homepage Welcome to WebmasterWorld Guest from 54.234.0.85
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Become a Pro Member
Home / Forums Index / Hardware and OS Related Technologies / Linux, Unix, and *nix like Operating Systems
Forum Library, Charter, Moderators: bakedjake

Linux, Unix, and *nix like Operating Systems Forum

    
New Allegations Against Linux
Will this have an impact on the SCO v IBM suit?
cyril kearney

10+ Year Member



 
Msg#: 1083 posted 7:18 pm on May 20, 2004 (gmt 0)

Here's the stories from eWeek and CNET

[eweek.com...]

and

[news.com.com...]

The Linux community needs to face this issue. It is not good enough to point to Microsoft and say that they have contributed to the same think tank that is making the allegation.

Linus Torvalds admits that he developed Linux on a computer running the Minix operating system. Torvalds claims he did not use the Minix source code as the basis of Linux.

Minix, a Unix look-alike was written by Andrew Tanenbaum. Tanenbaum was familiar with Unix's source code but Tanenbaum claims he did not use the Unix source code as a basis for Minix.

Even after accepting Torvalds and Tanenbaum's claims as true, a copyright issue may still remains. Are either or both of these systems derived from Unix?

I don't know what impact if any this new information will have on the SCO suit.

Just to make things more confused the publisher, Prentice Hall seems to own the copyright to Minix. Viacom is the parent of Prentice Hall.

A point to remember is that Torvalds only wrote a small portion of what is included in most Linux distributions. The GNU utilities really make the Linux kernel usable. When the GNU utilities are run with the BSD kernel the result is pretty much the same and with the Linux kernel.

 

bcolflesh

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 1083 posted 7:23 pm on May 20, 2004 (gmt 0)

Wow - way off - you may want to read the info on Torvald or Tanenbaum's sites, and the recent discussion on /.

All the claims about Linux mentioned in those articles are verifiably false.

bakedjake

WebmasterWorld Administrator bakedjake us a WebmasterWorld Top Contributor of All Time 10+ Year Member



 
Msg#: 1083 posted 7:51 pm on May 20, 2004 (gmt 0)

*yawn*

This story sounds like a think tank generating PR to sell a book.

bcolflesh

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 1083 posted 7:56 pm on May 20, 2004 (gmt 0)

sounds like a think tank generating PR to sell a book.

That would be SOP, but this seems more insidious/ridiculous:

[developers.slashdot.org...]

drbrain

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 1083 posted 8:13 pm on May 20, 2004 (gmt 0)

Please check out the UNIX family tree.

Linux is a descendent (not derivative) of Minix, which is a descendent of the UNIX Time Sharing System, Seventh Edition.

The disputed copyrights are for Unix SVRx (System 5 Revisions), which is itself descended from other descendents of the Unix Time Sharing System.

IIRC, the Unix Time Sharing System was placed in the public domain by AT&T.

Duckula

10+ Year Member



 
Msg#: 1083 posted 10:58 pm on May 20, 2004 (gmt 0)

Oh, WTH. I'm bored anyway.

The Linux community needs to face this issue.

No it doesn't. There's no need to face any issue by anybody except the people who puts their (C) at the file headings. Users have no real responsability over the "paternity" allegations.

Of course, if the "linux community" means really "the linux hackers and contributors community" then it's fine.

But then "face the issue" could mean "laugh a little for accusing us of skipping the work of initial kernel development without attempting to compare minix with Linux 0.x series".

Linus Torvalds admits that he developed Linux on a computer running the Minix operating system.

What a way to mislead. Attacking the man for the system he used. How low. While you're on it, do you want to blame him for the shirt he was wearing too?

Torvalds claims he did not use the Minix source code as the basis of Linux.
Tanenbaum claims he did not use the Unix source code as a basis for Minix.

And until there's some reason to believe the contrary I see no reason to not believe them, as they are both respected members of society and not known for false testimony.

Are either or both of these systems derived from Unix?

drbrain has the facts checked, even in the odd case of an actual derivation between linux/minix/unix that, as I'm asserting, has not been shown yet.

Just to make things more confused

Actually, no. There's nothing confuse over asserting a copyright over a work. If minix is copyrighted by somebody that's business as usual. So far there's no reason to believe Linux infringes minix's (C), given both systems show their codes and no proof has been presented, but even if that were the case it doesn't confuses things at all (unless one gets easily confused and/or is not well informed).

A point to remember is that Torvalds only wrote a small portion of what is included in most Linux distributions.

Another misleading attempt. If I remember well (and I do, since I've just reread the allegation) this not-really-quite-a-mess is over the kernel, not over userland. Userland has no place in a discussion over kernel's source origins.

The GNU utilities really make the Linux kernel usable

Given the last response it's pointless to answer to this. But, as I said, I'm bored anyway.

