I don't see how it can get much clearer that one of Project Monterey's goals was to include AIX on POWER and that SCO knew all about it at the time. The more I research, the clearer it gets. This is the third article just on this subject that Groklaw has published within a week (cf. this article and this one, both from April 4. See also this article on SVR4). Not only that, I now see that one other goal of Project Monterey was to have a UNIX that could run Linux applications, so that the eventual transfer to Linux, which is what IBM expected was the future for ebusiness, would be easy for businesses to make, when Linux was mature enough. The L in AIX 5L stands for Linux.
Needless to say, that pulls the rug out from under SCO in two ways. First, it has claimed it had no idea IBM was investing in Linux until around 2000. And it has apparently claimed that IBM had no right to use AIX on POWER, that IBM only did that after Project Monterey was declared dead in 2001. My research has convinced me that isn't the case. But, as usual, don't go by my opinion. I'll lay out the research we've done, and you can reach your own conclusions. I've marked the parts I think are most significant in red. Let's start by reviewing what SCO claims in its Second Amended Complaint about Project Monterey and IBM's Linux involvement.
Here's how SCO defined the purpose of Project Monterey in its Complaint:
53. As SCO was poised and ready to expand its market and market share for UnixWare targeted to high-performance enterprise customers, IBM approached SCO to jointly develop a 64-bit UNIX-based operating system for a new 64-bit Intel platform. This joint development effort was widely known as Project Monterey. . . .
57. On or about May 2001, IBM notified plaintiff that it refused to proceed with Project Monterey, and that IBM considered Project Monterey to be “dead.”
That IA/64 effort was only one part of Project Monterey, the part SCO was involved in, but as I have discovered in my research, there was a lot more to Project Monterey than the part SCO told us about. Here's how IBM, SCO and Sequent described the project in a joint press release dated April of 1999, just five months after Project Monterey began:
IBM, SCO and Sequent today said they have successfully completed initial tests of the Project Monterey operating system running on the Merced simulator for the Intel IA-64 architecture. Project Monterey is a high-volume, enterprise-class, commercial UNIX operating system initiative launched last October.
The participants in Project Monterey declared that in less than five months of development, the prototype for the UNIX operating system for IA-64 is up and running. The testing, conducted at a development center in the United States, marked the first development milestone in Project Monterey.
Also unveiled today were broad software vendor support, accelerated product roadmaps through 2001 and the launch of a comprehensive developer program
The goal for Project Monterey is to establish a high-volume, enterprise-class UNIX product line that runs across Intel IA-32 and IA-64 processors and IBM's Power processors in systems that range from departmental to large data center servers. In addition, UNIX vendors conforming to the UNIX Developer's Guide-Programming Interfaces (UDG-PI) specifications are supported by the Project Monterey family.
As part of the Project Monterey initiative, a UNIX operating system is being developed for Intel's IA-64 architecture using IBM's AIX operating system's enterprise capabilities complemented with technology from SCO's UnixWare operating system and Sequent's enterprise technologies. In addition, IBM will license AIX technology to SCO for inclusion in UnixWare and promote this offering to the IA-32 market. . . .
The Project Monterey developer program will include porting guides and a common set of APIs for IA-32, IA-64 and IBM Power processor platforms. Developers also will be able to leverage an extensive set of enterprise middleware from IBM and other software vendors who participate. Project Monterey will offer key developers access to porting centers worldwide, as well as ongoing developer events hosted by IBM, SCO and Intel throughout the year. The porting centers will provide developers with technical support, porting/enablement, performance testing and technical education on the new UNIX operating system on IA-64. Later this year, the IBM Solution Partnership Center in Waltham, Mass., is expected to extend its capabilities to support UNIX development on the Intel platforms.
