SCO may be contemplating suing SGI next, apparently.
[url]http://zdnet.com.com/2100-1104_2-5072061.html[/url]
--
100% of all human deaths occur within 100 miles of Earth.
Printable View
SCO may be contemplating suing SGI next, apparently.
[url]http://zdnet.com.com/2100-1104_2-5072061.html[/url]
--
100% of all human deaths occur within 100 miles of Earth.
In article <4krilvgtb3jcvrsodclkk7r8knpf7akvm4@4ax.com>,
Gerald Henriksen <ghenriks@rogers.com> wrote:
:Well I don't think SGI really qualifies as a "major" hardware maker
:which is CNET's first contention in deciding on SGI.
No, there's a reason before that that no-one happened to mention
on here, and which I never got around to posting about.
A couple of weeks ago, SCO has a presentation showing "evidence"
that code had been copied. They showed two samples in the
public presentation, obscuring them by using a greek font.
[They showed more samples in more private presentations that
required NDAs.] Someone took pictures, and people reversed the font
back to ascii, and then people analyzed where the source came from.
The first source that was shown turned out to be from [as I recall] a
section of a print spooler that had been written under a BSD- type
license and which SCO (or one of its ancestors) had then incorporated
into Unix System V. No offence for Linux to have used it.
The second source was from a memory allocator that turned out to
have been redundant and had already been removed from the Linux 2.5
kernel. According to the comments I saw, this code had been added
originally in order to support an SGI system that was never
actually sold, and a rare SGI system that not a lot of were sold.
So... SCO likely had SGI in mind: it's public presentation,
showing what would likely have been two of the best examples of
code copying, included one which was SGI-specific.
--
Positrons can be described as electrons traveling backwards in time.
Certainly many Usenet arguments about the past become clearer when they
are re-interpreted as uncertainty about the future.
-- Walter Roberson
Dans article <4krilvgtb3jcvrsodclkk7r8knpf7akvm4@4ax.com>,
[email]ghenriks@rogers.com[/email] disait...[color=blue]
>
> A lawsuit, no matter how frivolous, would likely put further downward
> pressure on SGI's stock price, something SGI cannot afford. So SGI
> may decide the removal of XFS/NUMA/etc.
>[/color]
They can't remove something they gave away. XFS and NUMA are available
everywhere, part of the stock Linux kernel, and certainly it's not in the
power of SGI to remove them...
--
Quis, quid, ubi, quibus auxiliis, cur, quomodo, quando?
Emmanuel Florac wrote:[color=blue]
> Dans article <4krilvgtb3jcvrsodclkk7r8knpf7akvm4@4ax.com>,
> [email]ghenriks@rogers.com[/email] disait...
>[color=green]
>>A lawsuit, no matter how frivolous, would likely put further downward
>>pressure on SGI's stock price, something SGI cannot afford. So SGI
>>may decide the removal of XFS/NUMA/etc.
>>[/color]
>
>
> They can't remove something they gave away. XFS and NUMA are available
> everywhere, part of the stock Linux kernel, and certainly it's not in the
> power of SGI to remove them...[/color]
Also, I expect there would be some license leftovers from Cray somehwere
at SGI.
Erik
On Sat, 6 Sep 2003 12:11:08 +0200, Emmanuel Florac
<eflorac@imaginet.fr> wrote:
[color=blue]
>Dans article <4krilvgtb3jcvrsodclkk7r8knpf7akvm4@4ax.com>,
>ghenriks@rogers.com disait...[color=green]
>>
>> A lawsuit, no matter how frivolous, would likely put further downward
>> pressure on SGI's stock price, something SGI cannot afford. So SGI
>> may decide the removal of XFS/NUMA/etc.[/color]
>
>They can't remove something they gave away. XFS and NUMA are available
>everywhere, part of the stock Linux kernel, and certainly it's not in the
>power of SGI to remove them...[/color]
Actually they can.
If SGI did not have the rights to the software in question then SGI
cannot re-license them under the GPL. Thus the code in question can
be revoked as it would no longer be GPL, and hence cannot be in the
Linux kernel.
This is no different than the requests by the Linux community to SCO
to reveal what the offending code is so that it can either be removed
or proven to not violate SCO's IP.
Now my personally opinion is that SGI did everything correctly and any
legal action by SCO would be harassment.
But as is so often the case in the legal world the costs of defending
yourself often exceed the cost of agreeing to a settlement.
Dans article <5crjlvcur2stfoqlp6ft7q6s8kmtfi7gdo@4ax.com>,
[email]ghenriks@rogers.com[/email] disait...[color=blue]
>
> Now my personally opinion is that SGI did everything correctly and any
> legal action by SCO would be harassment.