Not really. For a programmer, which is the one who really "talks" to the kernel, Linux is directly usable. It depends on the end user needs, but there's no reason (besides tradition and quality of the GNU tools) to not implement a full userland utilities kit pure from GNU influence.

Duckula

10+ Year Member



 
Msg#: 1083 posted 12:13 am on May 28, 2004 (gmt 0)

there's no reason to believe Linux infringes minix's (C), given both systems show their codes and no proof has been presented

Since I opened the door I feel responsible to post the non infringement proof.

Since, out of thousand of lines of code, only 4 small segments were found to be similar, and since in each case the similarity was required by external factors (the C standard, the POSIX standard, the minix filesystem format), it is highly unlikely that any source code was copied either from minix to linux or vice-versa.

Source comparison of early linux and minix versions [cs.vu.nl]

Isn't showing the code great?

bird

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 1083 posted 9:06 am on May 28, 2004 (gmt 0)

Torvalds claims he did not use the Minix source code as the basis of Linux.
...
Tanenbaum claims he did not use the Unix source code as a basis for Minix.

Even after accepting Torvalds and Tanenbaum's claims as true, a copyright issue may still remain

Those two statements in sequence simply make no sense at all.

Cyril, quite often you bring really interesting topics and provocative thoughts on the table here. But in this case you seem to have fallen for nothing but hot air.

SlowMove

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 1083 posted 12:36 pm on May 28, 2004 (gmt 0)

There's nothing there, but even if there was, open soucre is here to stay.

isitreal

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 1083 posted 6:33 pm on May 28, 2004 (gmt 0)

I'm surprised anyone fell for that nonsense, here's some fun reading:

1992 linus/tanenbaum exchange, just for some background [dina.dk]

linux insider interviews tanenbaum on that silly fake story [linuxinsider.com]

linuxworld interview with tanenbaum were he explicitly denies all the findings brown claims [linuxworld.com]

tanenbaum on this story, again in his own words [cs.vu.nl]

Brown is clearly a paid Microsoft employee whose job it is to generate bad press for Linux, whether or not he has any substance is totally irrelevant, the job is to get press.

cyril kearney

10+ Year Member



 
Msg#: 1083 posted 6:06 pm on May 30, 2004 (gmt 0)

bird,
I admit that I know more about literary copyrights than I do code copyrights. However the basic copyright law is derived from laws meant to protect written works.

If Shakespeare had a valid copyright now, one could not take the underling story of Romeo and Juliet, change the setting, the characters names and the method of suicide without having to worry about infringement.

The focus of most Linux commentators is just wrong. The work must also be taken as a whole not just snippet by snippet. If 10 lines of code perform a certain function, it may in fact be a copyright infringement to just rewrite the code to perform the exact function.

It may not be. If the function adds a to b and returns c, the code may be just too general to be a copyright infringement.

If there is many (and I don't know how many) that seem to have been rewritten in essentially the same order doing essentially the same thing, it is easier to sustain a copyright violation.

Brown may very well be a catís-paw or minion, I can't tell.

Here's the troubling part as I see it. Tanenbaum as I understand it taught an OS class or wanted to but certainly had access to some version of the Unix kernel. This access was legitimate and proper. He wanted a simpler OS to teach so he created Minix. He published the code (through Prentice Hall, I think) and distributed the source with the executable of Minix.

Torvalds wanted to produce a better kernel. He somehow had to understand what was in the kernel and how it needed to interface with other utilities and compilers. He might have taken a course or examined code to get this understanding. He might have intuited it from observing other OS's. Torvalds says that he ran both Unix and Minix but never derived anything from them.

This may be absolutely true but since others contributed code it is less certain that they did not derive their code or infringe on copyrights.

We do know that the GNU utilities work essentially the same for the Linux kernel as they do for kernels used by AIX, FreeBSD, Solaris, and System V. We do know that Linux is POSIX compliant. Given a clean slate how did this come about? Well others suggested and contributed code to make this happen. The SCO suit claims that IBM was the source of some infringing code.

SCO is claiming it has found vestiges of its copyrighted code to substantiate its claim. I think that it is something in the realm of possibility. The courts will eventually decide this. Even if some code infringes, the relief the court gives may not require a payment to SCO. It may require IBM to pay some money and may require some code to be removed from Linux. bird,
I admit that I know more about literary copyrights than I do code copyrights. However the basic copyright law is derived from laws meant to protect written works.

If Shakespeare had a valid copyright now, one could not take the underling story of Romeo and Juliet, change the setting, the characters names and the method of suicide without having to worry about infringement.

The focus of most Linux commentators is just wrong. The work must also be taken as a whole not just snippet by snippet. If 10 lines of code perform a certain function, it may in fact be a copyright infringement to just rewrite the code to perform the exact function.

It may not be. If the function adds a to b and returns c, the code may be just too general to be a copyright infringement.

If there is many (and I don't know how many) that seem to have been rewritten in essentially the same order doing essentially the same thing, it is easier to sustain a copyright violation.

