Why SCO won't show the code
The code in question is found in arch/ia64/sn/io/ate_utils.c in the 2.4 tree. It carries an SGI copyright. It seems that SGI was not entirely forthcoming in documenting the source of its source; some of the code in question was, indisputably, not written at SGI. So where does it really come from?
This code is from sys/sys/malloc.c in V7 Unix. It has been widely published; among other things, it can be found in Lion's Commentary on Unix (if you can get a copy). It was featured in this 1984 Usenet posting. And, crucially, it has been circulated with the V7 Unix source, which was released by Caldera (now the SCO Group) under the BSD license. SCO would like the world to forget about that release now, but the Wayback Machine remembers.
So...SCO's code demonstration, the one that it put up to convince its resellers of its case, comes from a version of Unix which first came out in 1979. The code was publicly circulated in the 1980's, and explicitly released under the BSD license by [the company now known as] SCO at the beginning of 2002. SCO might well have a complaint that SGI did not properly give credit for the code it used. But there is no possible way the company can argue that this code's presence in Linux is an infringement of its copyrights.
And this, of course, is why SCO refuses to show the code that, it claims, is copied. These claims do not stand up to even a few hours' scrutiny on the net. SCO may yet have an interesting contract dispute with IBM, but, from what we have seen so far, its claims of direct copying of code are hollow.
(Many thanks to those who commented on an
earlier LWN posting on this subject - those comments are the source for
just about everything that appears in this article. Many thanks are due to
LWN's readers; you have shown the best of what the community can do.
Update: see also: this analysis of SCO's
code by Bruce Perens.)
(Log in to post comments)
Why SCO won't show the code
Posted Aug 19, 2003 16:08 UTC (Tue) by rfunk (subscriber, #4054) [Link]
Don't forget about the official publication of the Lions book, with theSanta Cruz Operation's blessing, in 1996. :-)
Why SCO won't show the code
Posted Aug 19, 2003 16:15 UTC (Tue) by rfunk (subscriber, #4054) [Link]
Oh yeah, and if we go by the Lions book (Unix v6) example, then we candocument that the code dates back to 1976, rather than the 1979 date of
32V.
(I know, I'm picky.)
Why SCO won't show the code
Posted Aug 19, 2003 18:36 UTC (Tue) by ken (subscriber, #625) [Link]
ls -l ./usr/sys/ken/malloc.c-rw-r--r-- 1 ken ken 1119 Nov 27 1974 ./usr/sys/ken/malloc.c
I found it in V5 also.
The comment in that file is not the same and some whitespace differs but
it is the same code.
This is 30 year old source code!
Posted Aug 19, 2003 19:09 UTC (Tue) by mdrejhon (guest, #14189) [Link]
Found the V5 source code in its entirety:Photo of SCO Slide:
http://www.heise.de/newsticker/data/jk-19.08.03-000/imh1.jpg
Near exact match to 30 year old source code in V5:
http://minnie.tuhs.org/UnixTree/V5/usr/sys/ken/malloc.c.html
What the heck is SCO doing, claiming 30 year old source code as their own?
Why SCO won't show the code
Posted Aug 19, 2003 19:11 UTC (Tue) by rfunk (subscriber, #4054) [Link]
Hmm, you're right. In my Jan 2002 copy of the tuhs.org site I found PDP-11/Distributions/research/Dennis_v5/v5root.tar.gz, which contains this /usr/sys/ken/malloc.c:
#
/*
* Copyright 1973 Bell Telephone Laboratories Inc
*/
struct map {
char *m_size;
char *m_addr;
};
malloc(mp, size)
struct map *mp;
{
register int a;
register struct map *bp;
for (bp = mp; bp->m_size; bp++) {
if (bp->m_size >= size) {
a = bp->m_addr;
bp->m_addr =+ size;
if ((bp->m_size =- size) == 0)
do {
bp++;
(bp-1)->m_addr = bp->m_addr;
} while ((bp-1)->m_size = bp->m_size);
return(a);
}
}
return(0);
}
mfree(mp, size, aa)
struct map *mp;
{
register struct map *bp;
register int t;
register int a;
a = aa;
for (bp = mp; bp->m_addr<=a && bp->m_size!=0; bp++);
if (bp>mp && (bp-1)->m_addr+(bp-1)->m_size == a) {
(bp-1)->m_size =+ size;
if (a+size == bp->m_addr) {
(bp-1)->m_size =+ bp->m_size;
while (bp->m_size) {
bp++;
(bp-1)->m_addr = bp->m_addr;
(bp-1)->m_size = bp->m_size;
}
}
} else {
if (a+size == bp->m_addr && bp->m_size) {
bp->m_addr =- size;
bp->m_size =+ size;
} else if (size) do {
t = bp->m_addr;
bp->m_addr = a;
a = t;
t = bp->m_size;
bp->m_size = size;
bp++;
} while (size = t);
}
}
Copyright (C) Caldera International Inc. 2001-2002. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
- Redistributions of source code and documentation must retain the above copyright notice, this list of conditions and the following disclaimer.
- Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
- All advertising materials mentioning features or use of this software
must display the following acknowledgement:
This product includes software developed or owned by Caldera International, Inc.
- Neither the name of Caldera International, Inc. nor the names of other contributors may be used to endorse or promote products derived from this software without specific prior written permission.
USE OF THE SOFTWARE PROVIDED FOR UNDER THIS LICENSE BY CALDERA INTERNATIONAL, INC.AND CONTRIBUTORS ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL CALDERA INTERNATIONAL, INC. BE LIABLE FOR ANY DIRECT, INDIRECT INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Why SCO won't show the code
Posted Aug 19, 2003 23:50 UTC (Tue) by Xman (guest, #10620) [Link]
Note that because this has the advertising clause, it's actually GPL incompatible.
Copyright date
Posted Aug 21, 2003 20:31 UTC (Thu) by Ross (guest, #4065) [Link]
This isn't really relavant to the current issue, but how can they put a2002 copyright date on a file which was not changed since written and
published in 1973? The file even contains the original 1973 copyright
notice.
Seems like an invalid copyright extension to me.
Why SCO won't show the code
Posted Aug 19, 2003 19:37 UTC (Tue) by rfunk (subscriber, #4054) [Link]
"nsys" code, Aug 1973, dating between v3 and v4.http://minnie.tuhs.org/UnixTree/Nsys/sys/nsys/dmr/malloc.c.html
The v6 code (May 1975) is where these comments are introduced.
http://minnie.tuhs.org/UnixTree/V6/usr/sys/ken/malloc.c.html
SCO's copyrights
Posted Aug 19, 2003 16:27 UTC (Tue) by rfunk (subscriber, #4054) [Link]
SCO might well have a complaint that SGI did not properly give credit for the code it used. But there is no possible way the company can argue that this code's presence in Linux is an infringement of its copyrights.
Two words: advertising clause.
even without the advertising clause
Posted Aug 19, 2003 17:52 UTC (Tue) by stevenj (guest, #421) [Link]
Even without the advertising clause, the BSD license contains:Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.and essentially every other license contains a similar condition. IANAL, but it seems like there is a good chance that this code has been used illegally in the kernel if SGI replaced the copyright notice with its own. It's hard to imagine substantial claims of damages, of course.
(Note that it was a proprietary software company that did this, moreover, not the horrible hordes of hairy hackers. But, to be fair, I think that very few programmers are careful about this sort of thing.)
The code's being published in a book doesn't remove its copyrighted status or allow re-distribution, by the way, so that argument is a red herring (except regarding trade-secret claims).
SCO's copyrights
Posted Aug 21, 2003 6:33 UTC (Thu) by bojan (subscriber, #14302) [Link]
My understanding that this stuff appears in BSD 2.11 as well and under Regents of the University of California copyright:http://unix-archive.pdp11.org.ru/PDP-11/Trees/2.11BSD/sys/sys/subr_rmap.c
These guys have officially killed the advertising clause in 1999 on all versions of BSD software:
ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change
So, I'm guessing all the code is "off the hook", right?
Bojan
Make copies of that wayback machine page!!
Posted Aug 19, 2003 16:32 UTC (Tue) by rknop (guest, #66) [Link]
Not that it's a real worry, but:Wayback machine does let site owners remove things from the archive. I know, because I've done this. (www.dramex.org has plays people have put online, and sometimes they ask us to remove them. Once I discovered wayback machine, I told them to remove the old plays and stop archiving the new ones.)
SCO may figure out that they have left tracks behind and try to erase them. Print-outs or copies of those tracks should be made before they do this. (I'm to lazy to do this myself, but *somebody* probably ought to.)
Of course, I wouldn't worry, because this is the same company that has distributed Linux on it's own ftp site while claiming that those who distribute Linux under the GPL are violating its intellectual property. This is a company whose left hand doesn't know what its left hand is doing. But, still.
-Rob
Make copies of that wayback machine page!!
Posted Aug 19, 2003 22:50 UTC (Tue) by mmarq (guest, #2332) [Link]
Oh,no!But wich ones ?
I dont belive that they will throw more fiascos as this, plently commented in this site.
Now that the super fast response of the Linux/OSS community seamed to have made a hole in the SCO FUD warchest, and mark a clear victory... you are telling that they could be back in business ?
Guess that there must be copies of "everything" UNIX related,... maybe OSI could hire a guy to google and copy everything Unix related,... how much days could it take ?
Make copies of that wayback machine page!!
Posted Aug 19, 2003 23:58 UTC (Tue) by Arker (guest, #14205) [Link]
Call me a paranoid, but it has saved my life at least once.I won't say there's no worry here. Please someone archive this stuff on your personal machine. And don't tell anyone it's there. Just keep it until it's needed, or this mess is over.
