all replaced, when there was the lawsuit over ownership which cause BSD to stall for 3 years or so. Finally in 1994, the remaining Bell labs code was stripped out or replaced, and the first freely distibutatable version of BSD, one where you didn't have to buy a license from AT&T in order to run it, was created. However, if the FSF inspired BSD, then there needs to be a time machine involved there somewhere.

Re:I call it linux, here's why, in a few short poi

[Anonymous Coward]

> What the .. are you talking about? I fail to see either where GNU "failed" or Linus succeeded withough them. GNU has a reputation for providing a highly reliable OS that provides it users with freedoms not available with other OS's.

The failure is that there is no complete GNU OS yet; without the kernel, the GNU tools are add-ons to someone else's OS. Yes, the GNU tools span the range from boot (grub) to userland (GNOME), but without a kernel there's not complete OS, and that was the original GNU goal (a free UNIX clone).

Once the Hurd kernel is ready we'll be debating different issues, but until then, Linux is the OS that counts.

READ THE FAQ

[FooBarWidget]

They are not telling you to change the name! They are politely ASKING you to CALL it GNU/Linux!

Re:Why are they so desperate ?

[JdV!!]

cd, *giggle*
<pedantic>

[jan@forterie jan]$ ls /bin/cd /usr/bin/cd
ls: /bin/cd: No such file or directory
ls: /usr/bin/cd: No such file or directory
[jan@forterie jan]$

cd is a built-in of your shell. A popular joke is to have someone write cd.c and have him wonder why it doesn't work... /<pedantic>

Re:Why are they so desperate ?

[netphilter]

Linus didn't develop an OS, he developed a kernel. This is exactly what the FSF is saying...Linux is not an OS, it's a kernel, and the GNU tools play as much a part of the OS as the kernel does. Try to do something in GNU\Linux without a kernel...now try it without GNU software. Neither work.

Hey! I don't appreciate that!

[dalutong]

I've met RMS -- and he didn't spit at me. I used "GNU and Linux" or "GNU Linux" when speaking to him as I do when I speak anyone else (though i made an effort not to let the occational "linux" slip through.) Why? Because I respect him tremendously.

He doesn't advocate GNU/Linux so much because he wants the personal fame. Remember -- GNU doesn't sound like RMS as Linux sounds like Linus. He knows he worked most of his adult life to make these tools. He created the GPL. He started it all. If he hadn't done all this -- Linus wouldn't have found the GNU tools and wouldn't have decided to change Linux from a terminal emulator to a kernel. GNU made it possible.

I don't care if people like to say that their systems have little GNU code. I don't care if they say that "linux is what makes it a distinctive OS." The fact of the matter is, if there was no GNU, there would be no Linux. (Or, if we had it, it wouldn't be the leader of the Free Software Movement.) If GNU and the GPL were not around, neither would linux be.

So I have no problem giving the tools the GNU credit. I have no problem giving Free Software (since, had it not been around, Linux may not have been under a GPL-like license) credit. And I have no problem acknowledging the FSF and GNU's role in the wonder computer system I do have.

Thank you RMS. Thank you Linus. Thank you GNU. Thank you FSF. Thank you X. Thank you Mozilla. Thank you Larry Wall. Thank you Open Office. Thank you all those who i forgot. And don't listen to the parent comment -- he's a fool. (But don't spit on him like you didn't spit on me please.)

Re:Why are they so desperate ?

[gorilla]

Actually, the tools which were used to create Linux and make it into a self-hosted development enviroment were often Minux. At the time of creation of Linux, a lot of the people who jumped on the bandwagon were Minux users who had reached the limits of what Minux could do. That`s why the announcement was in comp.os.minux, Linux has support for Minux file systems, and for a while, you needed Minux to compile the kernel. Read the Original announcement [google.com] for the details.

They are NOT enforcing a name change!

[FooBarWidget]

They are asking you to CALL it GNU/Linux, not to change it's name! What they are *really* after is the general awareness of GNU's existence. They are not enforcing anything, and even if they can, they won't (as stated in the FAQ).
You don't have to call it GNU/Linux. If you want to call it just "Linux" and educate the public by explaining the whole story in 10 minutes, go ahead (the FAQ says the same thing).

Also read this:
http://www.gnu.org/gnu/gnu-linux-faq.html#T OCwhysl ash

"Following the rules of English, in the construction "GNU Linux" the word "GNU" modifies "Linux". This can mean either "GNU's version of Linux" or "Linux, which is a GNU package." Neither of those meanings fits the situation at hand.

Linux is not a GNU package; that is, it wasn't developed under the GNU Project's aegis or contributed specifically to the GNU Project. Linus Torvalds wrote Linux independently, as his own project. So the "Linux, which is a GNU package" meaning is not right."

why they made the faq

[_KhlER3L]

The FAQ @ http://www.gnu.org/gnu/gnu-linux-faq.html#afterker nel says:

It's not exactly GNU--it has a different kernel (that is, Linux). Distinguishing GNU/Linux from GNU is useful.

khl

Re:don't deserve equal mention.

[FooBarWidget]

Yeah and without GNU there wouldn't be a userland and then the Linux project wouldn't even have started!

You're making a mistake here. The FSF is NOT claiming ALL credits!

