> These are improved versions of programs originally from BSD. > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > > And this means that GNU inetutils cannot be part of the GNU system? > Amazing.
Of course a compilation consists of parts. And non-compilations can exists of parts. GNU is a single entity, of which GNU inetutils is a component. > > Emacs was to a significant degreed developed by third parties, I > > guess it too isn't part of the GNU system. > > Is is a component of a compilation, but as such is a single entity > (with very few exceptions (c) FSF due to the practice of copyright > assignments) and has been developed mostly as a single entity. > However, there are subsystems (like calc) which have historically > been distributed as separate entities. So parts of Emacs can be > considered aggregated. There is no necessity for drawing a line > here, however, since copyright and license for the components in > distribution rest with FSF and the GPL. > > Copyright doesn't state if something is or isn't part of a operating > system. You were arguing against GNU systems being compilations. GNU _system_, singularis. Variants of the GNU system can of course be compilations, many GNU/Linux systems are just that, compilations of a bunch of thingies. I'm talking about _the_GNU_System_. And that's an issue defined by copyright law. No, it isn't. Copyright law does not define if GNU is a compilation or not. But for the sake of the argument, even if it did, it would not define GNU as a compilation, since many of the tools have either been written specifically for GNU, or modified for GNU. So it is not a `compilation' in either case. > The license and copyright have nothing to do with this. The GNU C > Library is part of the GNU system, doesn't matter if you do not like > it or not. Of course, a compilation has parts, and the C library was not a topic of discussion, anyway. I'm not talking about the C library, I'm talking about the GNU C library. The C library that is part of the GNU system. But the "operating system" is not a single entity. In our case it is. All the tools developed by the GNU project are develoepd to be run on GNU. You can take, for example, the ftp client and compile and use it under a different operating system. I can take any free software and run it under a different operating system, it is called porting. It does not and never will make a thing `not a single entity'. What next, printf is not part of the C standard because one can take it out? That's what the autoconf stuff is all about: portability. And something which is portable and maintainable as a separate entity is a component of a compilation. That is nice, doesn't make GNU a compilation in either case. > > GCC isn't fully copyrighted by the FSF, neither are many > > projects, yet they are GNU projects, then there are non-GNU > > projects which are part of the GNU system. > > Which, for that reason, is mostly to be considered a compilation > with regard to the copyright situation. > > Nope, it isn't. Please read up on what a compilation is, and please > stop confusing compilations with the GNU system. Next thing we know > is that OpenBSD isn't _really_ a single entity; a operating system, > but a compilation of totally disjoint tools that Theo thought was nice > to publish. With regard to copyright law, of course it is a compilation. No, it isn't. It has been written as a single piece, it has been developed as a single piece, lots of things have been _modified_ for GNU to work on GNU. It isn't a matter of taking a couple programs and saying `this is GNU'. _______________________________________________ gnu-misc-discuss mailing list [email protected] http://lists.gnu.org/mailman/listinfo/gnu-misc-discuss