I'd just say I've done that myself, as I've done in past cases (I have an untouched copy of 2.4 source from Caldera for instance,) but it's almost 2am in my timezone and I've done enough for the day. I know there are thousands of geeks who haven't, and I know a lot of us have a little hard drive space to spare. Grab this stuff. If only one of us has it, it means nothing, but if a couple hundred have byte-identical copies with the same time and date and the same story on how it was obtained, we have a legal chain of evidence that can be proven beyond a reasonable doubt. So please, just in case, do it now. Burn it to a CD or something, along with a description of exactly when and how you obtained it. You'll almost certainly be wasting a CD, but they're cheap, and if it does become an issue, you'll be glad you did.
I'm going to bed now, I leave it up to you.
Make copies of that wayback machine page!!
Posted Feb 11, 2006 23:26 UTC (Sat) by kbolino (guest, #35879) [Link]
It was a real worry. SCO has used robots.txt, which the company that operates the Wayback Machine recognizes and applies retroactively, to block access to all of its pages. I hope that someone did get these saved, because SCO apparently doesn't want their own history to be known.
Why SCO won't show the code
Posted Aug 19, 2003 16:36 UTC (Tue) by apwiggins (guest, #14171) [Link]
It will be interesting to see the reactions or silence from Laura Didio and Rob Enderle, given their past statements, analysis, and opinions.==============================
http://lwn.net/Articles/41516/
http://lwn.net/Articles/37013/
==============================
The Lions book (more nitpicking)
Posted Aug 19, 2003 17:05 UTC (Tue) by rfunk (subscriber, #4054) [Link]
The title of the Lions book is actually "Lions' Commentary on UNIX 6th Edition, with Source Code", and the author is John Lions.The "6th Edition" part is part of the title (actually part of the name of "UNIX 6th Edition"), and refers to the version of Unix it covers. It is not the 6th edition of the book, as Amazon and others seem to think. (The "with Source Code" part is the subtitle.)
Also, it's Lions', not Lion's, since the author's name is John Lions rather than John Lion.
The Lions book (more nitpicking)
Posted Aug 19, 2003 18:15 UTC (Tue) by ikegami (guest, #14174) [Link]
: Also, it's Lions', not Lion's, since the author's name is John Lions rather than John Lion.
If you're gonna nitpick on spelling, get it right! The possessive form of "Lions" is "Lions's" because it's singular.
(singular) pro, James, Lions
(plural) pros, Jameses, Lionses
(possessive singular) pro's, James's, Lions's
(possessive plural) pros', Jameses', Lionses'
The Lions book (more nitpicking)
Posted Aug 19, 2003 18:27 UTC (Tue) by rfunk (subscriber, #4054) [Link]
If you're gonna nitpick on spelling, get it right!
I did. I gave the title of the book as it is on the cover, not what some manual on English says it should be. Then I explained why it's that way on the cover.
The possessive form of "Lions" is "Lions's" because it's singular.
That's a matter of some debate. For example:
http://www.meredith.edu/grammar/plural.htm#Possessive
http://www.alt-usage-english.org/excerpts/fxwheret.html
It's a living human language, therefore there are inconsistencies and uncertainties. Deal with it. The title of this book, however, is not uncertain.
The Lions book (more nitpicking)
Posted Aug 20, 2003 8:19 UTC (Wed) by rjamestaylor (guest, #339) [Link]
ikegami, I'd give it up. He's likely to bring in his partner Wagnall to buttress his arguments. Then you'd really be toast.;)
The Lions book
Posted Aug 19, 2003 23:47 UTC (Tue) by nicku (subscriber, #777) [Link]
John Lions taught me operating systems at The University of New South Wales. Sadly he's dead now, though the overseas students who suffered under his criticism of their standard of English may not regret so strongly.I have the two books from the class, which I treasure; one is A commentary on the UNIX Operating System, fourth printing, the other is UNIX Operating System Source Code Level Six. There, on Sheet 25 is the code that matches the slide.
They are printed on the big old Computer Centre line printers in fixed width font on A4 paper, landscape orientation, bound in yellow (the commentary) and red (the code). The commentary is printed justified in three columns, the code in two.
Why SCO won't show the code
Posted Aug 19, 2003 17:24 UTC (Tue) by dwalters (guest, #4207) [Link]
Good. Now if you're a journalist, perhaps you would consider doing the SCO shareholders a favour, and write about this.I keep seeing FUD articles about SCO in eWeek, CNet, etc. It would be nice to read about some fact instead of FUD for a change.
Why SCO won't show the code
Posted Aug 19, 2003 18:37 UTC (Tue) by freethinker (guest, #4397) [Link]
A favor? Publishing a story that would only drive the share price down is a favor? :)
Why SCO won't show the code
Posted Aug 19, 2003 19:06 UTC (Tue) by dwalters (guest, #4207) [Link]
What I mean is do the shareholders a favour in terms of educating them, so that they can make better informed decisions about buying and selling SCO stock.To my mind, the simple fact that SCOX is trading at over $10 a share, tells me that the shareholders may only be hearing one side of the story.
