Why is Fedora not a Free GNU/Linux distributions?

Rick Stevens ricks at nerd.com
Thu Jul 17 22:06:32 UTC 2008

Antonio Olivares wrote:
> --- On Thu, 7/17/08, Les Mikesell <lesmikesell at gmail.com> wrote:
>> From: Les Mikesell <lesmikesell at gmail.com>
>> Subject: Re: Why is Fedora not a Free GNU/Linux distributions?
>> To: "For users of Fedora" <fedora-list at redhat.com>
>> Date: Thursday, July 17, 2008, 12:33 PM
>> Rick Stevens wrote:
>>>> The history is really much more complex than this.
>>  Wikipedia has a 
>>>> nice graphic of how the open/commercial parts
>> developed at 
>>>> http://en.wikipedia.org/wiki/Unix.  But basically
>> since the 
>>>> government-regulated monopoly (AT&T) that did
>> the initial work could 
>>>> not sell it directly, they licensed it for
>> research purposes to 
>>>> universities where the original BSD additions were
>> components that had 
>>>> to be installed on top of the AT&T code.
>>> I've known Wikipedia stuff to be wrong and part of
>> this is.  In the old
>>> days, you couldn't call it "Unix" unless
>> you had a source code license
>>> from Bell Labs (not AT&T).  I know, I was involved
>> in the negotiations
>>> our company had with BL to get System V source.  They
>> wanted, IIRC, $50K
>>> in 1982.  We said "too much, guys."
>>> You could, however, get a license for BSD for a LOT
>> less ($5k, I think),
>>> and that's what a LOT of people did (including
>> Sun, DEC, IBM, Data
>>> General, Silicon Graphics and others too many to
>> name).
>> This doesn't make much sense until the completion of
>> the standalone BSD 
>> that I thought happened a lot later.  Originally you had to
>> have an AT&T 
>> license to run the BSD additions. And at these prices
>> it's pretty easy 
>> to see why everyone was running Windows a few years later -
>> I still 
>> blame AT&T for that.
>>> Many companies DID use SVR4.2 as the base for later
>> versions of their
>>> OSes.  Sun's Solaris (SunOS 5.x) is SVR4.2-based,
>> whereas the original
>>> SunOS (SunOS 4.x) was BSD-based.  They renamed it
>> Solaris to
>>> differentiate it from the BSD-based earlier OS. 
>> DG's later versions of
>>> their DG/UX was SVR4.2-based.  The first PC-esque
>> SVR4.2 I used was on a
>>> (blast from the past) Amiga 2000 (Motorola 68020),
>> followed by "E-NIX"
>>> (from Everex Computers) on actual i386 hardware.
>>> DEC got so pissed off at the Unix title owner that
>> they went to OSF/1
>>> (Mach-based) for the Alpha products (eventually called
>> "Tru64") and
>>> dropped BSD and SVR4.2 completely.
>> Sensible pricing could have changed everything.  Dell had
>> one of the 
>> least expensive versions of SysVr4 that was still around
>> $1k per box and 
>> it was one of the few that would adapt to generic SCSI
>> drives instead of 
>> being limited to the vendor's set compiled into the
>> kernel (like AT&T's 
>> own retail version).  It mysteriously disappeared right
>> when Windows95 
>> came out.  Of course after the court revelations about
>> Microsoft's 
>> anti-competitive practices, it wasn't so mysterious.
>> -- 
>>    Les Mikesell
>>     lesmikesell at gmail.com
>> -- 
> I have been keeping up with this thread and am actually surprised that no one has mentioned SCO, and if they did, I missed it:)  
> SCO is back at it again and they want Linux to pay for some stuff, and according to the arcticle(which follows) only Microsoft and SunOS are safe from their threats :(
> http://linux.sys-con.com/read/614015.htm

They've never given up and the vast majority of their arguments are
specious at best.  Several judges have already determined that SCO's
previous claims had no merit and the latest seem to simply be rehashes
of things they've claimed in the past.  In some ways it's rather
pathetic.  A company that can no longer compete effectively falls down
the pit of filing bogus suits in the hope that someone will give them
money to just "shut up and go away."

Note that the first SCO offering was called "Xenix" (not even they could
use "Unix" as they didn't have a license from Bell Labs...that ought to
tell you something).  A large part of Xenix was written by Locus
Computing in Santa Monica, CA.  Locus was also responsible for a huge
part of IBM's AIX operating system (again, no "Unix" because no source
license from Bell Labs and IBM could sure-as-hell afford one).

> I have read somewhere(don't recall where) that many of the software that was written for BSD/SunOS made its way into linux since the previous were for pay.  That is where Linux came in, then the big companies started selling it and now is getting bigger everyday, unfortunately there are many who do not like it like Microsoft, Apple, SCO, etc.  
> Despite the arguments into calling it GNU/Linux or simply Linux, it has its place for many of us users who are very happy to run it. 

I think the consensus is that Linux is the kernel and Gnu makes up most 
of the rest of the OS.  Gnu can use Hurd as its kernel as well, but it's
nowhere nearly as common as a Linux kernel.  I've seen Gnu run on a Mach
kernel as well.

As long as the kernel offers entry points and return values conforming
to the POSIX standard, I think Gnu will work with it.  I could be wrong
as I've not dug really deeply into the guts of Gnu.  I'm getting too old
to do that!  :-)
- Rick Stevens, Systems Engineer                       rps2 at nerd.com -
- Hosting Consulting, Inc.                                           -
-                                                                    -
-  Any sufficiently advanced technology is indistinguishable from a  -
-                              rigged demo.                          -

More information about the fedora-list mailing list