Read the FAQ:
http://www.gnu.org/gnu/gnu-linux-faq.html#ju stgnu

They ARE giving Linux credits!

Re:My First, Last, and Only Word on this subject..

[vidnet]

My preference at the time was to just get rid of the trademark, to get it declared invalid because of prior use in the industry. We had enough paperwork to show that Linux had, in fact, a history of prior use. The trouble was, our lawyer convinced us that it would be a wasted efforted, that we should not even try to get Linux declared a public domain instead of a trademark. The only way for it to really be in the public domain, he explained, was for it to become generic. And Linux at the time wasn't that generic. The trademark office probably wouldn't even concider it to be generic today. We could lose the battle, he said. Or if we invalidated the old trademark, somebody could possibly come along and trademark it anew.

The solution he suggested was to transfer the trademark to somebody else. My vote went to Linux International, but there was a lot of opposition to that. Linux International was young and unproven. People were worried about Linux International being taken over by commercial interests.(...)

So all eyes looked on me.

'Just for Fun', by Linus Torvalds.

Re:What about everything else?

[jdavidb]

Hi, I'm the article submitter. :)

where does one stop with the attributions?

Did you read the FAQ? I was hoping a few folks would and think about the ideas presented, even if they don't agree.

I did a little experiment today; I downloaded all the source code for Linux From Scratch [linuxfromscratch.org], and moved all the GNU code into a directory. The uncompressed GNU code takes up 341648 bytes. The uncompressed Linux code (counting the kernel, the manpages, and modutils) takes up 155872 bytes.

Since you mentioned X, I uncompressed XFree86 4.1.0 and counted it: 289624 bytes. (I was actually surprised; I expected X to be bigger than GNU.)

For the record, this is not all the GNU software, either. Emacs, for example, is not counted (that would've put it way over the top), and LFS chooses many alternatives where GNU packages exist.

Now, when you talk about the tail wagging the dog, if you want to call GNU the tail, the tail is bigger than the dog. :)

Is the kernel the OS, or are the utils the OS?

Did you read the FAQ? This issue is addressed. There's some truth to both views.

Does kernel32 or command.com makes Windows the "Windows OS"?

That's what GNU is saying. Most people would say the Windows OS consists of those pieces, plus the GUI, plus many utilities. And when you say you got RedHat Linux, do you mean you got the version of Linux, the kernel, distributed by RedHat, or do you mean you got an OS comparable to Windows? Which sense are you using the term OS in there?

RMS and a bit of history

[msobkow]

In 1993-4, the VAX systems at university ran BSD (not Unix -- no one could use the name except AT&T.) The compilers, utilities, etc. were not written by FSF for BSD. Emacs would take over the system, GCC was buggy as hell, and most of the command line utilities downloaded from the gnu.org site didn't exist yet in any form other than a "to do" list.

BSD was created so that educational institutions could run a Unix-like system without paying outrageous fees to AT&T. VMS, RTSX, HP RTE, HP MPE, and other proprietary systems ruled the computing landscape. DEC bought the rights to commercialize a copy of BSD, ran some scripts to replace the BSD credits with their own copyright, and raised a stink with campus, forcing us to use their older BSD version (bringing back the kernel panics and other bugs that had been fixed.)

It was in that expensive proprietary world that RMS toured universities, speaking about the GPL and the ideals of a system that wouldn't require paying corporations to use. He was eloquent, forthright, and convinced many of us that this was a good idea, particularly in light of the DEC Ultrix bugfest that had replaced our much stabler BSD core.

Rick's early speeches were about free-as-in-beer software, with the GPL providing a means of safely producing that environment. His current rants about "philosophical" freedom only became an emphasis several years later. He might have been thinking in those terms early on, but he sure didn't focus on those ideals in his speeches.

I do believe that Rick has lost his connection with the real world somewhere in the past decade. Open discussions and speeches followed by question periods have been replaced by political rants that are an embarassment to those of us who remember the days when the goal was non-corporate computing environments we could all use.

When I see RMS claiming that GNU "inspired" BSD, or Gates claiming credit for the popularity of PCs, it makes me ill. Both of these egomaniacs are quite content to rewrite and ignore history, forgetting that there are thousands of us who lived and worked through the history that they try to rewrite.

The network code has more roots in BSD than GNU or Linux, Sun's BSD-based SunOS and DEC's BSD-based Ultrix provided the development environment for most of the community. Thousands of people have contributed to the hundreds of projects involved, and the vast majority of them are not members of the FSF nor direct contributors to the GNU code.

You want credit, Rick? Fine. Emacs is a nice editor, you did a good job of getting LISP going, and you helped get gcc rolling. Most importantly, you evangelized and inspired many people to help free us from the tyranny of corporate licensing. But you did not do all or even the majority of the work, nor did the FSF as a whole.

Re:Hmm, not so sure about the BSD bit

[ChaosDiscord]

However, if the FSF inspired BSD, then there needs to be a time machine involved there somewhere.

They didn't claim to have inspired BSD's creation, they claim, They claim to have inspired the BSD developers to make their code free software. The BSD source didn't go truly free until much later. No time machine required. (I make no claims as to the accuracy of their statement, simply that's is feasible.)

Um... no.

[Anonymous Coward]

The point is that the GNU project inspired the BSD people to replace all of the AT&T code for the 1994 release, not that they inspired BSD in the first place.