An article on LinuxJournal includes a copy of the NDA required by SCO of anyone who wants to look into their claims about trade secret/patent/copyright infringement vis-a-vis Linux.

It's not surprising to me that this NDA is as rock solid as it appears to be. In particular, they don't want those who sign the document to be able to disclose which files are at issue or any other identifying information about the alleged infringement.

I have an interpretation of this, although I will admit that I have no information outside of this document, the way that SCO has been behaving throughout this entire debacle, and a general understanding of how these kinds of cases go. I'm not a lawyer and I don't play one on TV.

With that out of the way, I think they probably have located some files (a handful) that contain bits of code they are pretty sure infringe their rights. However, their advantage in this case is the uncertainty that abounds in the UNIX community right now and therefore, the last thing they want to do is provide the community with a scope of work. Why?

For one thing, if you know what files are causing the problems, you can hurriedly release a version of the operating system that does not make use of those files, concepts, designs, etc. It might take a few months, but if the community is successful, SCO would have a hard time intimidating people going forward. Basically, a call would go out for everybody to upgrade to a post-SCO version of Linux and then it would be up to the folks at SCO to prove infringement and damages from the individual 1500 companies.

For another thing, just knowing the scope of the problem would provide the entire community with a way to determine what their level of risk is. For example, if the code in question only executes on certain platforms, it may be irrelevant on others. If, for example, the problems don't raise their heads on x86 platforms, the folks at SCO are going to have a hard time scaring most of the community. If you don't think this is a possibility, then consider that the folks at IBM have Linux running on more different processor platforms than any other vendor. Operations that are appropriate only for mainframes or PowerPC chips would not have much of an affect on the rest of the community, but IBM would have trouble with them.

Be all of this as it may, it doesn't change the fact that the Linux community is looking for a few good people to take up the challenge and see what SCO might or might not have. Unfortunately, under the current NDA, nobody related to Linux is interested in taking the risk, and the information that everybody wants to know will be kept secret under the NDA. At this point, it may be necessary to just wait until everything is disclosed (or not) at the IBM trial.