Why SCO won't show the code
Posted Aug 19, 2003 17:36 UTC (Tue) by gups (guest, #14053) [Link]
I wonder what McBride has to say about this. He will probably come up with defensive statements even more moronic than ones he made before.
McBride's volley of folly
Posted Aug 20, 2003 8:43 UTC (Wed) by rjamestaylor (guest, #339) [Link]
He won't care about these revelations. Instead of trying to come up with more direct copying, he'll attack in three directions:
1) Eternal derivatives
Using the answer that since UNIX was taught in university Comp-Sci of course commonly known code and methods will show up in Linux, he'll argue that all the knowledge required to create a modern OS came from UNIX (it didn't) and therefore the eternal derivatives clause of the AT&T license prevails. This will be difficult to prove, but it must be preferable to him than just selling the office furniture and calling it quits.
2) GPL/BSD license nullification: targeting Caldera
He isn't trying to nullify the GPL to screw Linus, Alan, et al, but rather to screw Caldera, that is, to nullify Caldera's previous management's action of releasing Unix 1-7 to the public under the BSD-style and GPL-style licenses. Once he revokes that he'll use his Eternal derivatives theory to claim the entire Linux-base has his own. (This is also why they have no trouble dissing the GPL in the morning and announcing SVR6 with major features provided under GPL, MPL, APL, BSD, et al, licensing -- since it builds on UNIX, it's theirs, thanks for the free work now give-it-to-ME!).
3) Contractual breaches with UNIX licensees
They're going to nail SGI. We gave them that little nibblet through our research. But they will continue to hammer away on the contractual issues, which I don't believe but I have to admit not understanding the intracacies of contract law.
I'm not on McBride's side (in fact I'm TheOneToWhomMyRefers on Yahoo! Finance [Hi, Walter]). I don't think he will prevail. I also don't think he'll be working in technology once this is over -- IOWs, he's burnt the bridges, put up the walls, and peed in the pool. This is larger than SCOX vs IBM. This is the battle for the legitimacy of Open and Free software but also proprietary software built using common techniques and learned practices. If SCOX prevails then the most innovative technologies will belong to a company that gave up innovation for litigation to stave off corporate death. I expect even Microsoft to come around and oppose SCOX, at least on point #1.
Why SCO won't show the code
Posted Aug 19, 2003 17:47 UTC (Tue) by BrucePerens (guest, #2510) [Link]
The wayback machine page says the code is for educational use only, this is belied by Caldera's license letter, which says it is under the BSD license.Bruce
It seems we do have a real copyright/license violation then
Posted Aug 19, 2003 17:55 UTC (Tue) by JoeBuck (subscriber, #2330) [Link]
Then it appears that Linux does have a problem that must be corrected. There is no Caldera copyright notice in the Linux copy of the code, and the license described in the letter includes the GPL-incompatible advertising clause.
It seems we do have a real copyright/license violation then
Posted Aug 19, 2003 18:05 UTC (Tue) by hch (guest, #5625) [Link]
Well, that whole file already is gone from Linux for some time(a month or two).
Removed due to severe crappieness...
Information on "SCO Ancient Unix" web page appears to be factually incorrect.
Posted Aug 19, 2003 18:10 UTC (Tue) by dwalters (guest, #4207) [Link]
Indeed. The wording on the web site is:
Use of all versions is strictly for educational use only, and is restricted by the license agreed upon by entering this site.
If the code was already released under the BSD-style license, the above statement is inaccurate and misleading. The BSD license-covered code is most certainly not limited to educational use only. Either SCO doen't realize this, or they're trying to close the stable door after the horse has bolted!
However, unless the code in question was also released under some other BSD-license which does not include the Caldera advertising clause, any advertising materials mentioning features or use of the code must acklowledge Caldera to be in compliance with the license.
Information on "SCO Ancient Unix" web page appears to be factually incorrect.
Posted Aug 20, 2003 3:20 UTC (Wed) by lakeland (guest, #1157) [Link]
If it is their code, they're allowed to dual licence it. They can quite happily give it awayunder BSD, and now under a more restrictive licence. However it does seem somewhat
pointless since anybody can get the old BSD version.
The "Greek" says (in English); Typoe; WayBack
Posted Aug 19, 2003 18:03 UTC (Tue) by leonbrooks (guest, #1494) [Link]
"As part of the kernel evolution toward modular naming, the functions malloc and mfree are being renamed to rmalloc and rmfree. Compantibility will be maintained by the following assembler code: (also see mfree/rmfree below)". Greek is phonetic: I can read real Greek (badly) and Greek listeners will understand what I'm saying (while wincing at my accent) even if I don't. So just treat the text like a cypher and it's easy.If you presume that the editor was in a hurry, all that happened was the they didn't go rightwards enough words (or perhaps went back a word instead of forward) before appending the closing parenthesis to the ulong_t cast.
Naturally, all of this is moot since the code is BSD-derived anyway.
WRT WayBack, Reasons To Believe moved to a totally dynamic website not long ago, even for pages with static content, and the only reason I can see for this is that archive.org will then not cache it, but there will be no record of them (nor will they ever have to admit) asking WayBack to remove it. The reason for this is that their story changes often, and if their ideas aren't crazy and inconsistent enough as-is, any archive of their site would completely shred the credibility of what remains.
