...and if SCO is right...? [LWN.net]

As a general rule, the reaction to SCO's lawsuit against IBM has been one of derision and disbelief. It is generally assumed that SCO does not have a legal leg to stand on. SCO's tactics (ever-expanding FUD while refusing to point out the allegedly infringing code) have certainly served to reinforce that perception. But it is worth taking a moment to consider what could happen if SCO turns out to be right. Forewarned, as they say, is forearmed.

The Linux kernel (which is the subject of at least some of SCO's claims) is, as a whole, clearly an independent development. The development history is sufficiently public to make that clear. But it is worth considering a few things:

The source to various proprietary Unix systems tends to be more widespread than many people think. Numerous companies have source licenses, and, despite careful procedures, copies can leak out.
There is considerable reputation value in making contributions to the Linux kernel. Perhaps more than any other free software project, the kernel is surrounded by developers who would like to get their names into the changelog, even if that means submitting spelling fixes.
Some people are lazy or unable to program at the level required for kernel development (or both). Some of those people may have access to some flavor or other of proprietary Unix. And some of them might just be sufficiently dishonest to present somebody else's code as their own.

It is also worth bearing in mind that there is no process for checking the pedigree of code submitted to the Linux kernel. Kernel developers (like other free software developers) have more than sufficient integrity to keep them from stealing code, and the process relies upon that fact. If a developer can convince Linus or another major kernel hacker that a patch makes sense, in it goes. Some kernel code is heavily reviewed, but there are vast amounts of code that may not have ever had a serious look by anybody other than its author.

Beyond all that, of course, is the unpleasant scenario of tainted code being deliberately submitted to the kernel with the express intent of creating legal problems.

The end result is that there might be code of dubious parentage in the kernel. Such code is probably small, and not in the kernel core. But the existence, say, of a purloined device driver somewhere in the kernel would not be entirely surprising. The kernel community might just wake up one morning to find that there are plagiarists in its midst.

What happens then? Obviously, a code purge would be called for. Unless SCO explicitly puts any offending code under the GPL (which it might have to do to preserve its own right to distribute the kernel), any infringing code must be pulled from the kernel. That code could be excised even if SCO does release it; its presence would certainly be galling to a number of people. A big "purge and rewrite" operation could, among other things, delay the release of the 2.6 kernel.

Future code contributions would receive a higher degree of scrutiny - this may well happen regardless of how the SCO suit turns out. Even if it has not yet happened here, free software projects are vulnerable to injections of tainted code. Developers may have to be prepared to explain how they came up with a particular patch. It is hard to imagine the kernel adopting a bureaucratic mechanism where develpers must sign code releases with warranties and indemnification agreements, but it could happen. Adding that kind of friction to the system can only serve to slow down development, of course.

Most frightening, perhaps, is what happens if the kernel development community discovers that one or more of its members has been polluting the well with unfree code. The resultant shattering of trust could impair that community's ability to work together for a long time. In the worst case, if important developers are implicated in dishonest activities, a major fork of kernel development is not out of the question.

A successful suit would also make waves in the business world, of course. In the worst case, companies could move away from free software out of fear of lawsuits; this scenario seems unlikely, however. But companies could hold back on code releases or contributions to free software projects out of fear of being accused of illegal copying. A general chilling effect which slows adoption of Linux is a real possibility.

Happily, the most likely outcome is that SCO and its lawsuit go down in flames. They have picked on, perhaps, the most transparently developed piece of code in history by way of a huge company with seriously scary lawyers, deep pockets, and the will to defend itself. But the worst-case scenario is worth keeping mind for this simple reason: even if the Linux community doesn't get burned this time, it could happen in the future. We need to pay a great deal of attention to where our code comes from.

to post comments

...and if SCO is right...?

Posted May 16, 2003 17:11 UTC (Fri) by JoeBuck (subscriber, #2330) [Link] (7 responses)

There's been a long-running argument between Linus and the FSF over the issue of copyright assignments. Contributors to key FSF projects, such as GCC, sign a binding legal contract with the FSF. The standard language (which most contributors have signed) contains the language

The Assigner hereby represents and warrants that it is the sole copyright holder for the Work and that it has the right and power to enter into this contract. The Assigner hereby indemnify and hold harmless the Foundation, its officers, employees, and agents against any and all claims, actions or damages (including attorney's reasonable fees) asserted by or paid to any party on account of a breach or alleged breach of the foregoing warranty.

