On Jul 19, 2008, Steve Underwood <steveu@coppice.org> wrote:

> Alan Cox wrote:
>> On Sat, 19 Jul 2008 08:55:18 -0700 (PDT)
>> Antonio Olivares <olivares14031@yahoo.com> wrote:

>>> They(FreeBSD) should be protected, the users can get the ports from
>>> source, they do not ship binaries(except the installation *.tbz
>>> files).

>> That makes no difference to US patent law.

> Its the same with anyone else's patent law, too. Shipping things as a
> kit of parts never successfully circumvented any patent laws.

But what about shipping only software, rather than computer systems?
Wasn't there a recent court ruling favorable to a software distributor
defendant? AT&T vs Microsoft, Supreme Court, IIRC.

--
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

On Jul 19, 2008, Antonio Olivares <olivares14031@yahoo.com> wrote:

> Your meaning of object code is binary code?

The GPL defines it anything that's not source code, i.e., that's not
the preferred form for making modifications ot the software.

> and the modifications is the source code?

Modifications are made to the source code, yes.

> If I understand your point, they released versions of Linux only as
> binary, but without providing the sources(source code) for their
> modifications.

Without providing corresponding source code, or offering to provide
it, yes. Whether they modify it or not is irrelevant.

> You can do whatever you want with the code, just not release it to
> anyone and if you do, you have to provide the source code?

> Is that correct?

This pretty much sums up GPLv2. GPLv3 is has further requirements in
some details, and is more lax in others.

--
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

On Jul 19, 2008, Les Mikesell <lesmikesell@gmail.com> wrote:

> If someone wants to go looking for binary distributions, there are
> probably some of the vmware appliances at
> http://www.vmware.com/appliances/directory/ that haven't properly
> followed the strict GPL requirements to provide all the corresponding
> sources.

Would you please notify the copyright holders? They're the only ones
with legal standing to do anything about it and have their wish,
encoded in the GPL, respected.

--
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

On Sun, Jul 20, 2008 at 01:06:58 -0430,
Patrick O'Callaghan <pocallaghan@gmail.com> wrote:
> On Sat, 2008-07-19 at 13:22 -0500, Bruno Wolff III wrote:
> > On Thu, Jul 17, 2008 at 20:16:33 -0430,
> > Patrick O'Callaghan <pocallaghan@gmail.com> wrote:
> > >
> > > I saw it running on an LSI-11 (a minimal PDP-11 with no memory
> > > manegment). It wasn't much use for anything except data capture as it
> > > couldn't have more than 64k of memory.
> >
> > I used to use a PDP-8e timeshare system with less memory than that. I think
> > the one I used had 4 memory boards each with 4k 12bit words of memory
> > and that it would support up to about 16 simultaneously logged in users.
>
> Not on Unix I assume. Perhaps some kind of RSTS system?

TSS8. The base OS came from DEC, but one of the university's programmers
significantly hacked on the version we ran.

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

On Sat, Jul 19, 2008 at 16:12:54 -0500,
Les Mikesell <lesmikesell@gmail.com> wrote:
> Bruno Wolff III wrote:
>>
>>> When the AT&T monopoly was sensibly split up, it was then able to
>>
>> I think just plain split up is a better description. I think history has
>> shown that the split wasn't done "sensibly" and unfortunately we aren't
>> going to get a do over any time soon.
>
> You haven't noticed the dismembered Bell's crawling back together to
> resurrect the monster? Plus of, course devouring Cingular. (I'm not a
> big fan of huge corporations...).

And part of the reason for that is the split was done incorrectly. It should
have been split so the local loops and the CO's were owned by regulated
monopolies (in each region) and those companies prohibited from providing
services over the copper (or fiber). Then the incumbents wouldn't be able
to eliminate competition by sabotaging it.

There isn't going to be the political will to do a split like that any time
soon.

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

Bruno Wolff III wrote:

>>> I think just plain split up is a better description. I think history has
>>> shown that the split wasn't done "sensibly" and unfortunately we aren't
>>> going to get a do over any time soon.
>>
>> You haven't noticed the dismembered Bell's crawling back together to
>> resurrect the monster? Plus of, course devouring Cingular. (I'm not a
>> big fan of huge corporations...).
>
> And part of the reason for that is the split was done incorrectly. It
> should have been split so the local loops and the CO's were owned by
> regulated monopolies (in each region) and those companies prohibited from
> providing services over the copper (or fiber). Then the incumbents
> wouldn't be able to eliminate competition by sabotaging it.

Was the split-up good?
As a complete outsider, it doesn't seem to me to have had a great effect,
good or bad.

What I found, and find, rather puzzling was the anti-trust ruling
that did not allow AT&T to compete in software, or IBM in telephony.
On the face of it, this seemed to preserve monopolies
rather than the opposite.

To me, the saddest thing about the whole story was the demise of Bell Labs,
at least as it used to be.
A bit like the end of the Venetian republic.



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

> > They have said that they were going to block non GPL
> modules
>
> Who did?
http://www.linuxactionshow.com/forum/comments.php?DiscussionID=247

http://kerneltrap.org/node/1735
>
> This is not something that can be done in a Free Software
> copyright
> license any more than the GPL already does. The only way a
> non-GPL
> module can exist today is by claiming to not be a derived
> work, and if
> it's not a derived work, no copyright license (!=
> contract) can
> possibly lay any claim on it.
>
> --

Regards,

Antonio




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

Timothy Murphy wrote:



I think just plain split up is a better description. I think history has
shown that the split wasn't done "sensibly" and unfortunately we aren't
going to get a do over any time soon.

