|This article needs additional citations for verification. (June 2012)|
Taligent (a portmanteau of Talent and Intelligent) was the name of an object-oriented operating system, and the company dedicated to producing it. Initially started as a project within Apple to provide a replacement for the Mac OS, it was later spun off into a joint venture with IBM as part of the AIM alliance, with the purpose of building a competing platform to Microsoft Cairo and NeXTSTEP.
The development process never worked, and Taligent is often cited as an example of a project death march. Apple pulled out of the project in 1995 before the code had been delivered, and went on to start the Copland project to replace it. Copland quickly proved even more disastrous than Taligent, while IBM finally delivered a working version known as CommonPoint but it saw little use and quickly disappeared from IBM's catalogs. Taligent was officially dissolved in January 1998.
Pink and Blue
What would eventually become Taligent started in a roundabout way in 1988. After Apple Computer's release of System 6 that year, key engineers met to decide the future direction of the System software. Ideas were written down on index cards and pinned to the wall. Ideas that were simple and could be included in a new version of the existing software were written on blue colored cards, those that were more advanced or took longer to implement were written on pink cards, and "far out" ideas on red cards.
A new operating system, code-named Pink, was planned based on the ideas written on the pink index cards. Pink was to be a completely new object-oriented OS implemented in C++ on top of a new microkernel, running a new GUI that nevertheless looked and felt like the existing Mac. In addition to running programs written for Pink, the system was to be capable of running existing Mac OS programs. Many ideas from the red cards were later folded in.
Efforts to develop Pink started around 1989, although at the time the effort was primarily a research effort. By this time, however, the team writing the system based on the blue cards (now known as the "Blue Meanies") were well advanced on what would be released in 1991 as System 7. The problem was that System 7 was so large in memory terms that it would barely fit onto existing Macintosh systems, meaning that if Pink were going to run Mac OS programs by emulating System 7, it would have no room left over for itself.
Meanwhile, corporate infighting at Apple doomed Pink. To those working on Blue, Pink was seen as a project that might steal visibility from their own work. As the turf war grew, engineers started to abandon Pink to work on Blue, and whole projects were brought into one group or another in a flurry of empire-building.
Magazines throughout the early 1990s showed various mock-ups of what Pink would be like. One true innovation of the system was the People, Places and Things metaphor that attempted to provide the user with tools to easily move documents around between people and things (like fax machines) as easily as they could print them using contemporary technologies. The system also added a component-based document model that was similar to Apple's OpenDoc.
As development dragged on, in 1991 Apple entered the AIM alliance with IBM and Motorola. This was ostensibly to develop a new reference platform for PowerPC based personal computers, but IBM and Apple went on to form two other partnerships at the same time, Kaleida Labs to develop multimedia software, and Taligent to develop Pink into a cross platform OS. IBM had extensive experience in object-oriented programming, notably their well-respected VisualAge Smalltalk programming system. They also had experience in microkernel design as a side-effect of their Mach based Workplace OS efforts.
The original Apple team was expanded with the addition of a very small number of IBM engineers, as well as a new CEO from IBM, Joe Guglielmi (apparently to the distaste of many of the Apple people).
- "In 1992, the earth shook: IBM and Apple clasped hands and pronounced themselves allies. From this union sprang Taligent, a small Cupertino, California, company that's now developing nothing less than a universal operating system."
- –Macworld, 1994
During its first year, IBM persuaded Taligent to replace its internally developed object-oriented microkernel, called Opus, with the microkernel that IBM was using as the base for IBM's Workplace OS. The change in underlying technology had both positive and negative aspects. On the positive side, Pink would become a personality on top of the IBM Workplace OS. This would create easy migration paths between OS/2, AIX, Mac OS, and Pink by allowing any combination of operating system personalities to run simultaneously on a single computer. On the negative side, this created issues over how to integrate Taligent's object-oriented device-driver model with Workplace OS's procedural device-driver model.
Development during the IBM-Apple collaboration and CommonPoint
Taligent spent much of its first two years developing their operating system (sometimes referred to as TalOS) and simultaneously trying to find a market for it. They started a large project surveying potential customers, only to find little interest in a new OS. It is a point of controversy whether the lack of interest was real or the survey fell prey to question-framing problems and political issues with investors. If asked the question "Do you want a new OS?", there were few who would say yes. The survey did, however, show there was sufficient support for the benefits TalOS would bring.
