LWN: Comments on "Why SCO won't show the code"

Make copies of that wayback machine page!!

kbolino — Sat, 11 Feb 2006 23:26:42 +0000

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.

The "Greek" says (in English); Typoe; WayBack

rmosler2100 — Fri, 04 Nov 2005 16:27:07 +0000

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.

Why SCO won't show the code

lundberg1000 — Mon, 01 Sep 2003 14:50:08 +0000

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

Should somebody call the SEC?

onetimepost — Thu, 21 Aug 2003 23:00:38 +0000

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 !!!!!

Copyright date

Ross — Thu, 21 Aug 2003 20:31:08 +0000

This isn't really relavant to the current issue, but how can they put a
2002 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.

Perens' analysis mirror

jbh — Thu, 21 Aug 2003 14:02:20 +0000

[mirror 1] [mirror 2]

SCO's copyrights

bojan — Thu, 21 Aug 2003 06:33:40 +0000

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

Color changing is easy.

frazier — Thu, 21 Aug 2003 05:43:21 +0000

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.

DNA Argument etc..

maguska — Thu, 21 Aug 2003 02:38:58 +0000

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

Getting through to LWN

corbet — Thu, 21 Aug 2003 00:36:03 +0000

You can change the color scheme any time you like; just go into the customization section of the "my account" screen.

Awaiting your subscription...:)

Getting through to LWN

Germ — Thu, 21 Aug 2003 00:04:31 +0000

I'll subscribe if they change this awfull color scheme...it hurts my eyes and offends my sense of asthetics...

Getting through to LWN

colink — Wed, 20 Aug 2003 16:43:48 +0000

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?

Not even patentable... 25 year limit just about over

egberts — Wed, 20 Aug 2003 14:53:47 +0000

This is not even patentable either.

So, just run it through 'ident' and recopyright it.

Why SCO won't show the code

luckybit — Wed, 20 Aug 2003 10:02:40 +0000

Standards the USA do not get under a copyright. The decision of court BSD vs. ATT

McBride's volley of folly

rjamestaylor — Wed, 20 Aug 2003 08:43:20 +0000

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.

The Lions book (more nitpicking)

rjamestaylor — Wed, 20 Aug 2003 08:19:36 +0000

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.

;)

Getting through to LWN

ekj — Wed, 20 Aug 2003 07:51:07 +0000

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.

Ancient Unix Mirror

mcbridematt — Wed, 20 Aug 2003 06:57:59 +0000

It turns out that PlanetMirror is still mirroring the ancient unix files

Why SCO won't show the code

vijayandra — Wed, 20 Aug 2003 03:42:13 +0000

The James Bond visuals and music indicates McBride's frustration.

Information on "SCO Ancient Unix" web page appears to be factually incorrect.

lakeland — Wed, 20 Aug 2003 03:20:02 +0000

If it is their code, they're allowed to dual licence it. They can quite happily give it away
under 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

Jotham — Wed, 20 Aug 2003 02:10:24 +0000

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

neoprene — Wed, 20 Aug 2003 00:33:30 +0000

http://www.heise.de/newsticker/data/jk-19.08.03-000/imh0.jpg

THE 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.

Make copies of that wayback machine page!!

Arker — Tue, 19 Aug 2003 23:58:32 +0000

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.

Why SCO won't show the code

Xman — Tue, 19 Aug 2003 23:50:21 +0000

Note that because this has the advertising clause, it's actually GPL incompatible.

The Lions book

nicku — Tue, 19 Aug 2003 23:47:42 +0000

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.

Clean BSD code

Arker — Tue, 19 Aug 2003 23:19:20 +0000

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?

Make copies of that wayback machine page!!

mmarq — Tue, 19 Aug 2003 22:50:22 +0000

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 ?

Why SCO won't show the code

djabsolut — Tue, 19 Aug 2003 22:36:33 +0000

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.

DNA Argument

nexex — Tue, 19 Aug 2003 22:28:00 +0000

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.

Too trivail to copyright (pseudo code)

mammothrept — Tue, 19 Aug 2003 22:17:16 +0000

"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.

Too trivial to copyright

rfunk — Tue, 19 Aug 2003 22:15:08 +0000

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.

Why SCO won't show the code

spshealy — Tue, 19 Aug 2003 22:09:58 +0000

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

renco — Tue, 19 Aug 2003 21:46:31 +0000

For the ancient licenses

The 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 !!

Too trivail to copyright (pseudo code)

darkonc — Tue, 19 Aug 2003 21:28:32 +0000

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.

Re: Editors, please note this.

lsweeks — Tue, 19 Aug 2003 21:14:45 +0000

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;)

Getting through to LWN

corbet — Tue, 19 Aug 2003 21:13:37 +0000

"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.

Do they think the BSD license is restrictive?

iabervon — Tue, 19 Aug 2003 21:09:05 +0000

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.

Clean BSD code

muon113 — Tue, 19 Aug 2003 20:59:35 +0000

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

The second, and third, shoes

ccchips — Tue, 19 Aug 2003 20:48:04 +0000

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.

Why SCO won't show the code

ak_hepcat — Tue, 19 Aug 2003 20:08:48 +0000

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..