Linus has rejected the concept of assignment and has argued rather vigorously against it. This means that the FSF is protected against dishonest folks in a way that Linus is not.

...and if SCO is right...?

Posted May 16, 2003 17:37 UTC (Fri) by josh_stern (guest, #4868) [Link] (2 responses)

I'm having trouble seeing Joe Buck's point here. Is there a worry about Linus
himself being sued? For what? Which act of negligence and what damages?
On the other hand, let's suppose that there was a copyright assignment to the
FSF in place, but that in spite of this, Evil Employee X of Corporation Y injected
tainted code into the kernel. If Evil X is wrongfully misappropriating IP of Corp
Y, then Corp Y doesn't lose copyright because of that - the code would still need
to be purged. If Corp Y was behind the actions of X then they have committed
fraud and would certainly lose their copyright, with or without the extra copyright
agreement. What SCO alleges, if true, puts the kernel team in a position
analogous to someone who unknowingly received stolen property. Corbet is
right to point out that this could cause more future caution about where stuff
comes from in the future, but people who steal other's stuff might also falsely
sign release forms, and the overall situation would not be significantly altered in
that case.

...and if SCO is right...?

Posted May 17, 2003 16:35 UTC (Sat) by donwaugaman (subscriber, #4214) [Link] (1 responses)

> On the other hand, let's suppose that there was a copyright assignment
> to the FSF in place, but that in spite of this, Evil Employee X of
> Corporation Y injected tainted code into the kernel.

The FSF covers this by making a representative of Corporation Y sign the copyright assignment in addition to or in lieu of Evil Employee X. If Evil Employee X signs it without Corporation Y's approval, Evil Employee X is on the hook for damages, since:

"The Assigner hereby represents and warrants that it is the sole copyright holder for the Work and that it has the right and power to enter into this contract."

The warrant means that Y cannot go after the FSF if X does the wrong thing; but they can sue X for misappropriation.

...and if SCO is right...?

Posted May 17, 2003 19:32 UTC (Sat) by josh_stern (guest, #4868) [Link]

>The FSF covers this by making a representative of Corporation Y sign the
>copyright assignment in addition to or in lieu of Evil Employee X. If Evil
>Employee X signs it without Corporation Y's approval,

But our thread topic here is about people knowingly misappropriating/stealing
stuff, and what might have to be done to remedy that. In that context, 'extra
protection' based on the idea that the person who would do such a thing would
not lie about who their employer is, fake a signature, etc. seems minimal to me.
Their plan is clearly to get away with something based on the idea that nobody
will every bother with the comparison necessary to discover their wrongdoing.

>The warrant means that Y cannot go after the FSF if X does the wrong thing;
>but they can sue X for misappropriation.

Almost every software/technology corporation has employment agreements
that would allow them to do this anyway. The one's that don't probably don't
consider their software to be proprietary in that way.

My argument isn't that the FSF procedures are wrongheaded, but rather they
are not all that germane to the important issues. The big core risk to free
software here is not lawsuits of FSF, Linus, etc. but rather possible fear on the
part of business users of free software who might worry that open source
makes fishing expeditions for IP violation easier and increases the possibility
that their business, that relies on said said software, could be interrupted
sometime in the future by an IP-based cease and desist order. It seems to me
that their fear could only be assuaged by either a) believing such an event is just
too improbable to worry about, or b) getting some legal protection in the form of
court precedent that any cease and desist would have to take due account of
their innocence and collateral damage (e.g. they would automatically get some
reasonably long grace period to comfortably make a change in their installation).
The 'protection' that they would have a stronger lawsuit against the, now
unemployed, EvilX just isn't all that relevant to them.

...and if SCO is right...?

Posted May 18, 2003 15:03 UTC (Sun) by jamesh (guest, #1159) [Link]

Why is a copyright assignment needed for this?

You can as easily make a contributor sign a similar agreement without assigning copyright. The Mozilla project makes people with CVS access sign such a form (it also requires that they provide attribution for other people's patches that they commit, to help with accountability).

...and if SCO is right...?

Posted May 22, 2003 4:44 UTC (Thu) by mmarkov (guest, #4978) [Link] (2 responses)

I am surprised this measure is necessary.
Isn't the innocence of the consumer (I mean,
the one that receives) pre-supposed? For
instance, when you buy something, you don't
need a special certificate that it is not
stolen/smuggled/etc. If it turns out afterwards
that the commodity is somehow illegal, you
are clean, unless it is proven that you bought
it, knowing about the illegality.

...and if SCO is right...?

