Pure objects must wait for mid-decade, but prototype Pink will be available soon
The second joint venture is charged with the development of a pure object-oriented software environment, based on Apple’s on-going Pink development work. With from 300 to 400 people, Taligent (derived from talisman – an object endowed with magic powers) is aiming to provide a product for mid-decade, but with a million lines or so already completed, software developers (under non-disclosure) are promised their first look by the end of the year. IBM is providing its system object model work and results from the Patriot Partners joint venture. While technologies from Pink will be incorporated into the PowerOpen Unix environment as they become available, and into 32-bit OS/2, John Sculley insisted that although object technology layered over existing operating systems would provide real benefits, the real goal was an environment where networking, communications, file systems and device drivers are freed to become objects themeselves. The new products would work in parallel to existing product lines, and layered implementations would ease the migration the pure object environment. The environment will also run on multiple hardware environment, including Intel and Motorola-based systems.
Despite the alliances, IBM and Apple will still compete, insists Kuehler
What happens to existing Apple and IBM products? The new software associated with PowerPC is tied to the chip architecture and will not appear on the current Motorola range of Macintoshes, but then again, A/UX has had a fairly limited circulation on Macs to date anyway. Apple hopes that now the alliance is in place and growth path set out, customers will be encouraged rather than discouraged from buying existing Macintosh products. This is the second phase of our mission to bring Macintoshes into the mainstream, said Sculley, who claimed that sales of the Mac over the last year had accounted for the majority of the industry’s growth. And Michael Spindler said that Apple’s networking arrangements with DEC would continue, with systems integration announcements due soon. As for IBM, Jack Kuehler insisted that the announcements would not change our plans for OS/2 one iota. He insisted that the Macintosh interface would remain entirely separate from OS/2, professed that the OS/2 2.0 Workplace Shell was rumoured to be coming soon, and said that despite the co-operation, IBM and Apple would continue to compete strongly to sell their individual product lines.
Intel, Lotus, Novell and Borland heap on the praise
With just over an hour’s satellite time used up on the announcement and signing ceremony, IBM and Apple found they still had some pennies left in the meter, and so they wheeled on some industry executives that happened to be in the audience to impart their opinions. The first to the rostrum, Ron Whittier of Intel Corp’s software products division, looked as if he’d rather be just anywhere else, and his halted, rather strangulated tones suggested that he was still attached to the electrodes. His enthusiastic endorsememt of the multi-media and object-oriented software efforts was rather lost on an audience amazed to see any representative from Intel – which figured not at all in the new announcements – within a mile of the building. Who next, we wondered, Bill Gates? But no, it was Jim Manzi of Lotus Development Corp, and Novell Inc’s Ray Noorda, who spent much of their time reminiscing about a historic baseball line-up consisting of Ted Williams and Joe Di Maggio in the same batting order (the old one two – IBM and Apple, geddit?). Finally Phillipe Kahn of Borland International Inc came on, stating that object-oriented software would be to software what the invention of the microprocessor was to hardware. Fair enough, but he then went on to stun the watching UK press corps at London’s Hyde Park Hotel by saying that some were not yet convinced. But then, there’s still a Flat Earth Society, many of the members in England, it’s kind of eccentric…. Asked later why Mi
crosoft Corp hadn’t been invited, Sculley said he hoped Microsoft would find the new platforms exciting ones to develop for, adding rather limply that Microsoft was a great applications developer.
Content from our partners
Motorola – the king is dead, long live the king…
Motorola Inc’s Jim Norling, president of Semiconductor Products, made the company’s position on the PowerPC RISC microprocessor abundently clear when he stood up to make his pitch between John Sculley and Jack Kuehler. It is clear that PowerPC is the platform for high-performance open computing, he said. Quite understandable, you might think, but also a damning statement from someone who, six months before, would have said exactly the same thing about Motorola Inc’s own 88000 RISC chip. But, as company officials explained afterwards, Motorola is primarily a semiconductor company, and has a need to keep its giant and hugely expensive 0.5 and 0.6 micron fabrication facility, conveniently located close to IBM’s own Unix operations in Austin, Texas filled up with work. So far the 88000, despite claiming some useful niche markets such as telecommunications, high-performance graphics and multi-processing servers, has looked an unlikely bet to provide that work in sufficient volumes. Motorola insists it will continue to develop and support the chip: the new 88110 and 88120 are due to be launched at the Microprocessor Forum in early November, and Motorola says that having its own chip will enable it to go off in its own directions, as opposed to following the requirements of IBM and Apple with PowerPC. As for 88000 volumes, the deal Motorola struck with Ford Motor Co (currently Intel Corp’s largest customer) back in June (CI No 1,693) to develop an automotive-oriented microcontoller version of the 88000 could involve up to 6m chips a year, although not until around 1995. This, of course, does nothing to secure its future use as a computer systems CPU, but then the point of RISC was meant to be that as an intrinsically simple plant, development of new generations should be much less costly than keeping complex instruction set microcessor families up to speed. – John Abbot