In point of fact TSG's conference stuff (from which I drew a contact list) isn't archived, and was hurriedly pulled by TSG (who will say "but it finished" despite leaving last year's page up for ages), but Google still has a cache of it as at now.
The "Greek" says (in English); Typoe; WayBack
Posted Aug 20, 2003 0:33 UTC (Wed) by neoprene (guest, #8520) [Link]
http://www.heise.de/newsticker/data/jk-19.08.03-000/imh0.jpgTHE TWO SECOND CRYPTO LESSON:
It is not greek, just greek letters. Most word-editors allow you to switch fonts,
switch to AA-Symbol and you look like a learned man already ;)
* As part of the kernel evolution
toward modular naming, the
* functions malloc and mfree are being
renamed to rmalloc and rmfree.
Compatibility will be maintained by
the following assembler code:
* (also see mfree/rmfree below)
* /
Is this supposed to be the great mystery code SCOX is hiding?
So lame.
The "Greek" says (in English); Typoe; WayBack
Posted Aug 20, 2003 2:10 UTC (Wed) by Jotham (guest, #14211) [Link]
Court Summons #45214,Dear Sir,
You have just HACKED our encryption method and further *published* your method. This is in clear controvention of the DMCA act and show that you and those in collusion, LWN.net, are clearly terrorists. We estimate that this has damaged our company to in excess of 1 billion dollars in lost revenue and FUD. See you in court.
Sincerely and best wishes,
Dr Evil
SCO Legal Department
The "Greek" says (in English); Typoe; WayBack
Posted Nov 4, 2005 16:27 UTC (Fri) by rmosler2100 (guest, #33631) [Link]
Uh oh... I think that you just violated DMCA by breaking their encryption with that devious program that lacks any legitimate use... I think it is called MS Word.
You should get a lawyer because SCO will be coming for you.
Bee-gees are imortal :-)
Posted Aug 19, 2003 18:33 UTC (Tue) by nraynaud (guest, #532) [Link]
The brand new ia64 code of the kernel embbeds a '76 disco-sytle first-fit allocator Che !http://www.flowgo.com/greetings/Disco_Penguin/Disco_Penguin.gif
PEACE MY BROTHER !
Well, more seriously, why a so bad designed allocator falled here ? I just don't care the legal problem, but on a technical point-of-view, I think the power-flower mallocs are a little bit over-dated.
Should somebody call the SEC?
Posted Aug 19, 2003 19:01 UTC (Tue) by tavis (guest, #14187) [Link]
The comments above show that McBride's presentation either deliberately misled shareholders about the status of the code, or that SCO didn't perform due diligence in invesigating the code's history, in a situation where the management team had a personal interest in not performing it. (Regardless whether the code is legitimately GPLed, SCO is clearly not in a position to claim monetary damages from the code's use.) In other words, he made misleading statements at a shareholder's meeting that he knows will inflate the value of the stock. This is highly illegal. Somebody should call the SEC to investigate.
Should somebody call the SEC?
Posted Aug 19, 2003 20:01 UTC (Tue) by tjc (guest, #137) [Link]
The comments above show that McBride's presentation either deliberately misled shareholders about the status of the code, or [snip]I'm curious if this is part of the code that SCO claims Linux "stole" and used to provide enterprise capabilities that it would otherwise not have. If so, it's interesting that a Caldera Product Manager and a Public Relations Manager referred to this code as "ancient."
If this is the best SCO has to show, then they're up to their armpits already.
Should somebody call the SEC?
Posted Aug 21, 2003 23:00 UTC (Thu) by onetimepost (guest, #14340) [Link]
I think its a moot point. I have just purchased the rights to the letter 'S' from Sesame Street and intend to come after every body that has ever used it to make any money what so ever. As you know, the letter is prominently displayed in the trademark SCO, so they are toast. As for the rest of you I demand that you immediately stop using the letter 'S' with out licensing it from me. I will however offer you a special introductory rate of $25 per occurrence for any existing published work that contains this wonderful letter. My lawyers will be contacting your lawyers if this royalty payment isn't submitted at once !!!!!
In other news ..
Posted Aug 19, 2003 19:08 UTC (Tue) by nx_in (guest, #14162) [Link]
The development version of the linux kernel (2.5) has achieved a significant improvement ( 20% ) in its memory allocation routines (malloc) by removing one file containing an ancient malloc implementation developed by The SCO group.
Why SCO won't show the code
Posted Aug 19, 2003 19:26 UTC (Tue) by TheQuietGuy (guest, #14190) [Link]
Did anyone else notice that the slide shown in the second link, allegedlyshowing code in the Linux kernel, shows code that could not be compiled
successfully because it has a syntax error?
