Les Mikesell wrote:

> 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.

Actually, Unix edition 5 was free from Bell Labs, at least to universities.
There was even a microUnix which was supposed to run
on what were then called micro-computers.
But I never heard of anyone actually getting it (microUnix) to run.

I recall getting Unix edition 5 on a tape from Bell labs,
with a note attached that the tape was not guaranteed to contain anything.

But I didn't think AT&T were allowed to charge anyone for Unix.
I thought there was a legal agreement that AT&T must not sell software,
and IBM must not enter telecommunications.



--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list

On Thu, 2008-07-17 at 09:45 -0700, Gordon Messmer wrote:
> John Cornelius wrote:
> > This discussion is becoming both increasingly religious and somewhat
> > oblique in its depictions of the elements under discussion. It may be
> > instructive to review the classic definitions of some of these elements
> > in order to clarify in the minds of zealots from the several sides of
> > the discussion and thereby promote a more rational discussion.
>
> Can you cite any consensus based definition of "operating system" other
> than what you've provided? I think that the POSIX specification is
> generally agreed to be the definition of one operating system interface,
> and it includes the shells, editors, compilers, etc that you've decided
> aren't part of an operating system.
>
> That's what we're getting at. GNU/Linux is an operating system. Linux
> is one of the kernels that GNU *can* use, and one of the most common
> that it does.
>
> > GNU is not an operating system it is, and as far as I know always has
> > been, a tool kit that is platform and operating system independent.
>
> I think that the GNU developers disagree with you. What makes your
> opinion more valid than theirs?
>
> > GNU is not Linux and Linux is not GNU, it's just an evolution of a
> > movement started by Ken Thompson and Dennis Ritchie nearly 40 years ago.
> >
> > Whoda thunk?
>
> I think you're giving Ken and Dennis too much credit. As far as I
> understand it, Unix was only distributed free of charge because ATT was
> concerned that its monopoly status prevented it from entering new
> markets. Look at Plan 9. Free Software? Nope.
>
> GNU modeled its operating system after Unix because it was a common
> system, not because there was any particular sharing of ideals or goals.
>
The US government paid for the development of UNIX. That was the
original source of its being "free". The people of the US owned it.

I got that from some folks very close to the source.

--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list

On Jul 17, 2008, Alexandre Oliva <aoliva@redhat.com> wrote:

> On Jul 17, 2008, John Cornelius <jc@hangarpilot.net> wrote:
>> Operating System:

> [kernel and kernel inspection and filesystem utilities, no more]

> Per your definition, UNIX wouldn't have ever been regarded as just an
> operating system.

And then, UNIX philosophy would probably be limited to "everything is
a file" because, by this definition, everything else we understand as
UNIX wouldn't apply to the operating system, but rather to this
broader concept for which we don't seem to have a name.

Nothing like "many small programs, each doing a single simple task
very well, that can be combined through pipes and a powerful shell
programming language" would be part of the UNIX philosophy, because,
well, these small programs wouldn't be part of UNIX per this narrow
definition.

Nothing like "the same low-level programming language usable all the
way from the guts of the kernel to applications, and a well-defined
system API available to that programming language" would be part of
the UNIX philosophy, and it wouldn't have a C library (and a C
compiler) as fundamental building blocks.

Heck, you wouldn't even have an editor (teco, ed, vi, or the file
editing abilities of emacs) to adjust config or rc files if they
weren't part of the operating system. You wouldn't have a user
interface (textual or graphical) to even get to them.

Of course, you can remove a few of these components and end up with
something that could still pass for an operating system, but defining
operating system so as to exclude all of this doesn't match historical
practical (as opposed to academic, I guess) use of the term.

> Which is wrong, the definition or the understanding as to what UNIX
> is?

--
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}
FSFLA Board Member ¡Sé Libre! => http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}

--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list

Alexandre Oliva wrote:



[kernel and kernel inspection and filesystem utilities, no more]



Per your definition, UNIX wouldn't have ever been regarded as just an
operating system.


And then, UNIX philosophy would probably be limited to "everything is
a file" because, by this definition, everything else we understand as
UNIX wouldn't apply to the operating system, but rather to this
broader concept for which we don't seem to have a name.


I look at at this way: a unix-like operating system is the part that
makes everything look like a file and applications that are portable
across them only need an API of creat(), open(), read(), write(), and
ioctl() with all needed permission/device/socket/fifo magic handled by
the OS during open(). That's not completely true and of course there
are library layers above that, but it is the concept that distinguishes
unix. It's closer to 'everything is a file descriptor' than everything
is a file since they are often inherited without the app even knowing
about the associated file/device/socket.



Nothing like "many small programs, each doing a single simple task
very well, that can be combined through pipes and a powerful shell
programming language" would be part of the UNIX philosophy, because,
well, these small programs wouldn't be part of UNIX per this narrow
definition.


That's a good idea under any OS, not particularly unique to unix.



Nothing like "the same low-level programming language usable all the
way from the guts of the kernel to applications, and a well-defined
system API available to that programming language" would be part of
the UNIX philosophy, and it wouldn't have a C library (and a C
compiler) as fundamental building blocks.


