| gabriel rosenkoetter on Fri, 14 Jun 2002 23:10:15 +0200 |
|
On Fri, Jun 14, 2002 at 04:39:59PM -0400, Noah silva wrote: > I think RMS's argument is entirely invalid. > > "tar' isn't all of userland. and GNU tar wasn't very usefull intil it had > a real kernal to run on. What's more, I could easily write TAR, CP, LS, > etc. I wouldn't try to write anything liek the linux kernal. We don't > have to use GNU ls, we could use BSD ls, or any other. What's more > userland's upper levals are what is making linux more popular, and those > are things like KDE and apache, mozilla, etc. It doesn't matter whether you *could*, you haven't. You're using software released under the auspices of the GNU Project. And some of that software is significantly non-trivial, like find(1), grep(1) (another one where NetBSD uses the GNU version, the AT&T source being POSIX.2 non-compliant and really crufty to boot), and emacs (say what you will, it's complicated), to pick just a few. Just as soon as you *are* using an at least mostly non-GNU userland, *then* RMS's argument would be entirely invalid. Till then, it holds a bit of water. Not that it doesn't come off as whiny, mind you. -- gabriel rosenkoetter gr@eclipsed.net PS, I'm far from being a Stallman apologist. I think he's a jerk with some really whacked out ideas, both about software licensing and a variety of other things, but that's just my opinion and has nothing to do with the facts of the whose-userland-Linux-distros-use situation. Attachment:
pgpFRrGNkyD13.pgp
|
|