>[/color]
Actually it's extremely unlikely that companies such as SGI or IBM did
provide anything under the GPL without first getting sure that's all
right using a hord of lawyers...
--
Quis, quid, ubi, quibus auxiliis, cur, quomodo, quando?
Emmanuel Florac <eflorac@imaginet.fr> wrote in message news:<MPG.19c425bed0d7e1bb98c9b7@news.free.fr>...[color=blue]
> Dans article <5crjlvcur2stfoqlp6ft7q6s8kmtfi7gdo@4ax.com>,
> [email]ghenriks@rogers.com[/email] disait...[color=green]
> >
> > Now my personally opinion is that SGI did everything correctly and any
> > legal action by SCO would be harassment.
> >[/color]
>
> Actually it's extremely unlikely that companies such as SGI or IBM did
> provide anything under the GPL without first getting sure that's all
> right using a hord of lawyers...[/color]
Bah... I'm soooo tired of all the SCO crap. I really don't understant
where they're trying to get to with it all. Why can't the rest of the
world issue a "put up or shut up" type order like our German friends
did?
Personally, I'm all in favour of Linux succeeding. I like the ideals,
and I'm all for sharing knowledge. That said I'd never run it on a
production box. It's still cool that I can have a (semi) robust
UNIX-like OS running on my commodity laptop though.
The problem is that the UNIX genepool has become so distributed of the
years that it's totally impossible to attribute all of the original
SysV code to any particular organisation. Also, some of the code SCO
is disputing appears to be just textbook implementations of things
like memory allocation or scheduling. A patent has to be "non
obvious", but their supposed precious code is just no-brainer first
year computer science stuff.
James
In article <5crjlvcur2stfoqlp6ft7q6s8kmtfi7gdo@4ax.com>,
Gerald Henriksen <ghenriks@rogers.com> wrote:[color=blue]
>On Sat, 6 Sep 2003 12:11:08 +0200, Emmanuel Florac
><eflorac@imaginet.fr> wrote:
>[color=green]
>>Dans article <4krilvgtb3jcvrsodclkk7r8knpf7akvm4@4ax.com>,
>>ghenriks@rogers.com disait...[color=darkred]
>>>
>>> A lawsuit, no matter how frivolous, would likely put further downward
>>> pressure on SGI's stock price, something SGI cannot afford. So SGI
>>> may decide the removal of XFS/NUMA/etc.[/color]
>>
>>They can't remove something they gave away. XFS and NUMA are available
>>everywhere, part of the stock Linux kernel, and certainly it's not in the
>>power of SGI to remove them...[/color]
>
>Actually they can.
>
>If SGI did not have the rights to the software in question then SGI
>cannot re-license them under the GPL. Thus the code in question can
>be revoked as it would no longer be GPL, and hence cannot be in the
>Linux kernel.
>
>This is no different than the requests by the Linux community to SCO
>to reveal what the offending code is so that it can either be removed
>or proven to not violate SCO's IP.
>
>Now my personally opinion is that SGI did everything correctly and any
>legal action by SCO would be harassment.[/color]
You can't take code from BSD-licensed software, strip the copyright header,
and release it under the GPL. It's a comparatively minor sin, but the
memory allocator in that device driver from SCO's slides would be an
example of, at the very least, that. Very sloppy work on SGI's part; and
particularly bad since the party licensing said software under the BSD-style
license in this case was SCO.
There is, of course, some question as to whether the copyright of the
code in question is valid (it came out during the USL/BSDI lawsuit that
it was distributed without a copyright statement before the Berne convention
flipped the default copyright status of a work in the U.S. to "on"), but
that issue was never conclusively addressed by any court before the
USL/BSDI/UCB cases were settled. If the copyright on the code was valid,
SCO has every right to force it to not be distributed under the GPL, since
the only open-source license they ever offered it under was a BSD-style
license (if SGI had it otherwise, which clearly they did, they had it under
_much_ more restrictive licensing terms. :-( ).
Could SCO obtain punitive damages? Probably not. Could SCO obtain
injunctive relief forcing those who have distributed copies of this code
in violation of its licensing terms to get back as many copies as they
can out of the marketplace and destroy them? Maybe so. Now think
how many machines, for instance, IBM has shipped preinstalled with Linux;
the cost could be huge, and this could do real damage to companies that
have supported Linux heavily. SGI's carelessness has done a real disservice
to the Linux community here, if you ask me.
--
Thor Lancelot Simon [email]tls@rek.tjls.com[/email]
But as he knew no bad language, he had called him all the names of common
objects that he could think of, and had screamed: "You lamp! You towel! You
plate!" and so on. --Sigmund Freud
On 6 Sep 2003 16:10:21 -0700, [email]james@microcosmos.co.uk[/email] (James Holden)
wrote:
[color=blue]
>Bah... I'm soooo tired of all the SCO crap. I really don't understant
>where they're trying to get to with it all. Why can't the rest of the
>world issue a "put up or shut up" type order like our German friends
>did?[/color]
Red Hat filed just such a suit on August 4th.