You haven't noticed the dismembered Bell's crawling back together to
resurrect the monster? Plus of, course devouring Cingular. (I'm not a
big fan of huge corporations...).

And part of the reason for that is the split was done incorrectly. It
should have been split so the local loops and the CO's were owned by
regulated monopolies (in each region) and those companies prohibited from
providing services over the copper (or fiber). Then the incumbents
wouldn't be able to eliminate competition by sabotaging it.


Was the split-up good?
As a complete outsider, it doesn't seem to me to have had a great effect,
good or bad.


It was horrible, but so was the monopoly. People can afford to make long
distance calls now and you can connect equipment made by someone other
than Western Electric to the phone jacks. And you can credit all of the
fiber laid by competing companies that gives us the internet bandwidth
we enjoy now to it. Many of those companies went bankrupt since they
weren't able to get AT&T-like prices in a competitive environment -
meaning the monopoly had so wildly distorted the pricing that it
confused normal business planning.


The problem the breakup caused was that they used their huge coffers of
cash to buy up other companies (e.g. NCR) and mis-managed them into the
ground since they had absolutely no experience in competitive business
or pricing. And the problem the reconstruction is causing is about the
same. As the sales and billing operations of Cingular and the Bells is
shifting back to central control it is beginning to take months to get a
simple T1 installed and next-to-forever to get a credit that was due to
a screwup in what used to be one of the other companies.



What I found, and find, rather puzzling was the anti-trust ruling
that did not allow AT&T to compete in software, or IBM in telephony.


The idea was that no single company should control communication from
end to end as computers/software became significant parts. A company
should not be able to use its established monopoly position at one end
of the line to force the equipment at the other end to match. And the
communication line providers should not be motivated to provide better
service for their own end points than competitors. Had AT&T been
permitted to control the communication lines we'd probably have a split
among OSI networks with telco-designed layers and AT&T-controlled
directory services instead of near-universal IP networking with a
universal DNS as we have now.



On the face of it, this seemed to preserve monopolies
rather than the opposite.


Monopolies in one field aren't considered as bad as using that existing
monopoly to expand control into another field. The current
'net-neutrality' issues are along the same lines. Broadband internet
providers want to be directly in the media distribution business and
they have the capability and motivation to sabotage connections from
their customers to competing providers (as evidenced by Comcast actively
breaking bittorrent until recently).



To me, the saddest thing about the whole story was the demise of Bell Labs,
at least as it used to be.
A bit like the end of the Venetian republic.


Agreed - if you overcharge enough you can fund some nice unrelated
projects, but things like unix really didn't have a reasonable way out
of the lab into the real world. It sort-of survived in educational
environments on merit, but imagine what could have happened if there had
been a company behind it that would have competed on price against
everyone else. When AT&T did start selling unix it was well over $1,000
a box with the compiler and X as extra-cost options. Businesses had no
choice but to install Windows and the rest is history.


--
Les Mikesell
lesmikesell@gmail.com


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

Les Mikesell wrote:
I don't see how anyone can forget that he plainly wrote that demanding
that modules be GPL'd is both legally and morally wrong. Or have any
question about the meaning of this portion of the Linux license:


Again, he said no such thing. In fact, he argued explicitly that even if
it legally correct, it is morally wrong.



"NOTE! This copyright does *not* cover user programs that use
kernel services by normal system calls - this is merely considered
normal use of the kernel, and does *not* fall under the heading
of "derived work".


This is clearly talking about syscall interface and unrelated to the
kernel modules interface.



"Well, there really is no exception.


So his own quote, and the FSF legal counsel's understanding of the terms
as he stated them were both wrong?


The portion I quote is from a mail from Linus. He clearly says there is
no exception


Read again,

http://www.win.tue.nl/~aeb/linux/lk/COPYING-modules.txt
http://web.archive.org/web/20060202062935/people.redhat.com/arjanv/COPYING.modules

FSF didn't give a direct interpretation of his quote and that cannot be
held more authoritative than the original copyright holders anyway.
Simply put, FSF doesn't hold copyright over the Linux kernel. The
copyright holders (which is more than just Linus) intention's matter.


You were told about the problems earlier on too and you choose to
ignore it. CDDL was deliberately designed to be incompatible with GPL


Deliberate? _Everything_ that is not the GPL is incompatible with the GPL.


This one is a clear lie. There are dozens and dozens of GPL compatible
licenses. Any license which has no additional restrictions above and
beyond GPL requirements are compatible with it.


http://fedoraproject.org/wiki/Licensing

Rahul


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

* Rahul Sundaram <sundaram@fedoraproject.org> [20080720 19:11]:
> Les Mikesell wrote:
[snip]
>>
>> Deliberate? _Everything_ that is not the GPL is incompatible with the GPL.
>
> This one is a clear lie. There are dozens and dozens of GPL compatible
> licenses. Any license which has no additional restrictions above and
> beyond GPL requirements are compatible with it.
>
> http://fedoraproject.org/wiki/Licensing

Just so that I understand this right.

Any license which will permit being replaced by the GPL, when the
software it covers is combined with software under the GPL - is
compatible with the GPL?

And any license that does not permit itself to be replaced or
over-ruled by the GPL - is hence incompatible - even if it explicitly
permits combination with the GPL for any derived work or combination
work.

Am I understanding this right?

/Anders

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