Posted May 22, 2003 8:27 UTC (Thu) by ekj (guest, #1524) [Link] (1 responses)

Only clean in the sence that you will not be punished for the theft. You still have to give the stolent good back to the real owner, without refund of your costs, unless you can find the person who sold it to you and demand a refund from him/her.

...and if SCO is right...?

Posted May 23, 2003 16:46 UTC (Fri) by brouhaha (guest, #1698) [Link]

But having a signed transfer of ownership document doesn't avoid that. If it's found to be stolen property, the recipient will still lose it, and his recourse is still only with the party that provided it.

How will the court decide?

Posted May 16, 2003 17:28 UTC (Fri) by tjc (guest, #137) [Link] (7 responses)

The thing that concerns me most is that (as far as I know, which isn't very far in legal matters) there is no legal precedent for judging source code. Will the court have enough understanding to discern the difference between programming interfaces and the actual implementation of such an interface? Will they recongnize that there are certain idioms for coding that are commonly used and are not part of anyone's IP? Will they rely on the advice of "technical experts" (who may have an agenda of their own) in making a decision?

Beyond that, how will they determine if code has been deliberately obfuscated? As a simple example, it's easy to see that this:

if (foo) a = 1; else a = 2;

is about the same as this:

z = bar ? 1 : 2;

...but there is no way to prove that this code has been deliberately changed to mask it's origins. Code translation is not a lossless process in that one's intentions can not be determined by looking at a chunk of code. Speculation is as far as this can go.

There are a lot of things that will be decided by people who may not be qualified to make a reasonable decision.

How will the court decide?

Posted May 16, 2003 17:55 UTC (Fri) by dd9jn (✭ supporter ✭, #4459) [Link] (5 responses)

In the view of the copyright law, your examples are different. The law does not take semantics into account but only the actual text. The GNU coding standards make this pretty clear to every GNU developer. This is also the reason why we need this long legal blurb in each and every non-trivial file.

How will the court decide?

Posted May 16, 2003 21:15 UTC (Fri) by tjc (guest, #137) [Link] (3 responses)

The law does not take semantics into account but only the actual text.

That sort of takes the teeth out of the license, don't you think? If someone can come along and just reformat the code, change variable names, and restructure things a bit, then there isn't much practical difference between the GPL and the X11 license.

As an example, suppose I start with a window manager that's licensed under the GPL and rewrite every line of code as described above and release it under an X11 license: have I violated the GPL? I used GPLed code during the development phase of my project, but at the end, none of it remains.

Now supposed someone did this with SCO code in the Linux kernel. It can't be proven that this was done unless a chunk of code was copied in unchanged. But SCO seems to be saying that code has been copied into the Linux kernel and "obfuscated", and they expect a court to make some determination as to whether this was done or not. If a programmer can't tell for certain, what's the judge going to do?

How will the court decide?

Posted May 17, 2003 9:51 UTC (Sat) by arcticwolf (guest, #8341) [Link] (2 responses)

If a programmer can't tell for certain, what's the judge going to do?

Easy: innocent until proven otherwise.

How will the court decide?

Posted May 17, 2003 20:48 UTC (Sat) by pflugstad (subscriber, #224) [Link] (1 responses)

> Easy: innocent until proven otherwise.

I believe that's a criminal assertion. I don't think it applies to civil courts, does it?

How will the court decide?

Posted May 19, 2003 1:49 UTC (Mon) by Baylink (guest, #755) [Link]

The standard in civil litigation is "the preponderance of the evidence".

That might be a *much* harder standard to meet in this case, requiring
*someone* to spend lots of money on lawyers to provide said evidence.

Oh, that's IBM. Never mind...

How will the court decide?

Posted May 23, 2003 16:49 UTC (Fri) by brouhaha (guest, #1698) [Link]

That's not really true. Copyright law explicitly does cover translations of texts into other languages. That can *only* be evaluated in terms of semantics.

Also, I really doubt that you can take a Steven King novel, change the names of the characters, reword each sentence slightly, and have the result found by a court to not infringe the copyright.

Copyright protects the expression of an idea, not the idea, but it is up to a court to decide whether a different work is just a variant of the same expression, or a new and different expression of the same idea.

How will the court decide?

Posted May 18, 2003 15:06 UTC (Sun) by jamesh (guest, #1159) [Link]

I would expect that both sides of the case will provide expert witnesses to help persuade the court. Judges are not expected to know about everything personally.

...and if SCO is right...?

