Here's some interesting evidence from 1999 and 2000, and it supports the evidence Groklaw has already found, namely that not only did IBM know, at some point during Project Monterey, that Linux was the future and that Project Monterey would be an upgrade path to Linux for many of its customers, so did everyone else, including SCO. Which SCO? Both of them. Santa Cruz and Caldera. They all knew.
Also, someone who worked for Caldera back then, Kurt Wall, has now stepped forward. He says he was at a Caldera company meeting after Project Monterey was killed, and that at the meeting, Ransom Love said it was a good thing that Project Monterey was no more:
I remember very clearly Ransom saying (in front of one of the all-hands meeting I mentioned earlier) that Caldera were pleased that the Monterey project had been cancelled because it would allow IBM to focus its efforts on AIX-5L, which would benefit both Linux and Caldera. Again, the emphasis was growing Linux to the point it would run efficiently on seriously high-powered hardware and get Linux into the data center.
I asked him to verify this statement, and he has conferred, he tells me, with a colleague, who confirms his memory. Remember that the Project Monterey contract was between Santa Cruz and IBM, not Caldera (now SCO) and IBM. More from Mr. Wall in a moment. The article, "Caldera-SCO deal may scuttle Monterey," by John Taschek, also presents a very different picture of SCO's financial realities than the SCO v. IBM complaint. As you recall, they painted a rosy picture of the company's business success back then, interrupted only by IBM's breaking the Project Monterey contract. The article, though, says UnixWare was already slowly dying, before IBM made the contributions to Linux that SCO is in a snit about, and that everyone saw that there was a shift happening to Linux as a business platform.
Here's what Taschek says about SCO's business situation and Linux, and I've marked in red the pertinent parts:
Caldera is in a predicament. Linux "enthusiasts" almost universally despise the company, yet Caldera remains one of the few trusted distributors of Linux software for business users. So what does Caldera do? It scores SCO, which is also almost universally despised by Linux enthusiasts. SCO, however, still has distribution power, although lately, of course, the company hasn't distributed a lot of anything.
I feel bad for SCO because its Unix Ware operating system is clearly class work. Unfortunately, the company has been manipulated nearly out of existence by a series of purchases involving UnixWare, market shifts and a couple of bad choices . . .
In this case, the shift is clearly to Linux as a business platform. Everyone can see that this is happening, but no company really knows how to take advantage of it, particularly because Linux is free and so are the enthusiasts who do most of the support work on it right now.
So business wasn't good, and Caldera was able to buy SCO assets as a result, and everyone saw a shift to Linux as a business platform. Why, then, did Caldera want to buy UnixWare assets, if it was dying a slow and agonizing death? Was it for the code, or for something else? Let's let the article tell us:
Caldera clearly believes that to get its technology into play, it has to have a distribution channel and a support staff. The stock/cash agreement is an inexpensive way to accomplish both goals.
What's completely unclear about this deal is what will happen with UnixWare. Caldera CEO Ransom Love said in published reports that it's status quo for UnixWare, which apparently means it will continue to suffer a slow, agonizing death until the IBM-backed Project Monterey rolls around.
And what about Project Monterey? Would that save them? Here's Taschek's opinion, back in August of 2000:
It once sounded like a killer operating system, but now I don't see a lot of potential in it. "Only good as an upgrade path from UnixWare." Could it get any clearer? Because of the delays on Merced, by the time Project Monterey was going to be ready, 64-bit Linux would beat it. Now can you understand what happened to Project Monterey? Why would any company choose to push a product that would be beaten hands down by Linux out of the gate by the time it was ready? Corporations are supposed to rationally analyze the market and give customers what they clamor for, are they not?
IBM has been marketing the heck out of Linux lately. The company is boasting about Linux on mainframes, Linux on NetFinity and even (eventually) Linux on AS/400s. There's not a lot of room for Monterey, especially since it's easy to speculate that 64-bit Linux will beat 64-bit Monterey out of the gate. This means that Monterey is only good as an upgrade path from UnixWare.