Doug Michels, president and CEO of SCO, said, "We've not only completed the initial port to Merced in record time, but we've also set up an integrated product road map and developer program to guide our business partners in deploying this exciting high-volume UNIX system to businesses of all sizes. The Project Monterey partners already drive a huge portion of a worldwide UNIX System business that totals billions of dollars every year, and our momentum continues to increase as other OEMs and ISVs join Project Monterey.". . .
Unix Developer's Guide- Programming Interface
In a separate initiative, a group of industry leaders, including IBM, SCO, Sequent, Compaq and Intel, recently announced the UNIX Developer's Guide - Programming Interfaces. The UDG-PI is a collaborative effort to develop and publish guidelines that software developers and system manufacturers can use for UNIX operating systems running on the Intel microprocessor architecture. These guidelines are designed to help reduce development, maintenance and test costs for software developers and system manufacturers.
So, did SCO know about POWER? They obviously did. Doug Michels, then CEO of SCO (Santa Cruz) is quoted in the press release that mentions it as a project goal. It also refers to the programming interfaces that SCO helped to provide. So they knew. They helped. Obviously, oldSCO didn't think there was a license issue regarding AIX on POWER.
Here's another press release from March of 2000 from Lawson Software, and it also defines the goals of Project Monterey:
ST. PAUL, March 14, 2000 - Lawson Software, the Enterprise behind the Internet, today announced that it has joined Project Monterey, a UNIX operating system (OS) initiative led by IBM, along with SCO, Intel, Compaq and other industry leaders. Lawson also announced that it is "Monterey-ready," meeting the criteria to port its applications to the Project Monterey UNIX operating system for Intel IA-64 processors (code-named Monterey/64). . . .
"We feel IBM's Monterey initiative plays a key role in extending enterprise-ready e-business applications into the volume UNIX marketplace on PowerPC and Intel platforms," said Bill Keatts, vice president, Research and Development, Lawson Software. "As an early adopter of open systems, Lawson knows from experience the vitality and benefits of open platform computing. Our n-tier, component-based software architecture is ideally suited for the kind of flexibility Project Monterey is designed to support. Our customers will gain additional options for deploying their Lawson e-business solutions and the reassurance of quality from our close association with other top e-business vendors."
As part of Monterey/64, an enterprise-volume, shrink-wrapped, UNIX operating system is being developed for Intel's Itanium (IA-64) processors. The OS uses technologies from the IBM AIX operating system, SCO UnixWare operating system and NUMA-Q's Dynix/ptx enterprise technology.
"We are committed to providing our customers with world-class enterprise e-business solutions on UNIX," said Dr. Tilak Agerwala, vice president, UNIX marketing and product management, IBM Enterprise Systems Group, IBM. "Lawson's commitment to Monterey/64 demonstrates its position as an open computing e-business leader. Our customers will get significantly better flexibility to deploy enterprise e-business solutions on IBM POWER and Intel UNIX platforms"
The objective of the Project Monterey initiative is to establish an enterprise-volume UNIX product line that runs across Intel IA-32 and IA-64 processors (Monterey/64) and IBM's Power processors in systems that range from departmental to large data center servers. Project Monterey has garnered widespread support from major ISVs in the e-business, business intelligence and enterprise resource planning (ERP) market segments. Additional information on IBM and UNIX can be found at www.ibm.com/servers/aix. More information on Monterey/64 is available on the Web at www.ibm.com/servers/monterey; www.sco.com/monterey; and www.sequent.com/products/software/operatingsys/monterey.html. . . .
Here's a May 2000 press release by VERITAS:
VERITAS* Software Corporation (Nasdaq: VRTS), the leading enterprise-class application storage management software provider, today announced a business agreement with IBM in support of AIX/Monterey*. As part of the agreement, VERITAS Software will be porting and optimizing its complete set of storage management solutions, including VERITAS Volume Manager and VERITAS File System, to AIX/Monterey for IBM POWER and Intel IA-64 processor-based systems.. . .
AIX/Monterey is a code name that refers to both AIX for the IBM POWER processors as well as the future delivery of the UNIX operating system for Intel's IA-64 Itanium processor-based systems.