Posted May 16, 2003 20:02 UTC (Fri) by dkite (guest, #4577) [Link] (4 responses)

I'll make a wild prediction:

This case will not be settled in court in my lifetime.

I'd make a bet, but there would be no way for me to collect.

Did the previous unix legal battle ever get settled in court? I don't think so. The
longer the procedure dragged on, the more muddled the whole thing got, and the
more the plaintiffs were about to lose. This one will be the same. I don't think this
is about cut and paste code, rather trademark and trade secret stuff, which is
rather ephemeral at the best of times. There are literally thousands of individual
companies, consultants and developers that have a stake in this thing, each with
an intelligent case to make. Every claim that SCO will make can be and will be
countered by some argument with value.

How much do the lawyers get? 1/3 of a billion? How long will that last? How long
can they last without cash flow? Isn't that what the latest volley is about,
threatening users, some of which will call back and say "how much, who to?".

Most free/open projects respond to IP problems by quickly pruning anything even
remotely controversial.

Derek

...and if SCO is right...?

Posted May 17, 2003 4:41 UTC (Sat) by BrucePerens (guest, #2510) [Link]

Derek,

Yes, the previous Unix battle was settled. ATT paid the University of California's court costs so that the University would not counter sue them. The important point is that Mighty ATT lost the lawsuit to the BSD developers.

Here's my bet. This will be over much faster than you expect.

Bruce

...and if SCO is right...?

Posted May 19, 2003 17:16 UTC (Mon) by llywrch (guest, #9903) [Link] (2 responses)

> This case will not be settled in court in my lifetime.

I would disagree with your assertion were it not for the words ``in court." Because after having thought about this possibility all weekend, if the SCO group were actually right, IBM would respond in the same manner it is currently: grind the oponent down with legal motions & maneuvers until they run out of money. And since it's been a few years since either SCO or Caldera made a profit, that's only a matter of time.

They could sell the rights to the case to someone else with deeper pockets, but the question then turns on just how much money would the suit yield, & would it justify the initial investment. And I'm not convinced that Microsoft (for example) would be willing to spend tens of millions of dollars -- or more -- just to cast doubts on status of the intellectual property surrounding Linux.

The chief challenge for IBM is whether the SCO Group can actually pull their license to the UNIX code in June. There is a weakness in the SCO Group's claim that just occured to me, one no one else seems to have mentioned:

The original license most likely had a clause that stated it would be enforced by the laws of the state AT&T decided -- probably either Delaware (a favorite of many large US corporations) -- or AT&T home state, which could be New Jersey. But in eithr case, the state of Utah does not have jurisdiction to let the SCO Group revoke the license. That should be a slam-dunk for IBM.

The SCO Group will then try to refile in the proper jurisdiction, & maybe then be able to revoke IBM's license. They may try to communicate to IBM that they will remain a thorn in Big Blue's side until they settle. However, IBM has been demonstrating the wisdom that if one doesn't fight nuisence suits like these as they arise, then more will spring up.

So in short, the case will probably end in our lifetime, when the SCO Group gone out of business & its management become pariahs. And we should know if this is so by the middle of next month.

Geoff

...and if SCO is right...?

Posted May 20, 2003 12:56 UTC (Tue) by Cato (guest, #7643) [Link] (1 responses)

Unix has changed ownership a few times since AT&T owned it - first Novell, then the original SCO, and now the new SCO (originally called Caldera). I would guess that all Unix licenses have been updated to account for each change of ownership, so the jurisdiction issue is unlikely to help. Of course, I'm not a lawyer...

...and if SCO is right...?

Posted Jun 7, 2003 1:12 UTC (Sat) by mcbridematt (subscriber, #10302) [Link]

And what if Unix code is in NetWare? Novell might be saying something, but in a careful manner so SCO doesn't think Unix code is in NetWare.

...and if SCO is right...?

Posted May 16, 2003 20:50 UTC (Fri) by StevenCole (guest, #3068) [Link]

There is considerable reputation value in making contributions to the Linux kernel. Perhaps more than any other free software project, the kernel is surrounded by developers who would like to get their names into the changelog, even if that means submitting spelling fixes.
Oh bugger! Ken Thompson knew how to spell "address", but quite a few kernel hackers didn't. I hope my fixes don't cause the sky to fall. On the other hand, it may be that "printer on fire" thingy they're so ticked off about.
Seriously though, it might be worthwhile to revisit The 1992 Unix System Labs vs Berkeley Software Design, Inc. Lawsuit. It looks like out of 18,000 files, only three had to be redacted. And find . -type f | wc -l only gives 14085 for 2.5.69.