Why SCO won't show the code
Posted Aug 19, 2003 19:35 UTC (Tue) by lyda (guest, #7429) [Link]
these people have written articles on sco's side of the story. i think people should politely point them at this article: peter_galli@ziffdavis.com, prooney@cmp.com, LDiDio@yankeegroup.com, lisa.bowman@cnet.com.
Editors, please note this.
Posted Aug 19, 2003 19:53 UTC (Tue) by nx_in (guest, #14162) [Link]
The code which they are displaying as from linux kernel, is NOT actually from linux kernel. See in the second picture: if (size == 0) return) ((ulong_t NULL); Now, this code doesn't even compile!! The actual code that exists in linux is this: if (size == 0) return((ulong_t) NULL); Now, this could be really dangerous. Moreover, this file is part of only the ia64 port which not even 2% of linux users use. Just why I am supposed to pay SCO ??
Re: Editors, please note this.
Posted Aug 19, 2003 21:14 UTC (Tue) by lsweeks (guest, #14198) [Link]
Surely you can't mean the same ia64 release of Linux that SCO released on April 15th of this year. That would mean that they had released it under the GPL --AFTER-- they had already made the determination that it was their UNIX-derived code. Surely those clever folks wouldn't make a mistake like that. (o;)
DNA Argument
Posted Aug 19, 2003 19:53 UTC (Tue) by burbank15 (guest, #6401) [Link]
The code in question may be relevant if SCO is really intending to press the "DNA" argument. The quote from McBride stating "The DNA of Linux is coming from Unix" is interesting. Combined with earlier statements concerning IP I wonder if this isn't another argument they're developing.Granted I would be surprised if this is really what will form the basis of their case. How could you argue that ideas and processes that are available in any number of college texts without requiring an NDA can really belong to one company? Wouldn't this line of reasoning require that any programmer who attended any classes on OS design in the last 30+ years be willing to enter into some sort of contract with SCO. And SCO would have to convince the courts that the OS design theories that influenced Unix can be used without any contractual encumbrance, while processes developed for Unix do have contractual encumbrances.
*sigh* The more I read about this case, the more I think that someone such as Arianna Huffington should be writing a column about SCO. Any suggestions for the title? Note that she has already used "Pigs at the Trough" for a book.
DNA Argument
Posted Aug 19, 2003 22:28 UTC (Tue) by nexex (guest, #14202) [Link]
someone such as Arianna Huffington should be writing a column -- I'm sorry but I think she is the last person qualified for something like that. She changes positions so many times her next book should be called Kama Sutra. She launches a run for Cali governor attacking people and businesses who use "loopholes" to lessen their tax burden, but lo and behold, she hardly payes taxes herself.
DNA Argument etc..
Posted Aug 21, 2003 2:38 UTC (Thu) by maguska (guest, #14290) [Link]
Well, Mr McBride might have forgotten that his DNA is more than 80% identical to Caenorhabditis Elegans (a "nice" worm with 302 neurons). I hope the small 20% difference will be enough to make him able to understand our reasons and pursuade him.the other:
Since Linux is open source, SCO could also used it, and it's hard for them to prove that Linux is the copy and not their UNIX parts.
3rd:
They could also show the copied parts in Linux's open source code, they don't have to show their one. This would make it possible to rewrite the problematical parts without signing secrecy agreements (it has another name, but I've forgotten it by now :) ).
4th:
I was thinking about copyright law, and..
It would be funny if - for instance - the A-bomb team (A. Einstein, Fermi, Neumann Janos, Wiegner Jeno, Szilard Leo and Teller Ede) announce sueing Russia breaking A-bomb copyright (or patent, I don't know which). Teller Ede is still living, so it could be happen!
again, excuse me for grammatical mistakes
Maguska
Why SCO won't show the code
Posted Aug 19, 2003 20:08 UTC (Tue) by ak_hepcat (guest, #14192) [Link]
With this recent relevation, and the possibility that the evidence could disappear before it ever made it to court, what is the possibility of getting the FSF's lawyers to file an injunction or whatever it is, to stop SCO from trying to delete the evidence from The Wayback Machine?Seems like if I were DMcB (i'd kill myself, ha!) that would be the first thing that I'd sic my hounds on -- destroying the evidence..
The second, and third, shoes
Posted Aug 19, 2003 20:48 UTC (Tue) by ccchips (subscriber, #3222) [Link]
LWN: Are you aware that this revelation has really made it hard to get through to you?Second shoe: When I was in NORML, I saw news "reports", on major television outlets, of sperm swimming in circles due to marijuana use. The scene they showed was actually a snippet from an old educational film about what sperm looked like under a microscope, while the announcer boldly announced these "important findings" by "research groups." Could this "code" actually be just something they threw up on a slide to mislead people that they were showing evidence?
The third shoe:
On this forum and otheres, I've occasionally seen quiet, second-hand and third-hand posts (a friend, a friend of a friend who worked at SCO) about SCO employees doing the *exact opposite* of what they claimed IBM did, but with no specification of where or when; in other words, these people claimed that they knew SCO employees who stole code, and were encouraged to do so.
