This is a good day for a stroll down memory lane.
SCO is accusing IBM of helping Linux to scale by donating "their" code to Linux inappropriately. But they forget that OldSCO wanted Linux to scale in precisely the areas IBM is alleged to have helped it to do so. Not only that, they told the world they were helping it to do so. Let's take a look at two articles from the year 2000 and you'll see what I mean, first SCO Answers Questions About Linux, dated May 5th, 2000 and then The latest Linux: SCO, dated May 12, 2000 (thanks to Carlos Pruitt, Jr. for finding them). Groklaw reported its research on OldSCO's contributions to Linux back on August 11, which you may wish to review also. After we've completed our little stroll, I think you'll agree that IBM has every reason to wonder what SCO is complaining about. You may also note an odd coincidence.
OldSCO had plans back in 2000, which it announced publicly, to port into Linux the same technology that they now say was stolen and inappropriately donated by IBM. Of particular interest is the first article, "SCO Answers Questions About Linux":
David McCrabb, President of SCO's Server Division in May 2000, answered
questions about Linux on Slashdot. Yes, Slashdot. McCrabb stated:
"SCO is accelerating its participation in, and contributions to, the
Open Source Community. In some cases, we will be taking current
technology that we think is needed in the Linux market and driving it
forward as the project maintainers. Right now, we are focusing on
bringing some of our high-performance Intel development tools to Linux."
McCrabb goes on to elaborate:
"Enterprises building their businesses on a server platform are
interested in reliability and availability. Although we believe in a
high degree of reliability that comes from the level of code inspection
provided by the Open Source Community, we feel it needs to be quantified
with benchmarking statistics like MTBSS. This opens a number of possible
further improvements -- journalizing file systems, support for hot-plug
PCI, multi-path I/O -- things that make is easier to never bring the
system down, or to recover the system more quickly."
Um...you mean, like JFS, by any chance? The second article, "The latest Linux: SCO", reveals that SCO said it was working on
putting some interesting technology into Linux:
SCO, which has been around for more than 20 years, this week will
unveil plans for its own brand of Linux, one that will come with the
kinds of management, clustering and Web serving technologies that have
helped the company become a leading Unix supplier. . . .
Were Old SCO's engineers already placing SCO technology into Linux back
in 2000? These articles indicate that the technology was at least being
developed in-house and there was a plan, publicly announced, to do exactly what they now are saying IBM did but should not have done. Isn't it an odd coincidence that these are the same technologies at issue in the IBM lawsuit?
SCO is expected to announce 32- and 64-bit versions of Linux for
Intel-based servers, which will be available in the fourth quarter of
this year. In early 2001, SCO plans to deliver a 32-bit Internet
Infrastructure Edition that will come bundled with a Web server and
other IP applications. The company is also working on a 64-bit edition
for service providers, including ISPs and application service providers,
which will feature special billing and management tools.
The company is also expected to explore the following areas:
- Building the Linux clustering capacity to be in line with SCO's
NonStop Clusters technology, which scales to 12 or more boxes with
advanced reliability for data and applications. Current Linux
clustering technology is generally limited to two or four nodes.
- Beefing up Linux's symmetric multiprocessing capabilities.
Currently the number of CPUs per Linux server is usually limited
to eight; UnixWare can run on servers with up to 32 CPUs.
- Managing multiple Linux servers as well as applications from a
single console as if they were a single system.
- Improving security and the ability of Linux to handle applications
such as e-mail, including instant messaging.
- Adding online support services and documentation.
And does it sound to you like they never intended to release this technology under the GPL, that it was unwittingly and very much against their will? Do they think we were born yesterday?
This might be a fruitful area for some more research, perhaps during discovery. How far did OldSCO get with implementing its plans? Is it possible OldSCO is responsible for the "infringing" code being in Linux?
SCO and OldSCO are not the same company, of course. OldSCO is today's Tarantella. And SCO was Caldera, plus part of the product line of OldSCO. But their legal documents make the claim that they have been offering UNIX for 20 years, which Caldera didn't (it wasn't in existence that long, for one thing), so they seem to be asserting a connection to OldSCO. Do you get the impression that SCO has forgotten OldSCO history? Or that after it decided to do a U-turn on Linux, it decided to act like OldSCO never said and did what these articles say they said and did? Well, if they honestly forgot, that's what Groklaw is for, to help them remember. So here you are, SCO, a helpful reminder, from our little stroll down memory lane.