...and if SCO is right...?

Posted May 16, 2003 21:12 UTC (Fri) by ctg (guest, #3459) [Link] (2 responses)

I've never seen any organisation with a "process" to ensure that code really belongs to
the author.

Code is created by individuals. It would be almost impossible not to use the lessions,
techniques and idioms learned in one job/project, and not use them in another, for
example.

I'm pretty sure that if you looked at the closed sourced projects that I've been involved
with (i.e proprietry systems), you would find code produced by individuals that was
somewhat similar to the sort of code that they had produced for other projects...
belonging to another company with a different copyright.

Because the Linux development process is so open - that it is almost the best process
that I've come across for ensuring that blatant code duplication does not exist - it
would be too easy to it do be discovered. In a closed system there would be much less
risk of discovery.

Of course, the big risk for Linux is because the code is open it is easy for someone to
find "patterns" or areas of commonality...

I'm pretty sure that what SCO has found, if anything, is pretty much likely to be the
result of people simply applying the lessons, techniques etc. that they have acquired
during the course of their careers to problems in the Linux kernel.

...and if SCO is right...?

Posted May 19, 2003 10:22 UTC (Mon) by mdekkers (guest, #85) [Link]

IBM (of all companies...) has a rather tight set of processes in place to guard against "IPR leakeage" from its Open Source developers. The process ensures that code wrtitten and submitted by IBM either belongs to IBM and is cleared for donation, or does not belong to anyone else.

...and if SCO is right...?

Posted May 22, 2003 7:44 UTC (Thu) by Liefting (guest, #8466) [Link]

Code is created by individuals. It would be almost impossible not to use the lessions, techniques and idioms learned in one job/project, and not use them in another, for example.

Code is protected by copyright law. Lessons, techniques and idioms are protected by patent law. That's a totally different ballgame altogether.

...and if SCO is right...?

Posted May 16, 2003 22:37 UTC (Fri) by kunitz (subscriber, #3965) [Link] (2 responses)

SCO is not right. Even if there were source code with SCO copyright
in the kernel, it would be very easy for SCO to resolve the issue by
exactly declaring which source code is infringing their intellectual
property. This source code would be removed from the Linux kernel
in short time.

An even more painless solution would be that SCO publishs the
UNIX source under an open source license, allowing everybody to
build upon the UNIX legacy.

SCO, particularly SCOSource, has the clear mission to extract as
much money as possible from developers, distributors and users of
UNIX-like operating systems, so a fast, cheap, and painless
resolution of the issue is not in their interest.

...and if SCO is right...?

Posted May 17, 2003 23:00 UTC (Sat) by donstuart (guest, #4550) [Link] (1 responses)

"SCO ... has the clear mission to extract ... money"

Well, yes. That's what companies do. If the kernel contains SCO's code, SCO is under no obligation to find a solution that is cheap or painless for the rest of us. It might be nice for them to open source everything but there is no legal or moral obligation for them to do so. To the contrary, they have a legal obligation to extract as much money as they legally can for their stockholders.

The catch is the part about "legally." While we don't know for sure, it seems improbable that they have a winnable claim or even one good enough to coax a settlement out of IBM. If they don't, then this stuff is at best wasting stockholders money and at worst criminal.

That doesn't mean I don't hope IBM buries them

Don

...and if SCO is right...?

Posted May 22, 2003 3:01 UTC (Thu) by torsten (guest, #4137) [Link]

"SCO ... has the clear mission to extract ... money"

Well, yes. That's what companies do.

I disagree. Companies build products and provide services, and collect money for these. The Canopy Group, which purchased SCO, by way of the Caldera DOS example, is using their purchases to extract money.

Their $400 Million USD settlement from Microsoft was clearly an extraction, not a purchase or exchange.

There exist societal leeches that survive solely by the extraction of funds from other companies. They are usually populated by lawyers who have gained control of a patent portfolio. I don't think these bloodsuckers deserve the title "company."

...and if SCO is right...?

Posted May 16, 2003 23:24 UTC (Fri) by flux (guest, #7502) [Link] (3 responses)

Assuming the worst case scenario, how long before SCO goes after the *BSD derivatives in order to generate more revenue?

...and if SCO is right...?