So POWER was no afterthought that IBM came up with post the death of Project Monterey, was it? SCO wasn't involved in that aspect of the project, because it wasn't involved in the POWER platform, only Intel, but Project Monterey was designed for POWER every bit as much as for Intel. Here's the part that SCO was involved in, according to a blurb on IBM's Project Monterey Participants page:
SCO - UnixWare 7
SCO, the market share volume leader of UNIX operating systems (and developer of UnixWare 7), and IBM will collaborate to accelerate enhancement for IA-32. Also, SCO and IBM will work together to co-develop and market this UNIX for the IA-64 based market.
The link to the SCO page is now blocked by robots.txt by SCO, so Wayback doesn't have it.
What about Linux? Was SCO in the dark about IBM's interest in Linux or its plans for the future? It shouldn't have been. Here is a February 1999 announcement by IBM and Red Hat that they were in an alliance to offer IBM systems running Red Hat Linux. SCO, in its Second Amended Complaint, implies that it wasn't until 2000 that SCO realized IBM was getting involved in Linux:
89. On or about August 17, 2000, IBM and Red Hat Inc., the leading Linux distributor, issued a joint press release through M2 Presswire announcing, inter alia, as follows:
IBM today announced a global agreement that enables Red Hat, Inc. to bundle IBM’s Linux-based software.
IBM said it would contribute more than 100 printer drivers to the open source community. With these announcements, IBM is making it easier for customers to deploy e-business applications on Linux using a growing selection of hardware and software to meet their needs. The announcements are the latest initiative in IBM’s continuing strategy to embrace Linux across its entire product and services portfolio.
Helping build the open standard, IBM has been working closely with the open source community, contributing technologies and resources.
This 1999 announcement, however, shows that 2000 is a misleading date. The 2000 press release SCO quotes even says that it was "the *latest* initiative in IBM's continuing strategy to embrace Linux," implying that there were earlier initiatives. So SCO had plenty of notice. In fact, according to this IBM paper from May of 2000, their involvement dates back two years, to 1998, which I believe would predate Project Monterey, if they mean May of 1998. Project Monterey began in October of that year.
Not only that, Caldera was participating with IBM in a project, Project Trillian, "formed early in 1999 to port the Linux operating system to the Intel IA-64 architecture", so SCO's apparent pretense that it didn't know is just that, unless they "forgot". Or could it be that the SCO of today doesn't really know what oldSCO, Santa Cruz, was doing, because they are two different companies? Here's the info on IBM's earlier Linux involvement, including Project Trillian with Caldera, which now calls itself SCO:
IBM is further embracing Linux and open source software as key components in taking e-business to the next level. Linux is evolving toward an industry standard. Along with other open standards, such as HTTP, XML and TCP, we view Linux as playing a pivotal role in bringing interoperability to disparate server platforms and providing customers with an open, integrated e-business structure. IBM continues to work with the Linux and open source communities to support this evolution and support critical standards across its servers.
IBM believes that Linux will help drive the long-term growth of the Internet by providing an open application platform that can harness leading-edge technologies and simplify customer choice. The common application platform will help ensure software interoperability across heterogeneous servers. IBM is an active participant in the open source community, not only embracing its software, but also contributing significant skills, technology and resources where appropriate.
While IBM is further strengthening its commitment to Linux and open source software, IBM's participation actually dates back two years. . . .
IBM is actively participating in the following established open source projects:
* Apache -- a working group that has created the de facto standard HTTP server used in most Web servers worldwide;
* Jakarta -- a working group dedicated to providing Java-based Servlet and JavaServer Pages implementation for the Apache Web Server;
* Mozilla -- a working group building an open source browser; and
* Trillian -- The Trillian Project also was formed early in 1999 to port the Linux operating system to the Intel IA-64 architecture. The project currently includes Caldera, CERN, Cygnus Solutions, Hewlett Packard, IBM, Intel, Red Hat, SGI, SuSE, TurboLinux and VA Linux Systems.