But, skeptics may say, that article was written in 2000; earlier, it was a different story. Was it? How about we go back to 1999, a full year earlier, to an article Jason Perlow wrote, "'Penguinitis' Sweeps Monterey UNIX Consortium."
We have only the first bit of the article, and we're wondering if any of you with Lexis can get the rest for us? Even if not, look what it says in the excerpt on Linux Today, dated August 20, 1999:
"Partners are increasingly Linux-happy--and even SCO seems to be hedging its bets.
"Linux fever is infecting even the staunchest Unix advocates, as evidenced this week at SCO Forum in Santa Cruz."
"While the partners involved in the Monterey Project--the initiative between SCO, IBM, Intel, Sequent Computer and Compaq Computer Corp. to create a high-volume unified UNIX--were upbeat on Monterey's prospects, they still had Linux on the brain."
So they *all* had Linux on the brain, back in the summer of 1999. Linux fever, he calls it. Even SCO.
UPDATE: Chris Brown and rm6990 found the complete article on Wayback. You can also read it, under the title "Unix Forum Cheers Linux." It adds the following information:
In a Project Monterey progress report issued this week, IBM explained its evolving to a multi-tier UNIX strategy, with Linux the operating system of choice for entry-level UNIX workstations and Internet servers, and Monterey replacing AIX for line-of-business and high-availability applications. The previous week, IBM announced at LinuxWorld Expo that it would be joining the Trillian Consortium, a group of companies, including SGI, Hewlett-Packard Co. and VA Linux Systems, working to port Linux to Intel's IA-64 architecture. . . .
SCO, too, has jumped on the Linux bandwagon, in spite of its role as one of the Project Monterey ringleaders. This week SCO announced its own Linux and open source professional services offering.
Many of the software vendors exhibiting at SCO Forum were demonstrating proudly Linux versions of their SCO offerings, and distributed demo CDs that ran on both platforms.
This is August of 1999, before Caldera bought the SCO assets. IBM was publicly announcing its Linux involvement, with SCO right there, at a Project Monterey press conference, no less. So when nowSCO told the court it had no idea IBM was supporting Linux and it was completely blindsided, was that true?
The market made any rational company go that way. SCO, oldSCO itself, was jumping on the Linux bandwagon. The handwriting was on the wall, and everyone knew it, the article says. To give you the full flavor of the time, here is the sole Talkback that has survived:
Murray Todd Williams - Subject:
Claims about the deliverable features of a product that is not even expected out for over a year (and mentioning features expected two years from now) reminds me a bit about NT and 95 before they were first released. As far as I'm concerned, Monterey isn't even in the picture until it actually manifests itself.
I'm amazed by Linux's laundry list for kernel version 2.4, which if released by year's end will be simply amazing. I don't think we can even begin to guess what Linux will have accomplished in that same amount of time!!
If you were IBM or any other rational company back then, what would you have done, faced with Merced delays and the explosion of interest in Linux and its increasing technical capabilities even back in 1999? Would you stick with a product no one much was interested in any more? That Linux probably could beat? Remember this is 1999, prior to IBM donating the code that SCO now objects to. The truly ironic thing is this: had Caldera stuck with Linux and IBM, they might have been the beneficiaries of Linux fever. Now, they are its road kill instead.
Now, about that ex-Caldera employee. His name is Kurt Wall, and he sent me the following information, and he also sent me a representative sample of the email he mentions he still has on a CD, so I could see it to verify what he writes.