The c library isn't unique to unix by design.

--
Les Mikesell
lesmikesell@gmail.com

--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list

On Jul 18, 2008, Les Mikesell <lesmikesell@gmail.com> wrote:

> I look at at this way: a unix-like operating system is the part that
> makes everything look like a file and applications that are portable
> across them only need an API of creat(), open(), read(), write(), and
> ioctl()

API = UNIX-like C library? On GNU+Linux, GNU libc, right? Linux per
se offers no such APIs.

>> Nothing like "many small programs, each doing a single simple task
>> very well, that can be combined through pipes and a powerful shell
>> programming language" would be part of the UNIX philosophy, because,
>> well, these small programs wouldn't be part of UNIX per this narrow
>> definition.

> That's a good idea under any OS, not particularly unique to unix.

Heh. Discarding part of a statement to make it fit others isn't very
nice :-) Good ideas are present in any OS, so... whatever conclusion
you might want to get to :-)

> The c library isn't unique to unix by design.

I don't understand what you're trying to communicate here, there are
several different possible interpretations. Please expand or use
different words?

--
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}
FSFLA Board Member ¡Sé Libre! => http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}

--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list

Alexandre Oliva wrote:



I look at at this way: a unix-like operating system is the part that
makes everything look like a file and applications that are portable
across them only need an API of creat(), open(), read(), write(), and
ioctl()


API = UNIX-like C library? On GNU+Linux, GNU libc, right? Linux per
se offers no such APIs.


In the original design, the things in section 2 of the manual (fork(),
open(), pipe(), etc.) were system calls and thus by definition what the
operating system does. Now everything is stubbed into the c library so
it is harder to tell the difference unless you look at the man section
number that documents the function.



Nothing like "many small programs, each doing a single simple task
very well, that can be combined through pipes and a powerful shell
programming language" would be part of the UNIX philosophy, because,
well, these small programs wouldn't be part of UNIX per this narrow
definition.



That's a good idea under any OS, not particularly unique to unix.


Heh. Discarding part of a statement to make it fit others isn't very
nice :-) Good ideas are present in any OS, so... whatever conclusion
you might want to get to :-)


The philosophy is loosely associated with unix and aided by it's ability
to fork fairly lightweight processes, but really isn't specific to
anything, much less whether something is GNU-ish or not. Emacs, the
very first GNU program, was probably the thing most widely accused of
straying from the 'small program that does one thing' concept in its
day. (Actually, I think emacs goes back farther than unix, but my
memory is a little fuzzy that far back).



The c library isn't unique to unix by design.


I don't understand what you're trying to communicate here, there are
several different possible interpretations. Please expand or use
different words?


We've strayed from GNU vs. not-GNU into unix vs. not-unix, but the point
of unix-like operating systems is to hide the differences among
processor types and i/o devices and present the system calls as
described in section 2 of the manual regardless of the hardware - and to
do some special stuff in process id 1 which is the only one not created
by fork() with inherited file descriptors. The point of the standard c
library is to hide the differences among operating systems.


But back to GNU vs. not-GNU. There is a GNU operating system. Nobody
uses it and there are reasons for that. Instead of fixing those
reasons, they want to tack their name onto the distributions that people
do use because some portable applications from their set are often
included. For reasons I've already covered, I'd rather see distributions
minimize the GPL-encumbered code and use things from the *bsd or
opensolaris projects instead - except for gcc which I don't think has a
less restricted counterpart. Then the GNU project could just go its own
way in the isolation the GPL demands while the rest of the world
cooperates and interoperates at the component level.


--
Les Mikesell
lesmikesell@gmail.com

--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list

> But back to GNU vs. not-GNU. There is a GNU operating
> system. Nobody
> uses it and there are reasons for that. Instead of fixing
> those
> reasons, they want to tack their name onto the
> distributions that people
> do use because some portable applications from their set
> are often
> included. For reasons I've already covered, I'd
> rather see distributions
> minimize the GPL-encumbered code and use things from the
> *bsd or
> opensolaris projects instead - except for gcc which I
> don't think has a
> less restricted counterpart. Then the GNU project could
> just go its own
> way in the isolation the GPL demands while the rest of the
> world
> cooperates and interoperates at the component level.
>
> --

While ideally that would be nice, it won't happen

Which license should be used then, if the above happens?

There are other free licenses/open source licenses around, but which is the best one to use?

As I understand, if one uses Linux

Linux <===> GPL
*BSD <==> BSD/GPL(for KDE/Gnome other DE)
Solaris <===> CDDL

A license that is often mentioned indirectly is the Beerware license

http://en.wikipedia.org/wiki/Beerware

Would that one be accepted as an appropriate license?

People have to pay bills, support their families, gas up their vehicles, from where would the motivation come to continue giving up some work for no $, Which is the best way to continue and support your favorite OS and applications at the same time?

There are many revisions to the GPL as of now, and I believe that one was posted on this list, how can cooperation be achieved and also protect the code as well?