I am waiting. If this is true, it's got to be heavy on *someone*'s conscience.
Getting through to LWN
Posted Aug 19, 2003 21:13 UTC (Tue) by corbet (editor, #1) [Link]
"LWN: Are you aware that this revelation has really made it hard to get through to you?"Trust us, we noticed. Our bandwidth is there, but our poor server is sweating under the load. I think there's people at Rackspace standing by with fire extinguishers.
That's why we should have released the article under subscription - it would have kept the load down.
Seriously, though, it's clear that a stronger server has got to be in our future plans at some point.
Getting through to LWN
Posted Aug 20, 2003 7:51 UTC (Wed) by ekj (guest, #1524) [Link]
It's ok. If your server is sweating, it means lots of people is reading this.Lots of people reading this can only be good. Let it sweat. Keep the extinguishers handy.
Getting through to LWN
Posted Aug 20, 2003 16:43 UTC (Wed) by colink (guest, #274) [Link]
Let this be a notice to all of us who aren't currently subscribed. LWN is not just looking to make a few bucks. They published this openly because it's very important that the community they server know this now! If you like LWN, think about giving up two cups of Starbucks per month and support these guys. I just did.And why doesn't somebody like OSDL just buy y'all? Wouldn't it be great if a n industry wide consortium of companies interested in Linux gave back to the community that way?
Getting through to LWN
Posted Aug 21, 2003 0:04 UTC (Thu) by Germ (guest, #14284) [Link]
I'll subscribe if they change this awfull color scheme...it hurts my eyes and offends my sense of asthetics...
Getting through to LWN
Posted Aug 21, 2003 0:36 UTC (Thu) by corbet (editor, #1) [Link]
You can change the color scheme any time you like; just go into the customization section of the "my account" screen.Awaiting your subscription...:)
Color changing is easy.
Posted Aug 21, 2003 5:43 UTC (Thu) by frazier (guest, #3060) [Link]
Sure enough, I changed the left column background to #ffffcc and I'm back to classic LWN color. Changing "Headline background" to #ffffff pretty much completes the look.That was easy.
Clean BSD code
Posted Aug 19, 2003 20:59 UTC (Tue) by muon113 (guest, #14196) [Link]
Thanks to Arker who pointed this out on slashdot. The code actually traces back to earlier origins. The code can be found in 2.11 BSD and later versions (which, as we all know thanks to AT&T v. University of California, is clean). A copy of the code is here: http://unix-archive.pdp11.org.ru/PDP-11/Trees/2.11BSD/sys/sys/subr_rmap.c
Clean BSD code
Posted Aug 19, 2003 23:19 UTC (Tue) by Arker (guest, #14205) [Link]
Hey,Thanks for crediting me. Just because of you, I made an account here.
I'm always late on these things. I've been posting to slashdot since just after they went live, but I have an embarrassingly high uid because I never bothered to register until it became necessary because of the trolls.
Enough about me, the case at hand, much as I'd like to agree with you I don't. It's my understanding that BSD before 4.4lite is not necessarily unencumbered. If someone can show that I'm wrong on that please do, I'd love to hear it.
However, the good news is that the comment does not prove copyright infringement, only the code could do that, (although the comment could be used as evidence to help convince a judge that this was a cut and paste rather than just a tight algorithm which any competent programmer would have hit on quickly) and that at worst what we have here is a failure of a contributor to include a legally required copyright notice, with actual damages being the statutory limit to the award against him.
So, IANAL but it looks to me like the best SCO can get out of this particular example is $1 from whoever originally contributed the code. No? Why not?
Do they think the BSD license is restrictive?
Posted Aug 19, 2003 21:09 UTC (Tue) by iabervon (subscriber, #722) [Link]
If I understand their claims about the GPL, they are saying that the conditions the GPL places on distribution (etc) are insufficient to prevent the work from falling into the public domain, and thus anything properly licensed under the GPL may be treated as public domain. (That is, they don't mean that the GPL is an invalid license, which would make them massive copyright violators in their own eyes, if no one else's, but rather that releasing a work under the GPL removes any copyright protection on it, and thus they can use it without being bound by the terms of the GPL)Surely if the GPL's restrictions are not sufficient to prevent a work from being essentially in the public domain, the BSD license, which has nothing but the advertizing clause to make it more restrictive than the GPL is similarly unenforceable. This means that they must believe the code on which they're founding their claims to be in the public domain.
Too trivail to copyright (pseudo code)
Posted Aug 19, 2003 21:28 UTC (Tue) by darkonc (guest, #14197) [Link]
Looking at that chunk of malloc code, it is extremely functional. It is a very straightforward and minimal implemetation of first-fit memory allocation from a free pool.static struct ( size_t m_size, char *m_addr } *chunk;
While(more chunks){
if current chunk biggern than request {
take what we need out of the chunk
point the pool pointer to the rest of the chunk
adjust the size indicator.
if we're using the entire chunk,{
move this node to the end of the list.
