They really want you to see this, apparently, so there will be "a replay of the teleconference, made available 60 minutes following the live Webcast. This replay will be available for 30 days following the live event." [Update: You can get it here]

Exhibit D -- Santa Cruz Operation Inc. Reference Source Code Agreement No. SOFT-2538

I find it curious that SCO bundled both Amendment X and SOFT-2538 together as the two documents seem to be independent of each other. Amendment X clearly modified the original agreements back to Exhibit A. As such it seems to apply to the System V source code and documentation listed in excruciating detail in the original contract (Exhibit A). The February 1998 agreement, SOFT-2538, explicitly applies to "reference software products" ordered after the signing of the agreement. Without a list of those orders SOFT-2538 seems to be essentially meaningless -- it states the detailed terms of a license for nothing.

Here is the language I think they are referring to:

Seller agrees that it shall use the Licensed Technology only (i) for internal purposes without restriction or (ii) for resale in bundled or integrated products sold by Seller which are not directly competitive with the core products of Buyer and in which the Licensed Technology does not constitute a primary portion of the value of the total bundled or integrated product.

This was done as a group project, with lots of volunteers working on one section and then putting it all together, and it was hard work, because the original was so unclear that a lot of the times, it was a matter of typing from scratch. Praise and thanks to all. I especially want to thank Edward MacNaghten for pulling all the pieces together into one document. Also to the team: in the order I heard from them, with more to go -- Steve Martin, beast, Philip Stephens, Chris Brewer, Rand McNatt, Bill Sharrock, and Thomas Le Page, and some others too modest to be mentioned but who also made a contribution.

[Update: Here is the PDF. This is not a complete version, however. Note that paragraph 4.11 mentions that the Schedules were not finished. Later, more were added, and the missing schedules are found here, along with various amendments to the schedules, along with amendments to this APA (1, 2, Technology License Agreement, and Bill of Sale, all on our Contracts page under the subheadings Novell and Santa Cruz.

Later, IBM also filed a copy of the APA [PDF] in SCO v. IBM, complete with all attachments, 98-pages in all, but it has a pagination error. Page 98 should go between pages 66 and 67. Compare here. And Novell also filed it in two parts, Part 1 and Part 2 [PDFs] in the SCO v. Novell litigation, and it is paginated correctly.

Also, with so many interpretations offered, it might help to look to the reasons each side gave at the time, as to what was intended. Here's Novell's 1995 press release and here's one from SCO, posted at the time to comp.unix.sco.misc. Both said that the idea in 1995 was to merge OpenServer and Novell's UnixWare.]

"November 14, 2003 - The Open Source Development Labs (OSDL), a global consortium of leading technology companies dedicated to accelerating the adoption of Linux, today confirmed that both OSDL and Linux creator Linus Torvalds had received subpoenas on Wednesday from attorneys for The SCO Group. The subpoenas were issued in The SCO Group's pending litigation with IBM. The subpoenas request that OSDL and Torvalds produce documents for use in that dispute.

"OSDL has agreed to fund legal representation for Torvalds and any other employees of the Lab who may become involved in the litigation.

"OSDL is represented by AterWynne LLP."

So, OSDL will pay for Linus' attorney, whatever firm he eventually chooses to retain. The fact that they are asking for documents only doesn't mean they will never seek to depose Linus; it just means that those of you booking flights in your dream world to get to watch Linus and rms being deposed can stop daydreaming. It may happen down the road, but it doesn't look like it'll be now.

Should there be such "high profile" legal moves, Skiba writes, the stock could pick up nicely, because he says there has been "a relative dearth of news flow over the past several months". He's marking on a curve, evidently. Note also the disclaimer at the end:

"Deutsche Bank and/or its affiliates(s) makes a market in securities issued by the following companies(s): SCO Group, Inc. (The)."

Oh. No agenda there, then.

Wall Street may go wild on this report, I suppose, briefly, because hype and emotion is all, it seems with those hungry dudes, at least the so-called "small" investors. The distinct impression I am getting, being new to finance, is that the big fish in the loop depend on the little fish outside of it going into a feeding frenzy on cue, but what do I know? I'm no financial analyst.

Dion Cornett of Decatur Jones Equity Partners, the only other analyst following SCO closely, disputes the DB math, and shows in his October 27 report on SCO that, viewed rationally, no matter what SCO does, in his opinion, there is no way SCO can reach a $45 target from the licensing program, even at its most successful, and his math and his logic is compelling. He rates SCO "Underperform", which they define as meaning "returns at least 15% below the appropriate index". Part of his reasoning is that there are some workarounds that will make it hard for SCO to prove who is and who isn't using SCO code, even if you accept that any of it is theirs to claim, including a survival strategy of sticking with the 2.2 kernel for now and just upgrading applications on top. It's an interesting idea that technically could work for most SMBs. Some more financial facts on SCO on this Forbes page.

With that introduction, here are relevant snips from each analyst's report, including details on the survival strategy.

This is SCO's new attorney for the Red Hat case, by the way, so it's our first chance to see him at work. They again tell the judge that they don't want to do discovery in Delaware until she rules on their Motion to Dismiss. If they win, they argue, then there is no need for discovery. If they partly win, they might not have to turn everything over that Red Hat is asking for. It wouldn't be fair to make them show their cards now, and then if they win, it's too late to take it back.

Again they argue that most of the issues are going to be handled in the Utah case (of course most isn't generally enough, but they slide right over that point), so in the alternative, they'd like to let the IBM case go to a resolution first and then do Red Hat, like in 2005 or 2006. Say, is that when Longhorn comes out? What a coincidence.