Finally, and most tellingly, here's the first three pages of a paper [PDF] from March of 2000 in which IBM explains why it thought Linux was going to replace Unix eventually as the inevitable choice for business and what its motivations and goals were for Project Monterey. It's called, "Why Linux? What About AIX/Monterey? And Why IBM?" The date the paper was posted by DataTrend Technologies is listed in the url as being March 20, 2000, more than a year before IBM declared Project Monterey dead, according to SCO's account, and five months before the first date SCO lists in its litany of complaints about IBM's contributions to Linux. But what it describes is why IBM started Project Monterey to begin with. This paper shows me that SCO's claims do not match the evidence I am finding:
Why Linux? What About AIX/Monterey? And Why IBM?
The evolution of e-business is presenting customers with a new set of challenges:
- Integrating multiple applications and data across a spectrum of operating systems, running on multiple servers, each with unique management requirements, distributed geographically, and crossing company borders.
- Responding to the business impacts of delivering performance and availability in environments where peak demands and business growth may be difficult to predict
Customers choose platforms based upon application availability, frequently making compromises and adapting their business to specific technologies. Should they outgrow the capabilities of their current platforms, implementing changes necessary to grow can be extremely disruptive to their businesses.
Linux is becoming a rallying point for the industry, one that will drive the evolution of the next generation of e-business. Linux is not just another operating system. It represents a collaboration of the best programmers in the industry coming together to create an operating system that works on any hardware platform. This industry change will address many of the new problems that organizations face today, allowing companies to adapt technology to their business, rather than having to change their businesses to fit technology.
As an operating system, Linux is already the fastest growing member of the UNIX** family. Linux is most popular on small to midsize, Intel based PC servers, as well as clusters of such servers. It is very popular as a platform for modest sized web and commerce servers, ISP/ASP applications, dedicated networking functions (such as web-infrastructure, file/print, LAN, firewall, DNS, e-mail...), supercomputing clusters, as an embedded operating system for new server and client application appliances, as well as the development platform for many innovative “next generation” applications.
As a technology, we expect that Linux, over time, will develop into the standard application development platform for the spectrum of applications that comprise our customers’ e-business solutions. By driving standardization of programming interfaces, Linux will do for applications what standards, such as TCP/IP and the Internet, have done for networking. Linux is uniquely positioned to create an environment where applications can be developed once and deployed across a broad range of servers and environments. This application portability is essential if customers are to be able to adjust to the dynamic nature of e-business.
The impact of Linux is that customers will be able to independently select: 1) the applications that are best suited to their business needs; 2) platforms matched to their performance and capacity requirements and 3) operating environments that impart critical characteristics upon the applications.
Over time, Linux will become a viable enterprise UNIX system, capable of running more workloads requiring high scalability and industrial strength. We will work with the Linux community to help build such an enterprise Linux offering. This will take years, with the rate and pace being determined by the Linux community. Operating systems evolve slowly, and it is uncertain when Linux will have comprehensive enterprise capabilities.
What About AIX/Monterey?
As Linux continues to mature, customers will continue to search for the most robust UNIX, one that will help them evolve to the future. AIX/Monterey, the next evolution of IBM’s UNIX operating system, AIX*, brings all the advantages of a highly scalable, available, industrial strength UNIX operating system to the world of Intel-- today.