Posted May 18, 2003 10:30 UTC (Sun) by MathFox (guest, #6104) [Link] (2 responses)

Well, ATT did go after BSD; and settled the case when it became clear that they would lose.
When SCO says that "UNIX is valuable intellectual property" I'ld like to point out that large parts of UNIX were written at universities (by students) and contributed back to the ATT source pool. With a small team of programmers you can create a small self-hosting UNIX system from scratch within a few months. (It took Linus 3 months to build 0.01)
The rise of Linux has lowered the value of the SCO UNIX licences; why pay tens or hundreds of dollars per processor to SCO if Linux is good enough (or better). Why pay thousands of dollars to see the SCO source if the Linux source is one download away?

...and if SCO is right...?

Posted May 18, 2003 14:17 UTC (Sun) by flux (guest, #7502) [Link] (1 responses)

I agree with the point that one of the advantages of Linux is that the source code is easily available, this leads to a collaborative and open development. However, some of the implied questions from my original post are:
(1)If SCO wins and then succeeds in demanding royalties from all distributions of the GNU/Linux OS, based on the inclusion of a small amount of code which is simialar to code that is also present in thier own OS, what measures have the other OS's taken to review all submitted code for Trade Secret and or Patent violation i.e. the *BSD code bases have changed since the original ATT court case and all code included since must therefore also be liable to a similar legal attack?
(2) On a wider issue, as already pointed out in an earlier post, this could persuade some business people that releasing source code in to the open will increase the risk of an IP attack.
The point often made by closed source proponents is that the it's not worth a business expense to fund open source solutions. Just pay the software vendor for a liscence to run propriety software. If this is enforced in law my argument would be that it is depriving people of the right of choice. If the software used by a business is ancilary to the main function, e.g. a hotel chain, then they should have the right to employ a consultant who creates a solution with open source software. A significant problem it seems to me is when trivial patents are awarded such that no software can be created with out falling foul of some patent.It is my opinion that if software patents are going to be awarded they should be for non-trivial innovations i.e. it would not be obvious to a tallented programmer.
(3) (semi-seriously) In order to level the playing field between open source and closed solutions, perhaps if a company is not willing to release the source code for a propriety program then there ought to be an independant review body which will review the code from both closed and open source enviroments. Perhaps the BSA could fund this such that if code is blieved to have been taken without permission from open source programs then the author is properly recompenced. You never know closed source authors/companies may also ask for each others code to be reviewed as well for patent/trade secret violations:-)

Hope the above rambling makes some sense, hopefully the SCO problem and other situations in future will be resolved such that collaboration on open source software is still viable and vibrant.

...and if SCO is right...?

Posted May 18, 2003 19:40 UTC (Sun) by MathFox (guest, #6104) [Link]

I must tell you that I live in Europe and that intellectual property law here is differing significantly from that in the USA. We could save Linux for most of the world by inviting Linus and RMS. (political asylum for open source software developers)

You can not review code on violation of Trade Secrets (hey, it's a secret) and it is better not to try that because you need to knowingly infringe to be punishable. IBM can be caught for it, because they have Non Disclosure Agreements in their source licenses from SCO; but that doesn't extend to you (unless you have downloaded one of the "free" SCO UNIX distributions).

For copyright infringement the playing field seems skewed such that the closed source companies stand a better chance of proving infringement in open source programs than the other way around. Despite that, there are more closed source companies caught red-handed than open source projects. I think that is due to the fact that closed source is (usually) unavailable to open source programmers; while copying open source into a closed source project is easy to do.

Software patents are a big pain (blame the USPO); I am glad they are not (yet) valid in Europe.

I have the feeling that for the Linux community this case will end the same way as the USL vs. BSDI case ended. SCO's allegations are less supported by evidence than USL's allegations were and this "billion dolar case" is bound to collapse. It could be that Linus is asked to remove a few dozen lines of code... The possibility that SCO is asked to open source its "very valuable" UNIX sources because of GPL inclusion isn't zero either!

I pay for this....

Posted May 17, 2003 17:37 UTC (Sat) by johnjones (guest, #5462) [Link] (2 responses)

for gods sake its bloody obvious and I pay for this ?

please dont write anything like this and ask money it's silly and does nothing for your
reputation

regards

John Jones

p.s. SCO had not even had its legal people read and understand the GPL.....

I pay for this....

Posted May 18, 2003 12:46 UTC (Sun) by amikins (guest, #451) [Link] (1 responses)

The particulars aren't 'bloody obvious' to everyone, and as I recall, the particular value LWN has to offer are the editorials.