Regards,

Antonio




--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list

Antonio Olivares wrote:


For reasons I've already covered, I'd
rather see distributions
minimize the GPL-encumbered code and use things from the
*bsd or
opensolaris projects instead - except for gcc which I
don't think has a
less restricted counterpart. Then the GNU project could
just go its own
way in the isolation the GPL demands while the rest of the
world
cooperates and interoperates at the component level.


--


While ideally that would be nice, it won't happen

Which license should be used then, if the above happens?


Licenses are difficult to change on existing works, so like now, you'd
have to keep the license as is on each component that you choose. It
only becomes an issue when you start combining components, and then only
with the GPL'd ones since the others don't care if you link with someone
else's library or vice versa.



There are other free licenses/open source licenses around, but which is the best one to use?

As I understand, if one uses Linux

Linux <===> GPL


Yes, that means you can't swap driver/filesystem code with old-style BSD
or cddl licensed versions, but it doesn't specifically mean you have to
use GNU/GPL'd userland code.



*BSD <==> BSD/GPL(for KDE/Gnome other DE)
Solaris <===> CDDL


Pieces can keep these licenses and still be redistributed together.
GPL'd parts can be aggregated in the distro and under certain
circumstances, certain GPL'd things can be linked with others (there's
the weird standard OS component exception but then you have to
distribute the parts separately). The *bsd's tend to avoid including
GPL'd work wherever there is a choice since their philosophy is to avoid
the restrictions but it's not a strict requirement at the standalone
program level.



A license that is often mentioned indirectly is the Beerware license
http://en.wikipedia.org/wiki/Beerware

>

Would that one be accepted as an appropriate license?


As long as it doesn't place requirements on other components it works
for me.



People have to pay bills, support their families, gas up their vehicles, from where would the motivation come to continue giving up some work for no $, Which is the best way to continue and support your favorite OS and applications at the same time?


That's why commercial licenses exist, and why it's good for code that
you wish to be distributed freely to play nicely with them. There are
plenty of people who appreciate having affordable commercial software
choices that may be better than the free alternatives.


There are many revisions to the GPL as of now, and I believe that one was posted on this list, how can cooperation be achieved and also protect the code as well?


Code that is freely available doesn't need protection as nothing can
happen to it other then someone else using and improving it which is a
good thing regardless of what else happens to that copy subsequently.


Long ago it might not have been completely predictable that many end
points of the longest-developed paths of unix development
(http://en.wikipedia.org/wiki/Image:Unix_history-simple.svg) would be
open-sourced but it was never out of the question either. Having that
big chunk isolated by the GPL and unable to share components is just bad
for everyone.


--
Les Mikesell
lesmikesell@gmail.com

--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list

> Code that is freely available doesn't need protection
> as nothing can
> happen to it other then someone else using and improving it
> which is a
> good thing regardless of what else happens to that copy
> subsequently.
I am sure many would disagree with this, The code has to be protected in some way to ensure that someone/or a company cannot claim the code to be theirs and start selling it and not give anything back. This is the good side of the GPL if there is one.
>
> Long ago it might not have been completely predictable that
> many end
> points of the longest-developed paths of unix development
> (http://en.wikipedia.org/wiki/Image:Unix_history-simple.svg)
> would be
> open-sourced but it was never out of the question either.
> Having that
> big chunk isolated by the GPL and unable to share
> components is just bad
> for everyone.
>
Unix is not GPL'd, Linux is or did I miss something here?

The components can be shared, you just have to use the GPL and license your work on it. This is like I scratch your back, but you will also scratch mine. Cooperation is the key and interoperability between compnents like you have mentioned.
> --

Regards,

Antonio




--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list

> > Long ago it might not have been completely predictable
> that
> > many end
> > points of the longest-developed paths of unix
> development
> >
> (http://en.wikipedia.org/wiki/Image:Unix_history-simple.svg)

Thank you for the link, a friend of mine has one in his classroom and it has 1969 as the birth of Unix, and in 1991/1992 the birth of Linux. Linux the kernel and combined with the GNU utilities started the Linux distribution. Another friend of mine also told me that many programs that are now in Linux, were either copied over from BSD/Solaris or ported to linux and were allowed to be shared across without restrictions However, we now now that there are GPL police out there and enforcing the GPL on people who modify the freely available code out there, but do not share their modifications

Is Novell the owner of Unix?
Is it SCO
http://en.wikipedia.org/wiki/SCO_Group ?

Does the GPL protect Linux users across the globe against malicious lawsuits claiming ownership of programs that are used?

These are other questions that affect the status quo and what is done to correct issues.

I also wonder if the GPL is a true open source model? Some posts here are pointing that it is not , I am still confused because of too much jargon present.

http://www.opensource.org/

The GPL is there along with its multiple versions

http://www.opensource.org/licenses/alphabetical

According to the definition of Open Source,

http://www.opensource.org/docs/osd

The GPL then violates #9 in the definition

<quote>
9. License Must Not Restrict Other Software

The license must not place restrictions on other software that is distributed along with the licensed software. For example, the license must not insist that all other programs distributed on the same medium must be open-source software.
</quote>

or does it really violate it?

Regards,

Antonio




--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list