Influenced by the results of the survey effort, Taligent changed its focus from creating an object-oriented operating system, to creating an object-oriented programming environment that would run on any modern operating system. It was thought this approach would preserve much of the benefit of the higher levels of the system while freeing Taligent from the OS wars. Because of the portability and modularity provided by the C++/object-oriented approach, this change in direction, while not trivial, was relatively easy to implement. The result was TalAE (Taligent Application Environment), later known as CommonPoint. TalAE consisted of more than a hundred object-oriented frameworks and nearly two thousand classes and was heavily pattern-oriented. It ran on top of AIX, HP-UX, OS/2, Windows NT, and a new Apple OS kernel called NuKernel. This made TalAE appear comparable to OpenStep conceptually.
At this time (in 1994), Taligent marketing types also began mentioning the possibility of TalDE (Taligent Development Environment) intended to be just as multiplatform as TalAE. Also at this time, IBM marketing types floated the idea of putting a TalOS "personality module" into IBM WPOS (Workplace OS) – a Mach kernel-based OS for the PowerPC intended as a successor to OS/2. However, both suffered the effects of corporate immune response, as teams in both companies saw Taligent as a potential threat to their own systems.
- "It's reminiscent of a classic Greek tragedy. A child is born, destined to kill its father and commit even more unspeakable acts against its mother. The parents love their child and are unwilling to kill it, so they imprison it in a secret dungeon. Despite its mistreatment, the child grows stronger, even more intent on committing its destined crimes."
The combination of C++ and IBM and Apple's names on the project suggested that it might prove to be more successful than OpenStep. Early in 1994, Hewlett-Packard became a Taligent partner as well, which was surprising given that HP decided in the same year to produce OpenStep on their platforms. Several existing OpenStep "end customers" stated they would move to Taligent as soon as it was ready. The first versions of CommonPoint shipped for AIX and OS/2 in mid-1995, but were met with a lukewarm response in terms of sales.
By 1995, Apple still didn't have an OS capable of running CommonPoint, and while work continued on the fabled Copland (which was designed to run CommonPoint), it was fairly clear to all involved that Apple had lost all interest in Taligent. Financial concerns at HP also caused their interest to wane. In early 1995 Guglielmi left Taligent for Motorola, and founding board member Dick Guarino became the new CEO. Guarino, though also from IBM, started Taligent on a new course as an object technology supplier in an effort to remain an independent company.
End of Taligent
In the fall of 1995, Guarino died of a heart attack while jogging, and that loss spelled the end of Taligent as a joint venture. It was decided that Taligent would become a wholly owned subsidiary of IBM, focusing on developing technology and leaving the marketing to IBM.
IBM used parts of CommonPoint to create the Open Class class libraries for VisualAge for C++. IBM spawned an open source project called International Components for Unicode from part of this effort. Taligent also created a set of Java- and JavaBeans-based development tools called WebRunner, a groupware product based on Lotus Notes called Places for Project Teams, and licensed various technologies to Sun which are today part of Java, as well as to Oracle Corporation and Netscape. HP released the Taligent C++ compiler technology (known within Taligent as "CompTech") as its "ANSI C++" compiler, aCC. HP also released some graphics libraries that had been developed at Taligent.
In 1997, when Bill Gates was asked what trend or development over the past 20 years had really caught him by surprise, his reply was:
In January 1998, after two years as a wholly owned subsidiary, Taligent was dissolved and the remaining engineering team members became IBM employees.
- "Apple surrenders the Pink (to Microsoft)", The Register, 3 October 2008
- Cotter 1995, A brief history of Taligent.
- Davis, Fred (May 1993). "Surrender the Pink!". WIRED Magazine. Condé Nast. Retrieved 28 June 2011.
- "Frenemies: A Brief History of Apple and IBM Partnerships".
- Potel, Mike (1996). MVP: Model-View-Presenter: The Taligent Programming Model for C++ and Java. Retrieved on 2009-01-18 from http://www.wildcrest.com/Potel/Portfolio/mvp.pdf.
- InfoWorld. March 3, 1997. Missing or empty
- Cotter, Sean (1995). Inside Taligent Technology. Addison-Wesley.
- Official website (last available archive) at the Wayback Machine (archived December 11, 1997)
- Taligent developer documentation
- Inside Taligent Technology (sample of book)
- Taligent patents at USPTO