If you don't like them, don't subscribe; a week isn't long to wait for the weekly edition, and 'straight up news' can be gotten elsewhere.

do you read ?

Posted May 18, 2003 13:34 UTC (Sun) by johnjones (guest, #5462) [Link]

the particulars are obvious if you have read the GPL and read about what SCO have said....
even without reading around the actions of SCO speak for themselves

I like LWN editorials what I dont like is this 'article' thats why I subscribe so I can encourage
and discourage when I see fit

what i object to is giving money to things I dont like... does that seem wrong to you ?

regards

John Jones

...and if SCO is right...?

Posted May 19, 2003 7:48 UTC (Mon) by ekj (guest, #1524) [Link] (1 responses)

This article is silly, naive and overlooks several rather obvious points. For starters:

The source to various proprietary Unix systems tends to be more widespread than many people think. Numerous companies have source licenses, and, despite careful procedures, copies can leak out.

Sure. No question about it. I've got source for several historical unixes myself. But this fact would tend to increase the chanse that code lifted from other unixes was discovered. Afterall, this means there's a lot of people reading linux-kernel that has atleast some knowledge of the source of other unixes. I'm not saying we'd nessecarily catch anything derived from one of them, but lifts of major subsystems would very likely be spotted.

Some people are lazy or unable to program at the level required for kernel development (or both). Some of those people may have access to some flavor or other of proprietary Unix. And some of them might just be sufficiently dishonest to present somebody else's code as their own.

This assumes that porting the code to fit in with Linux requires significantly less skill and understanding than writing the code in the first place. That is typically not so. I can see cases where reading code from other unixen would help you understand how to solve a certain problem, but only if you already understood the problem and the Linux-framework well. A pure 1-to-1 (obfuscated or not) copy of code from any other unix would simply not work in Linux, and furthermore it would be rather obvious it was written for something other than Linux.

Obviously, a code purge would be called for. Unless SCO explicitly puts any offending code under the GPL (which it might have to do to preserve its own right to distribute the kernel), any infringing code must be pulled from the kernel.

SCO ALREADY willingly distributed the Linux kernel under the GPL. I know it, because I downloaded my kernel from ftp.caldera.com myself. I did so the 15th of may, long after this suit was announced. It seems that at the moment they have removed the SRPM for the kernel, yet the RPM remains, meaning they're in GPL-violation. They cannot legally distribute the kernel without also offering source. Anybody with a line of code in the kernel can sue them at will.

Given that SCO themselves gave me my kernel, and did so under the GPL, I don't see how they can come now and attempt to reverse their stance. The license is not revokable. They gave me (nd millions of other people) the permission to distribute this under the GPL. We will.

Furthermore, it is well-worth pointing out that SCO themselves do not have a unix with the advanced "enterprice class" features they accuse us of stealing. It's rather curious that we could steal it fro mthem if they haven't got it. ("it" here being your choise of advanced feature. They're lacking NUMA, robust SMP, hotswap, usb2.0, firewire ....)

...and if SCO is right...?

Posted May 22, 2003 13:41 UTC (Thu) by jarto (guest, #3268) [Link]

Actually, SCO doesn't have to offer the kernel source on it's web server. They only have to give you the source if you ask for it. They violate the GPL only if they refuse to.

...and if SCO is right...?

Posted May 19, 2003 18:41 UTC (Mon) by southey (guest, #9466) [Link]

It is very clear that SCO is using this law suit a means to attack Linux. SCO has to show that the alleged code is SCO Unix (not BSD Unix) and came via IBM and not any other source including SCO. Since only SCO and license holder's can see the code I very much doubt any court will convict a third party that has no knowledge of any wrong doing or any way to verify it. So wining the suit means IBM has to pay up and Linux will only be affected if it included in the settlement such as removing that code. Although that would provide a counter law suit because of the GPL. Will the courts allow SCO to put the cat back in the bag after so long?

The Opposite Is True

Posted May 22, 2003 1:53 UTC (Thu) by mbcook (subscriber, #5517) [Link] (2 responses)

Let's just skay that this goes to court and the source code gets compaired. Maybe they're right. Maybe some code in the kernel was lifted. But how much code has been lifted out of the kernel (or other OSS projects) and put into SCO's codebase? I bet it's happened, and if it has it might be orders of magnitude more. What's to keep them from looking at the Linux driver for some network card and just copying it (and modifying it to work with their kernel)? I bet something like that has happened. It seems just as likely to me that if they do go to court, they'll be bitten by the reverse. The two seem just as likely to me.