AIX has been recognized year after year as the #1 UNIX for customers’ critical applications 1. With the rapid growth our customers face resulting from the evolution of e-business, the industrial strength capabilities of AIX are becoming even more important. The next release of AIX will be AIX/Monterey (which will soon be re-branded). AIX/Monterey is the result of a major UNIX operating system initiative, led by IBM, along with SCO and Intel, with support from leading industry software and system vendors. AIX/Monterey is a volume, enterprise-class, UNIX operating system that today runs across Intel’s 32-bit architecture (IA-32) and across the POWER architecture. It will be extended to IA-64, Intel’s 64-bit architecture, later this year. AIX/Monterey supports systems that range from departmental to large data center servers. AIX/Monterey will incorporate the best of AIX as well as IBM’s DYNIX/ptx, the operating system from Sequent that runs NUMA-Q; and SCO’s UnixWare, the UNIX operating system that runs on today’s volume Intel platforms. Later this year, IBM will deliver AIX/Monterey/64 2 offering customers who prefer systems based on Intel’s 64-bit Itanium 3 processors the same capabilities provided by AIX on the RS/6000*.
As Linux addresses our customers’ needs for high volume, low cost UNIX operating systems, AIX/Monterey will continue to address our customers’ need for industrial strength, enterprise UNIX platforms across Intel and POWER based systems. We will build strong Linux compatibility in AIX/Monterey to help make the deployment of Linux applications on AIX/Monterey easy as well as facilitate the deployment of AIX/Monterey applications on future releases of Linux. Application portability is expected to precede the maturation of Linux, providing the foundation for customers to make this shift smoothly. In addition, we will collaborate with the open source community to contribute AIX/Monterey technology to Linux in an effort to help build a better Linux.
No other UNIX vendor is as well positioned to help customers adapt to the Linux evolution. Customers who invest in AIX/Monterey today can do so confident in the knowledge that IBM plans to make applications that run on this powerful new operating system migrate easily to Linux in the future, helping to protect their investments in hardware, applications, data, processes and skills.
Why IBM: As customers today invest in IBM AIX/Monterey, they will be uniquely positioned to take advantage of Linux solutions as they emerge.
1. IBM’s UNIX/Linux strategy is to provide a high degree of application portability across system and software environments.
- Monterey/64 provides a growth path for 32-bit Intel based solutions on DYNIX/ptx and UnixWare to Intel’s IA-64 (Itanium) environment and provides the flexibility for today’s AIX solutions to be deployed on systems using both IBM Power and Intel microprocessors.
- The introduction of the Linux Application Execution Environment later this year is intended to provide both binary compatibility for many Linux binaries on AIX and Monterey/64 and source compatibility allowing Linux applications to be recompiled for native execution on AIX and Monterey/64.
- IBM continues to contribute enterprise technologies to the open source community. This contribution also will narrow the gap between our Linux products and UNIX, further increasing application portability.
2. As Linux becomes more mature as an operating system, customers will be able to begin to use their existing servers, applications, data, skills and business processes with little to no change.
Could IBM have been more forthright and open about its plans and its reasons? Project Monterey was the stopgap, in a way, I gather. It worked for the enterprise right away, and it was a path to smoothly move to Linux as it matured. And the benefits of Linux are clear as far as customers are concerned. You can run it on any hardware. Interoperability is the goal in a networked world, and Linux is perfect for that. So while proprietary Unix companies might not like it, the simple truth is, customers do and they did. They adopted Linux so fast, Project Monterey was dead before it was fully born. The market did that. Intel didn't help, I gather. But where is there any IBM deception or dirty tricks? I can't see any here.
What about SCO? How could they not know all this? Unless we attribute bad motives, which I am reluctant to do without clear proof, the only explanation I can come up with is that newSCO hadn't got a clear picture of what oldSCO was doing, and when it read some IBM emails in discovery, it misinterpreted, leaping to conclusions maybe. It's not hard to do that when you want something very much. And yet, as Caldera, then a Linux company, it certainly knew about Project Trillian, so what's the deal with that lapse of memory, if that is the problem? Everyone back then knew that Linux mattered, the evidence indicates, and they all were trying to help Linux mature. That includes oldSCO (Santa Cruz), and it includes Caldera, now calling itself SCO. Revisionism has reared its head, and the stories being told now are different. But what does the evidence indicate to you?
If we wish to be kind, let's say that newSCO didn't have a clear picture of what oldSCO was doing in Project Monterey before. It does now.