4 ms·

How long will they insist that we genuflect to them for the few pieces of legacy software that they bring to a typical Linux system?

Y_Y 3y ago
I think the naming debate is silly, and I just say "Linux", but (with notable exceptions) tons of the core software on an average Linux system is part of the GNU project, e.g. glibc, GCC, coreutils, make, emacs
- felipetrz 3y ago
  The article is basically saying that a non-gnu Linux distribution, like Alpine, is less linux than ChromeOS, just because it uses GNU. And while being pedantic about this, they still called Linux a Unix (which it's not) while ignoring the commercial unixes because they're dead (which, as much as I'd like it to be true, it's not in the case of MacOS). The person who wrote this article was dishonest or ignorant in so many levels.
  - GabrielTFS 3y ago
    > they still called Linux a Unix (which it's not) Uhhhhhhh, on what basis ? The only arguments I can see for this would be either of: - Pedantry, as technically the Linux kernel itself isn't UNIX certified by the Open Group, who owns the trademark - though multiple distributions have been certified to past standards, with no meaningful changes necessary. As an additional note, if you want to only count implementations conforming to the latest standard as being "real UNIX", then AIX is literally the only UNIX in existence. - Purity, as Linux doesn't directly use any of the original code from AT&T-created UNIX OSes. But by that standard, it seems like modern BSDs wouldn't qualify as "UNIX", given that all of the AT&T code was stripped out a long time ago, and MacOS would qualify even less given it uses a non-AT&T kernel and only a few parts of some BSDs in the userland. Also, if including AT&T-produced code suffices to be "UNIX" then I must ask whether a Linux distribution that comes with ksh qualifies (...or even if Windows with UWIN or Cygwin+ksh qualifies). Generally though, the argument just reeks of ancestor worship to me. ...and neither of them are satisfactory to me.
gsliepen 3y ago
You mean the rock solid userland foundation and the state of the art compiler? (Insert a modified "What have the Romans ever done for us?" sketch here.) I'll happily acknowledge their work I enjoy every day, GNUflecting doesn't seem a high price to pay ;)
- veave 3y ago
  The state of the art compiler has been LLVM for 10 years already.
  - pjmlp 3y ago
    Only until Google and Apple stop caring, now they have an honorable 3rd place in ISO C++ support, and are yet to support as many platforms.
AndrewStephens 3y ago
Forever, and furthermore they request you spell it ge-gnu-flect from now on.
jimrob4 3y ago
Gen-gnu-flect
yjftsjthsd-h 3y ago
> for the few pieces of legacy software Linux is just one piece, and also quite old; by your logic we should really just call it a GNU system. In fact, from a user perspective, a GNU/HURD or Debian GNU/kFreeBSD system far more closely resembles what you're calling Linux than, say, Android which literally uses the Linux kernel.

Y_Y 3y ago

I think the naming debate is silly, and I just say "Linux", but (with notable exceptions) tons of the core software on an average Linux system is part of the GNU project, e.g. glibc, GCC, coreutils, make, emacs

felipetrz 3y ago
The article is basically saying that a non-gnu Linux distribution, like Alpine, is less linux than ChromeOS, just because it uses GNU. And while being pedantic about this, they still called Linux a Unix (which it's not) while ignoring the commercial unixes because they're dead (which, as much as I'd like it to be true, it's not in the case of MacOS). The person who wrote this article was dishonest or ignorant in so many levels.
- GabrielTFS 3y ago
  > they still called Linux a Unix (which it's not) Uhhhhhhh, on what basis ? The only arguments I can see for this would be either of: - Pedantry, as technically the Linux kernel itself isn't UNIX certified by the Open Group, who owns the trademark - though multiple distributions have been certified to past standards, with no meaningful changes necessary. As an additional note, if you want to only count implementations conforming to the latest standard as being "real UNIX", then AIX is literally the only UNIX in existence. - Purity, as Linux doesn't directly use any of the original code from AT&T-created UNIX OSes. But by that standard, it seems like modern BSDs wouldn't qualify as "UNIX", given that all of the AT&T code was stripped out a long time ago, and MacOS would qualify even less given it uses a non-AT&T kernel and only a few parts of some BSDs in the userland. Also, if including AT&T-produced code suffices to be "UNIX" then I must ask whether a Linux distribution that comes with ksh qualifies (...or even if Windows with UWIN or Cygwin+ksh qualifies). Generally though, the argument just reeks of ancestor worship to me. ...and neither of them are satisfactory to me.

gsliepen 3y ago

You mean the rock solid userland foundation and the state of the art compiler? (Insert a modified "What have the Romans ever done for us?" sketch here.) I'll happily acknowledge their work I enjoy every day, GNUflecting doesn't seem a high price to pay ;)

veave 3y ago
The state of the art compiler has been LLVM for 10 years already.
- pjmlp 3y ago
  Only until Google and Apple stop caring, now they have an honorable 3rd place in ISO C++ support, and are yet to support as many platforms.

AndrewStephens 3y ago

Forever, and furthermore they request you spell it ge-gnu-flect from now on.

jimrob4 3y ago

Gen-gnu-flect

yjftsjthsd-h 3y ago

> for the few pieces of legacy software Linux is just one piece, and also quite old; by your logic we should really just call it a GNU system. In fact, from a user perspective, a GNU/HURD or Debian GNU/kFreeBSD system far more closely resembles what you're calling Linux than, say, Android which literally uses the Linux kernel.