The Opposite Is True

Posted May 22, 2003 18:57 UTC (Thu) by Baylink (guest, #755) [Link] (1 responses)

And, in practical fact, this has happened.

There is at least one case I remember where a hardware manufacturer
employee said (albeit off the record) that a driver wasn't being open
sourced because it either a) had stolen code in it, or b) had code in it
that violated someone else's patent. Video card, I think.

And there have certainly been a couple of cases where it's been
questionable as to whether a closed-source vendor had simply lifted GPL'd
code, and couldn't be caught at it since their code was closed.

So, certainly, they could have already been doing it to us...

But how do you prove it?

The Opposite Is True

Posted May 22, 2003 20:54 UTC (Thu) by piman (guest, #8957) [Link]

GPL clause 2c, while under strong contention as to its "freeness", is how lifted code is often found when proprietary software incorporates GPLd software. Running strings on the binary ends up pulling out the original author's copyright notices.

I imagine lots of proprietary software has similar strings (RCS id strings, for example).

What is the law on stealth copyright?

Posted May 22, 2003 18:55 UTC (Thu) by iabervon (subscriber, #722) [Link]

The situation I'm referring to is this: someone writes something, but never publishes it. The author shows it to someone, who copies it, sells it to a publisher, and then disappears. The publisher publishes it for a while, and then the author discovers it, and sues the publisher.

Obviously, the publisher isn't at fault; the content wasn't published at all and was entirely unavailable to the publisher to compare against, even if the publisher were able to compare each "new" work against everything that anyone had ever written before. The plagerist has disappeared long ago. The publisher has done some work and made some money, and the author hasn't made any money. The author may or may not want the content to be published at all, regardless of any royalties.

I suspect this isn't a situation which has happened much before; most authors actually intend their work to be published, and make it available for publishers to do due diligence against even before it is (with their copyright asserted, obviously). But it becomes significant with software, where the programmers write something, and then publish a work derived from it, potentially keeping the source secret.

If an author claimed that a publisher was publishing content owned by the author, but refused to reveal which book it was specifically because this would allow the publisher to stop publishing it, I doubt it would get very far. I'd suspect that it would be grounds for a countersuit, in fact.

Of course, SCO is still (which respect to the courts) at the stage of suing the accused plagerist. On the other hand, if they go after Red Hat next, they stand to lose badly; even if they don't they're in somewhat risky territory now, since they're refusing to provide information which Red Hat might need in order to follow the law. If Red Hat is forced to abandon or delay a project later due to not being informed of SCO's copyrighted work now, it seems like Red Hat would have a case for damages against SCO. And with the large number of companies now considering Linux, that could add up to a lot of money for SCO (or anyone who ends up with SCO's liabilities).

the missed point

Posted May 23, 2003 21:11 UTC (Fri) by mepr (guest, #4819) [Link]

The point that I have heard no one address is in regards to that "sco doesn't even have feature X" thing. If you read their claims a little more closely, you'll find that the likely place to look for infringements is in project monterey. Apparently IBM and SCO have shared claims on the IP developed in that project. It's entirely possible that code from there made its way into Linux. It is, of course, still the case that SCO most likely has distributed all of the code in question under a GPL licence, but there's been too much smoke blown all over the place about the old System V code. That is probably a red herring (where a red herring is a smelly fish rubbed into the ground to throw hunting dogs off the trail of their game). It seems most likely that SCO has been saying all fo these preposterous things precisely because they predicted that so many (including, sadly, Bruce Perens, et al) would go crazy picking out the gnats and swallow the smelly camel whole.

...and if SCO is right...?

Posted Jun 11, 2003 12:53 UTC (Wed) by cpm (guest, #3554) [Link]

SCO is not right.

Their argument is that Linux now has capabilities it did not
have prior to the alledged infection of the kernel with Unix
code by IBM.

The capabilities that SCO claims Linux did not have, Linux
in fact DID have years prior to IBMs contribution to the
kernel project.

The inverse is probably more true. SCO probably stole code.

SCOs posistion paper is so full of misrepresentations, false
statements and outright lies, that I really hope that IBM will
move for summary judgement.

ESR has done a wonderful paper on this whole thing.

You can read it at:

http://www.opensource.org/sco-vs-ibm.html

SCO is *NOT* right. It is not possible for them to be right.
It is much more likely that SCO included GPLed code into
it's product before IBM even came on board, but that is
pure conjecture, and isn't really germane.

SCO IS LYING.