Alas the US legal system is slow.
[color=blue]
>The problem is that the UNIX genepool has become so distributed of the
>years that it's totally impossible to attribute all of the original
>SysV code to any particular organisation. Also, some of the code SCO
>is disputing appears to be just textbook implementations of things
>like memory allocation or scheduling. A patent has to be "non
>obvious", but their supposed precious code is just no-brainer first
>year computer science stuff.[/color]
SCO is not arguing patents, they are arguing intellectual property
(protected by copyrights).
Dans article <1ef8c15c.0309061510.2b1b397c@posting.google.com>,
[email]james@microcosmos.co.uk[/email] disait...[color=blue]
>
> Bah... I'm soooo tired of all the SCO crap. I really don't understant
> where they're trying to get to with it all. Why can't the rest of the
> world issue a "put up or shut up" type order like our German friends
> did?
>[/color]
It's all simple : SCO's stock value rocketed since they began their
ridiculous IP crusade (x13 or 14 IIRC). The assho^W SCO executives
probably sold a lot of their stock and filled their pockets with good
dollars, and the poor stupid employees support them while they're killing
the company and destroy their jobs, doh.
--
Quis, quid, ubi, quibus auxiliis, cur, quomodo, quando?
Dans article <y8jj3T9CSXcE@ludens>, [email]edhellon@not-quite-ludens.elte.hu[/email]
disait...[color=blue]
>
> Why not? I'm just being curious here. AFAIK BSD-style licenses mean
> that it's yours for the taking, you can put it in your proprietary
> application or library or whatever, why can't you relicense it under the
> GPL?
>[/color]
Actually you can, and you must provide a file with an entry like "this
product contains code copyright UCB" or whatever.
--
Quis, quid, ubi, quibus auxiliis, cur, quomodo, quando?
Dans article <bjfh6e$geu$1@panix5.panix.com>, [email]tls@panix.com[/email] disait...[color=blue][color=green]
> >
> >Actually you can, and you must provide a file with an entry like "this
> >product contains code copyright UCB" or whatever.[/color]
>
> Um, no. The GPL states that "no further restrictions may be placed on
> the redistribution of the work"; the BSD license -- since it does not
> have null content, after all -- places precisely such restrictions.
>[/color]
It depends upon the BSD license : the latest version (freeBSD license) is
GPL compatible. see [url]www.gnu.org[/url] for details.
--
Quis, quid, ubi, quibus auxiliis, cur, quomodo, quando?
In article <MPG.19c51e6946b3bb6e98c9c8@news.free.fr>,
Emmanuel Florac <eflorac@imaginet.fr> wrote:
: The assho^W SCO executives probably sold a lot of their stock
Why say 'probably'? The insider trading reports clearly show waves upon waves of
selloffs by all the company principals.
<http://finance.yahoo.com/q/it?s=SCOX>
I gotta wonder, though, who are the suckers who are buying?
Cheers - Tony 'Nicoya' Mantler :)
--
Tony 'Nicoya' Mantler -- Master of Code-fu -- [email]nicoya@ubb.ca[/email]
-- [url]http://nicoya.feline.pp.se/[/url] -- [url]http://www.ubb.ca/[/url] --
On Sun, 07 Sep 2003 03:31:16 GMT,
Gerald Henriksen <ghenriks@rogers.com>, in
<mf9llvg9r66g69edgu0jio09eqsb03fn8s@4ax.com> wrote:
+> SCO is not arguing patents, they are arguing intellectual property
+> (protected by copyrights).
Unless they've amended their complaint, the suit against IBM is about
breach of contract and trade secrets (the weakest of the "IP pool").
James
--
Consulting Minister for Consultants, DNRC
I can please only one person per day. Today is not your day. Tomorrow
isn't looking good, either.
I am BOFH. Resistance is futile. Your network will be assimilated.
On 6 Sep 2003 20:35:42 -0400,
Thor Lancelot Simon <tls@panix.com>, in
<bjdugu$8lq$1@panix5.panix.com> wrote:
+> USL/BSDI/UCB cases were settled. If the copyright on the code was valid,
+> SCO has every right to force it to not be distributed under the GPL, since
+> the only open-source license they ever offered it under was a BSD-style
+> license (if SGI had it otherwise, which clearly they did, they had it under
+> _much_ more restrictive licensing terms. :-( ).
Yes, and? put the copyright notices back in and it becomes
redistributable as source. Problem solved.