# (so it doesn't block the search) #
} #endif
return pointer }#endif
}#endwhile
# couldn't find a big enough chunk
return(NULL) # error
It would be pretty difficult to produce a tight version of this algorithm without a high degree of duplication. I'd say you might as well cut and paste, because about the only changes that I can see making in a tight implementation would be to change the variable names. You'd be lucky to find 4 meaningful permutations of this algorithim, once you tighten up the code for the kernel.
Try to implement the pseudo code above, and see just how far away you end up.
BTW, this is not part of a block of duplicate code.. This is pretty much the entire thing. If that's the best that they can find, then they're SOL.
If you're not on somebody's shit list, you're not doing anything worthwhile.
Too trivial to copyright
Posted Aug 19, 2003 22:15 UTC (Tue) by rfunk (subscriber, #4054) [Link]
It would be pretty difficult to produce a tight version of this algorithm without a high degree of duplication.
Indeed, In his 1977 commentary, John Lions wrote about malloc and mfree:
The code for these two procedures has been written very tightly. There is little, if any, "fat" which could be removed to improve run time efficiency. However it would be possible write [sic] these procedures in a more transparent fashion.
Too trivail to copyright (pseudo code)
Posted Aug 19, 2003 22:17 UTC (Tue) by mammothrept (guest, #14201) [Link]
"It would be pretty difficult to produce a tight version of this algorithm without a high degree of duplication. I'd say you might as well cut and paste, because about the only changes that I can see making in a tight implementation would be to change the variable names. You'd be lucky to find 4 meaningful permutations of this algorithim, once you tighten up the code for the kernel."
The implication of what you are saying is that this section of code is not copyrightable under US law, regardless of whether a copyright notification is attached. First, algorithms themselves are not copyrightable because of a legal doctrine called the idea/expression merger. Ideas cannot be copyrighted. Only specific expressions of ideas can be. Further, if there are only a small number of ways to express an idea, then even the expression is not copyrightable. If there are only "4 meaningful permutations of this algorithm" then a developer could block anyone else from using this idea merely by writing the algorithm four different ways. This would be using copyright to attain patent-like protection which is not allowed.
Not even patentable... 25 year limit just about over
Posted Aug 20, 2003 14:53 UTC (Wed) by egberts (guest, #14248) [Link]
This is not even patentable either.So, just run it through 'ident' and recopyright it.
Why SCO won't show the code
Posted Aug 19, 2003 21:46 UTC (Tue) by renco (guest, #11927) [Link]
For the ancient licensesThe link to the ancient Unix license is removed!
You first need to obtain an Ancient UNIX license from SCO's web site. Once you have clicked on the license, go to the bottom and follow the hyperlink to obtain access to the Unix Archive.
After you fill in your details, you will be emailed the access details. The go to the Unix Archive sites list to find your nearest mirror of the archive.
Once you have been granted access, you can also order the Unix Archive on CD-ROM or other media. This is done on a volunteer basis, so it may take some time. Please volunteer to help out with this effort if you can.
SCO is making a lot of noise. I think I agree with one statement from SCO if I see the ode and the history. Linux is not the operating system that makes a lot of innovation. Do not use samba also anymore in you new OS.It also GNU !!
Why SCO won't show the code
Posted Aug 19, 2003 22:36 UTC (Tue) by djabsolut (guest, #12799) [Link]
I think I agree with one statement from SCO if I see the ode and the history. Linux is not the operating system that makes a lot of innovation.And you're making this conclusion based on a simple memory allocator?? There's not a lot of ways to implement a memory allocator like that, as one of the previous posts above elegantly demonstrated. There's a _lot_ more to the Linux kernel than just this basic allocator. Try SMP, try the scheduler, try the VM, try integration with XFree, various journaling file systems... one could go on and on.
Why SCO won't show the code
Posted Aug 19, 2003 22:09 UTC (Tue) by spshealy (guest, #14200) [Link]
Showing the code doesn't fit in with their business plan...I ran across the following flash cartoon with Darl in it... Pretty funny!
http://www.justinsanedesign.com/staff/justin/penguin.swf
Why SCO won't show the code
Posted Aug 20, 2003 3:42 UTC (Wed) by vijayandra (guest, #13932) [Link]
The James Bond visuals and music indicates McBride's frustration.
Ancient Unix Mirror
Posted Aug 20, 2003 6:57 UTC (Wed) by mcbridematt (subscriber, #10302) [Link]
It turns out that PlanetMirror is still mirroring the ancient unix files
Why SCO won't show the code
Posted Aug 20, 2003 10:02 UTC (Wed) by luckybit (guest, #14235) [Link]
Standards the USA do not get under a copyright. The decision of court BSD vs. ATT
Perens' analysis mirror
Why SCO won't show the code
Posted Sep 1, 2003 14:50 UTC (Mon) by lundberg1000 (guest, #14689) [Link]
In Sweden SCO has a very bad image recently. Also this hurts the image of Utah. I belong to an interesting forum on this which is open http://groups.yahoo.com/group/scofanclub/It would be interesting to have the opinion of more Linux people.
Lars