Slashdot Mirror
Do You Know UNIX Secrets?
ESR writes "You can help stop the SCO attack on IBM and the Linux community.
I'm looking for ways to prove that Unix trade secrets have been legally
nullified.
I want to know if you have ever had read access to proprietary Unix
source code (not just binaries and documentation) under circumstances
where either no non-disclosure agreement was required or whatever
non-disclosure agreement you had was not enforced. To help out, see my No Secrets page."
"I know nothing... nothing!"
Actually, that was Schultz, not Klink.
That won't do because he does not want "Version 7 and older". http://catb.org/~esr/nosecrets/
Sun licensed the Solaris 8 source code under its Community Source License, a few years back. It is no longer the case though;
http://wwws.sun.com/software/solaris/source/
There was an NDA though, one which I can not recollect the therms of. But anyway, Solaris 2.x is based on SVR4, so it qualifies as a derivative.
Wealth is the product of man's capacity to think. -Ayn Rand
Eww. Isn't that incest?
This is why he's helping IBM.
Back in the mid '80's, when I was working for U S WEST, the Amdahl machine I and a bunch of other engineers had logins on had the /src files all open (read only).
I used my access to the source of that version of UNIX (UTS) source a lot to help me with the Xenix system I was running at home.
Thing is, my racall of this is flakey enough that I cannot provide actual dates that the source on the "PN1" machine was open (about a one year window, after we moved from an IBM to an Amdahl mainframe, probably around 1985-86).
I don't think that's good enough, though, to have any effect on the SCO v IBM case.
(I wonder if the fact that U S WEST used to be a part of the Bell System - I went to U S WEST from Bell Labs, Holmdel - possibly made us feel a "part of the UNIX family" so we didn't seem to be as strict about holding the source inviolate. I dunno.)
The University of Southern California had a project in 1981-1982 to port UNIX from a VAX 780 to the Data General MV8000 (from "The Soul of a New Machine") using about 20 grad students. To my knowledge, none of the students (including me) had to sign anything to work on the project, and we certainly had access to the full source. One of the other guys was Fred Cohen, who has been widely credited with coining the term " computer virus".
Dipshit, that was Schultz..
20 secs
This the page you meant?
Doesnt this kinda invalidate the whole law suit?
And Solaris isn't System V Release 4?! Gee...I guess Sun have been lying to me :-p
SunOS Release 5.7 [UNIX(R) System V Release 4.0]
The first line line of text that appears when the kernel loads. So you're right.
I was a in the platinum beta test program for Solaris at the time, and the question of it being made open source came up when I was visiting. Sun was actually interested in making all of the Solaris source available to NDA-signing clients, but found themselves unable to do so for legal reasons.
Over the years, over 100 subcontractors, some of which no longer exist as companies, were involved in writing the code that makes up Solaris. It was impossible for Sun to get the okay from all those subcontractors to make those pieces of the source available to clients outside of Sun, and I suspect it quickly became a logistical nightmare of tracking which pieces of code were subject to which legal agreements with subcontractors, who had the IP rights of any subcontractors that were defunct, and so on.
If you think that ESR is "finally" getting things in gear, then you haven't seen his OSI Position Paper. The initial draft of this was released 4 days after the initial lawsuit.
Those of you who have read it awhile back may want to look again. A fair amount has changed in the past week or so.
the no
(Posted anonymously for obvious reasons.)
I'm sitting at a workstation here in my CS department computer lab reading parts of the Solaris source (a recent version). I didn't sign anything to be able to do this, I just happened to find it in a remote, out-of-the-way part of the filesystem. The permissions are such that anyone with an account can read it, although I doubt they were intented to be that way. I'm not sure if this includes the code for every piece of Solaris; I see just about all of the user-space commands and libraries, and some development tools, but I can't find the kernel source, and the directory structure is too confusing to look for it.
This leaves SCO with a supposed copyright infringement claim. For this to be valid, someone must have taken SCO's code, and directly copied it into Linux. But again, SCO has not divulged where any infringement of this sort actually is. Their only defense for their silence falls on their "trade secret" argument, which as I pointed out, is voided by the fact that loss of these so called trade secrets is imminent anyways. So SCO would have nothing to lose by telling us where the copyrighted code is so it can be removed. Further, SCO has made references to code that was supposedly copied from SCO but changed so as to appear to be different, but if there were enough changes to it to cause it to appear that different, then it would not be a copyright violation. Copyright does not protect ideas, only the content. Further, copyright does not extend to ownership of derivative works. [1]
It is also worth pointing out that even if the so-called infringing code were removed, it would not reduce the damages that SCO could collect if their allegations have merit. That is, they could have legitimate claim to substantial losses due to their trade secrets becoming public knowledge, even if it was due to them revealing where they were in the kernel, because the code to Linux was already public. Another point of interest is that, according to my legal sources, the amount of a trade secret infringement claim should not ever exceed (by much) the net worth of the company whose secrets were compromised, measured at the most recent point in time before the secrets could have been stolen. This could date no further back than the date that IBM (the supposed source of the trade secret infringement) began its official involvement with Linux, so SCO could ask for an amount on similar par with their maximum net worth at any point since then. The fact that they are asking for so much more than this (by orders of magnitude!) means that it would be impossible for this to be a trade secret violation claim alone.
[1] This is, interestingly, the heart of MS's claim against the GPL, but can be shown to be inapplicable because the GPL does not affect who owns the copyright on a product, it only affects who has the right to distribute the copyrighted work (in the case of the GPL, rights to distribute are granted by the copyright holder only to those who will not limit the rights to further distribute the original work or works derived from it). If a person wants to make a change to a GPL'd work, but does not want his changes subject to the GPL, he can simply choose to not distribute his changes, and he's fine. Also, since copyright does not protect ideas, one can freely examine the source to any GPL'd product for the purposes of self-education, and then apply that knowledge to a new work that would not be subject to the GPL. RMS would probably loathe this loophole, but it can't be helped --- knowledge, once learned, is inherently free to use for the person who has it, and without an NDA, no copyright clause can govern what a person does with it once it is in their head.
File under 'M' for 'Manic ranting'
No. The request is `tell me something I know everybody knows, and you know everybody knows, but SCO thinks is a secret.' The goal is to prove that everyone knows it.
There are reasons why democracy does not work nearly as well as capitalism.
-- David D. Friedman
Nope, it's Inigo. Read the book (as it's been said in another post, it is a good read).
SunOS (later the WIndowed version was called Solaris 1.x) up until 4.x was BSD 4.3 system...it was NOT a "free BSD" derivative, whatever you mean.
Solaris 2.x+ is very much System V Release 4. Sun were a member of that collective and Solaris still retains that structure.
You might want to check your facts.
-psy
Thats 32V and Release 7, not System V Release 4, they did make a few changes in the decade between the releasing of R7,32V and SVR4. Such as VM, SMP, ELF, shared objects, INET, kernel modules (SVR4.2), crap relating POSIX compliance, ooh and adding runlevels to /etc/init and then moving it /sbin/init, Screen control libraries, &c. The versions to which you refer are very primitive by todays standards, these things have 10k kernels (opposed to the 2meg kernels seen today) and considered 5 megs of FS space quite generous so obviosly they have added a bit of code to it.
IBM's OS/360 and VM/370 are in the public domain
but does that imply that Z/OS and Z/VM are public domain too? I think not.
ESR is married to a fiesty red-headed IP lawyer named Cathy. Maybe you should get a clue before trashing ESR and this strategy. I am sure that this will all make sense soon.
Things are seldom as they seem.