+> Could SCO obtain punitive damages? Probably not.
Not a chance in hell, not for that particular code.
+> Now think how many machines, for instance, IBM has shipped
+> preinstalled with Linux; the cost could be huge, and this could do
+> real damage to companies that have supported Linux heavily.
That's a rather dubious arguement. At worst, one would have to remove
/usr/src/linux/, and more than likely, you'd simply have to apply a
patch to /usr/src/linux to add the original comments back in.
As far as the binary kernel goes, well, comments are ignored by the
compiler. I would not be surprised if SCO will now attempt to imply
that the simple act of compiling will cause copyright infringement.
James
--
Consulting Minister for Consultants, DNRC
I can please only one person per day. Today is not your day. Tomorrow
isn't looking good, either.
I am BOFH. Resistance is futile. Your network will be assimilated.
In article <MPG.19c583714d6a7c1298c9e2@news.free.fr>,
Emmanuel Florac <eflorac@imaginet.fr> wrote:[color=blue]
>Dans article <bjfh6e$geu$1@panix5.panix.com>, [email]tls@panix.com[/email] disait...[color=green][color=darkred]
>> >
>> >Actually you can, and you must provide a file with an entry like "this
>> >product contains code copyright UCB" or whatever.[/color]
>>
>> Um, no. The GPL states that "no further restrictions may be placed on
>> the redistribution of the work"; the BSD license -- since it does not
>> have null content, after all -- places precisely such restrictions.
>>[/color]
>
>It depends upon the BSD license : the latest version (freeBSD license) is
>GPL compatible. see [url]www.gnu.org[/url] for details.[/color]
No version of the BSD license is "compatible" with having the license
and copyright header stripped off the code. All versions require that
you keep the header intact and present in the source files as a condition
of use. SGI didn't do that.
I'll note once more that *SCO/Caldera is the copyright holder of the
code in question (the "Ancient Unix" code and the SysV code)*. SGI could
have it from them under SGI's SysV license -- which unquestionably
forbids placing it under the GPL -- or, possibly, under the BSD style
license Caldera later released the "ancient unix" code under. But neither
of those licenses allow what SGI did: stripping all attribution from the
code and pasting it into a GPLed program.
--
Thor Lancelot Simon [email]tls@rek.tjls.com[/email]
But as he knew no bad language, he had called him all the names of common
objects that he could think of, and had screamed: "You lamp! You towel! You
plate!" and so on. --Sigmund Freud
In article <slrnblosa0.dbv.sy_nttvr@gurcragntba.pbz>,
I R A Darth Aggie <no-courtesy-copies-please> wrote:[color=blue]
>On 6 Sep 2003 20:35:42 -0400,
>Thor Lancelot Simon <tls@panix.com>, in
><bjdugu$8lq$1@panix5.panix.com> wrote:
>
>+> USL/BSDI/UCB cases were settled. If the copyright on the code was valid,
>+> SCO has every right to force it to not be distributed under the GPL, since
>+> the only open-source license they ever offered it under was a BSD-style
>+> license (if SGI had it otherwise, which clearly they did, they had it under
>+> _much_ more restrictive licensing terms. :-( ).
>
>Yes, and? put the copyright notices back in and it becomes
>redistributable as source. Problem solved.[/color]
Not for all of the infringing distributions that are already out there
in the world.
[color=blue]
>+> Now think how many machines, for instance, IBM has shipped
>+> preinstalled with Linux; the cost could be huge, and this could do
>+> real damage to companies that have supported Linux heavily.
>
>That's a rather dubious arguement. At worst, one would have to remove
>/usr/src/linux/, and more than likely, you'd simply have to apply a
>patch to /usr/src/linux to add the original comments back in.[/color]
I think you're being rather optimistic about what sort of remedy a court
might require. It's not at all uncommon to require the publisher of an
infringing work to get his hands on every copy he possibly can get back
out of the market and destroy them. In IBM's case, that would probably
mean, at least, pulling every machine they have under service contract
and replacing the Linux sources, if present, all at IBM's expense. That's
not chump change we're talking about, and a lot of smaller Linux vendors
can probably afford it a lot less than IBM can. Hell, having to pull
a zillion copies of prepackaged Linux from store shelves might be enough
to convince ChumpUSA to stop stocking them, for instance.
The short-term net result for random Linux users? Probably not too bad.
The net result for companies that have supported Linux? Quite possibly
a lot of pain -- which can't be good for Linux in the long term. Again,
a real "own goal" for SGI here as far as I can see.
--
Thor Lancelot Simon [email]tls@rek.tjls.com[/email]
But as he knew no bad language, he had called him all the names of common
objects that he could think of, and had screamed: "You lamp! You towel! You
plate!" and so on. --Sigmund Freud