Wall was a tech writer at Caldera back when the company was working on the Linux Kernel Personality. I'm sure you remember that it was code that SCO-Caldera worked on, to make it possible to run Linux applications on UnixWare and/or OpenServer, and no doubt you recall the accusations about Linux code being copied into the LKP, without giving back the modifications, as per the GPL. There is a dispute about that issue, but the point Mr. Wall makes is that SCO and Caldera worked very closely together on the LKP, and the whole point of the LKP was to let customers get what they wanted: Linux. Of course, back then, Caldera (now SCO) was a Linux
company. For that reason, one would think that IBM's support for Linux was to their benefit:
My name is Kurt Wall. I worked for Caldera from September 1999 to
February 2001. I started as a technical writer and eventually became the
manager of the technical documentation group. I wrote much of the
documentation for three releases of the OpenLinux desktop product, one
release of the OpenLinux server product, and the only release (that I
can recall) of the OpenLinux eBuilder product. I was actively involved
in the day-to-day work of creating and maintaining Caldera's Linux
Within perhaps two months of the announcement that Caldera was going to
buy the OS and professional services pieces of the Santa Cruz Operation
("old SCO"), personnel in both of Caldera's facilities (Erlangen and
Orem) were working with personnel from old SCO to combine operations.
We worked increasingly closely with old SCO from shortly after the
purchase was announced until I left in early 2001. I find it difficult
to believe that SCO's corporate memory is so bad, especially insofar as
some of the people still there were involved
at a high level with the IBM-related activities.
As Caldera proceeded with the purchase of SCO, there was a lot of
talk about the "Linux Kernel Personality," also known as "LKP." LKP
was an effort (actively pursued at SCO and at Caldera) to run
Linux on top of UnixWare in order to help Linux scale better on
large SMP systems. I attended numerous meetings about this, including
a couple of company-wide meetings in which Ransom Love, then Caldera's
CEO, used PowerPoint slides to illustrate the performance differences
between UnixWare on SMP and Linux on SMP (in 2000, the differences were
pretty stark, and Linux was coming out on the bottom). The intent was
*expressly* to leverage UnixWare's performance while taking advantage
of Linux's low cost and popularity and get Linux into data centers, a
niche into which Linux had not yet moved. Remember, we were all still
using kernel 2.2.
The point here is that SCO's own engineers, especially the kernel
engineers in the New Jersey office, worked very closely with Caldera's
engineers to get LKP working. I recall a release being delayed to wait
for some last-minute fixes. I have a CD-ROM full of email discussing
precisely this, and more. In another meeting, a conference call with
developers from New Jersey, Erlangen, and Orem, during which issues
relating to problems getting UnixWare and Linux to share file
descriptors and device nodes was discussed. In yet another meeting, we
debated whether and how to lay out the CD containing the LKP itself,
finally having to settle on the SCO engineer's insistence on stuffing
everything into the root directory (for a reason that eludes me now) -
those from Caldera were dismayed at that disorder and poor planning
such a CD layout betrayed, but that's a gripe for another day.
To be sure, as a technical writer, I wasn't involved in the the lowest-level, nitty-gritty detail. But, I was, as a member of Caldera's engineering group, close
enough to what was going on to be very clear about who was doing what.
Indeed, after I became the manager of the technical documentation, I
was even more plugged in to the deluge of email and meetings in which
some of the discussions took place.
Indeed, if any code was copied or at least studied back then, it was
Linux kernel code, because UnixWare was being modified to allow Linux
to run on top of it.
SCO has recently claimed to know nothing about IBM's intentions
vis-a-vis Linux and Project Monterey. It simply isn't so. I remember
very clearly Ransom saying (in front of one of the all-hands meeting I
mentioned earlier) that Caldera were pleased that the Monterey project
had been cancelled because it would allow IBM to focus its efforts on
AIX-5L, which would benefit both Linux and Caldera. Again, the emphasis was
growing Linux to the point it would run efficiently on seriously
high-powered hardware and get Linux into the data center.
So, Caldera and SCO had engineers looking at Linux code and Unix code at the same time? I'd guess a reasonable question would be whether there was any effort to keep them clean-roomed off from each other? Or was it a free-for-all, with the same engineers working on both?
If there are others out there who have the courage to step forward, Groklaw is here. Hey, you know the old expression: if everyone stands up, they can't shoot *all* of us.