Basic problems with Shiney Object Intellectual Property (SOIP)


July 23, 2006

Other than the fact that I have never heard of it before, and was specifically taught in law school that it does not exist, and IMHO, the contract cannot be read to create it, I believe there are other problems with SOIP. (SOIP is the innovative theory that a contract term can prevent another party from disclosing home grown code when the first party does not even know what is in the code.

First, what, exactly, does it cover?

Does it cover information in the public domain? IIRC, D.E.Knuth's Art of Computer Programming, Vol. III (1973) discusses semaphores. This puts semaphores in the public domain. If the argument from "The SCO Group" is taken literally, then if IBM puts semaphores in AIX, it can never again sell a product with semaphores other than AIX or Dynix, and can never again discuss semaphores. This relates to any disclosure or use, whether or not it is part of an operating system. To me, this is patently absurd.

Suppose it is information in the public domain, but patented by IBM before inclusion into Dynix or AIX. IBM may prevent the use of the method by anyone else, but by contract, IBM may not use it except in AIX. Never mind that AT&T and successors did not consider it because it was not *invented* at the time of the contract. To me, this is patently absurd.

Suppose it is a new technique discovered by IBM, and IBM files a patent application on it before including it in Dynix or AIX, but before the patent is granted. Under SOIP theory, IBM violated the terms of the contract by not immediately dismissing the patent petition, to prevent public disclosure. IMHO, this directly violates the spirit of the Patent statutes, and is against enacted public policy. In my opinion, this is plain nonsense.

Suppose it is a secret technique owned by someone other than IBM, and IBM includes it in AIX or DYNIX by license. Under SOIP theory, IBM could not use the licensed technique in any other product after that, and could not disclose it, even if the licensor granted them specific permission to do so. Again, this is a technique that is unknown to AT&T or successors. This is absurd.

Suppose it is simply home-grown code from another project, but later added to AIX or Dynix. The code itself is held as confidential, but IBM makes no effort to protect the methods and concepts. In the absence of confidentiality, there is no trade secret status. Yet IBM would be held to have breached the contract. Again, this is absurd.

If it were a home-grown trade secret, not suitable for patenting, there is still the problem of IBM using it in AIX or Dynix, then using it in another product. If the product is not an O/S, then "The SCO Group" is not damaged, much less damaged in a non-recoverable way. How would "The SCO Group" even know that a breach ocurred, without the discovery process of a lawsuit to ferret out the disclosures? The next, and uglier step: to identify such disclosures, "The SCO Group" would have to have an unlimited right to review all home-grown trade secrets in AIX or Dynix, and compare them with all home-grown trade secrets in every other IBM product to make sure that IBM was not covertly releasing the information. Just try getting some licensed attorney to believe that that clause, as it stands, alone, grants an unlimited privilege to AT&T or its successors to review every trade secret of every product sold by IBM. This guts the whole notion of "trade-secrets". It is a reverse-trade-secret clause. That the plain meaning of the words in the clause might have this result is simply absurd.

It makes no sense to me.
AllParadox - Retired Attorney, no legal opinions, just my opinion.
< EOM >

11:59:02 PM

Re: Basic problems with Shiney Object Intellectual Property (SOIP)


July 24, 2006

[Just try getting some licensed attorney to believe that that clause, as it stands, alone, grants an unlimited privilege to AT&T or its successors to review every trade secret of every product sold by IBM.]

Apparently SCO managed to convince their own licensed attorneys that SOIP was a valid theory. I suspect they are going to have somewhat more difficulty convincing a Federal Judge that IBM, the holder of the world's largest patent portfolio, inventor of too many operating systems to list, agreed to allow AT&T to control their IP portfolio in exchange for (at the time) a few hundred thousand lines of C code. I think they would have a hard time convincing a Judge that Fortune Ssytems would have agreed to such a license, or any other 'for profit' Unix licensee.

Aside from the fact that noobdy that signed the agreement interprets the terms the way new SCO does, and new SCO wasn't even involved in those agreements in any way.

< EOM >

3:16:21 PM

Source: Investor Village SCO Board [ ]

Copyright 2006