Brown may very well be a catís-paw or minion, I can't tell.

Here's the troubling part as I see it. Tanenbaum as I understand it taught an OS class or wanted to but certainly had access to some version of the Unix kernel. This access was legitimate and proper. He wanted a simpler OS to teach so he created Minix. He published the code (through Prentice Hall, I think) and distributed the source with the executable of Minix.

Torvalds wanted to produce a better kernel. He somehow had to understand what was in the kernel and how it needed to interface with other utilities and compilers. He might have taken a course or examined code to get this understanding. He might have intuited it from observing other OS's. Torvalds says that he ran both Unix and Minix but never derived anything from them.

This may be absolutely true but since others contributed code it is less certain that they did not derive their code or infringe on copyrights.

We do know that the GNU utilities work essentially the same for the Linux kernel as they do for kernels used by AIX, FreeBSD, Solaris, and System V. We do know that Linux is POSIX compliant. Given a clean slate how did this come about? Well others suggested and contributed code to make this happen. The SCO suit claims that IBM was the source of some infringing code.

SCO is claiming it has found vestiges of its copyrighted code to substantiate its claim. I think that it is something in the realm of possibility. The courts will eventually decide this. Even if some code infringes, the relief the court gives may not require a payment to SCO. It may require IBM to pay some money and may require some code to be removed from Linux.

Torvalds might even have a possible claim against IBM for feeding Linux tainted code.

bird

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 1083 posted 7:05 pm on May 30, 2004 (gmt 0)

The focus of most Linux commentators is just wrong. The work must also be taken as a whole not just snippet by snippet. If 10 lines of code perform a certain function, it may in fact be a copyright infringement to just rewrite the code to perform the exact function.

The Convention of Berne, it's succeding treaties, and the resulting national legislation treat software quite a bit differently than literary works, and for good reasons.

Literary innovation is not hampered if you can't rewrite Romeo and Juliet in your own settings.

The purpose of a coded function, and the way that function is called and returns results, is called its API (Application Programming Interface). Software innovation *is* hampered, if you can't re-implement an existing API with your own code. Based on the relevant copyright regulations, it has been established many times in court (in several countries) that software APIs can't be copyrighted. The explicit purpose of that rule is that people can write software that interacts with existing software.

The specification of UNIX (parts of which are also known under the name Posix) is nothing more than a collection of API definitions. This means you can implement your own UNIX/Posix compatible operating system without violating any coprights. And not surprisingly, this is exactly what Linux tries to do.

This UNIX specifications (and the trademark UNIX) are maintained and owned by the Open Group. SCO only owns the code to one specific implementation of that specification. Therefore there is no way for SCO to rightfully claim copyright infringment other than by producing evidence that significant portions of their code have been copied.

Since Tannenbaum didn't copy any code from AT&T (which owned the relevant implementation of UNIX at that time), and Linux didn't copy any of Tannenbaums Minix code, the story that started this thread is really just what I called it a few posts ago: hot air.

py9jmas

10+ Year Member



 
Msg#: 1083 posted 8:03 pm on May 30, 2004 (gmt 0)

SCO is claiming it has found vestiges of its copyrighted code to substantiate its claim.

They've been claiming this for over a year now. In this time they've even published examples of infringing code. They have all been ridicled. One example (code from bpf, the Berkeley Packet Filter) SCO doesn't even own the copyright to the code they claimed was copied. And don't forget that SCO's ownership of the Unix copyright is disputed by Novell.

Duckula

10+ Year Member



 
Msg#: 1083 posted 10:24 pm on May 30, 2004 (gmt 0)

We do know that Linux is POSIX compliant. Given a clean slate how did this come about? Well others suggested and contributed code to make this happen. The SCO suit claims that IBM was the source of some infringing code.

Cyril should be very confident of that fact since he wrote it twice.

Ok, let's see...

The IBM contributions at my 2.6.6 tree 'grep -r IBM /usr/src/linux ¶ grep opyr' (I'm too lazy to try it at other versions) are mostly from 1999 and later with a few noticeable exceptions: an scsi string (that I haven't found in use in code) from 1994 and arch specific stuff in 1998; none of them related in any way to POSIX compliance.

Linux is mostly POSIX compliant, and running Gnu tools, since, uhm, way before 1999.

The allegation "IBM made Linux POSIX compliant" is frivolous at best.

Or maybe the paragraph was intentionally misconstructed to give the false impression of some relation between POSIX related contributions and IBM contributions.

Global Options:
 top home search open messages active posts  
 

Home / Forums Index / Hardware and OS Related Technologies / Linux, Unix, and *nix like Operating Systems
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Home ¦ Free Tools ¦ Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About ¦ Library ¦ Newsletter
WebmasterWorld is a Developer Shed Community owned by Jim Boykin.
© Webmaster World 1996-2014 all rights reserved