Apple needs chip czar for Son of Newton
Back to the future
An intriguing vacancy at Apple Computer suggests the company is looking to regain control of its long-term technical destiny.
The position paves the way for Apple's re-entry into the handheld business, which will gladden the hearts of Newton devotees.
The job description calls for an industry chieftain with 10 years experience to manage "the development of new high-performance microprocessors for future Apple computer products".
In other words, Apple is dipping its toe into the chip design business, an ambitious advance on its position today, which sees the company heavily dependent on key component suppliers. Or at least, one big component supplier - Motorola.
But the following qualification specified in the ad suggests that Apple is more interested in striking out into new markets than in reshuffling its PC component supply chain.
The applicant "will manage processor co-development and production start-up across processor vendors (our emphasis), Apple VLSI chipset design, and Apple computer system design teams".
Which of course is how today's handheld and cellular phone manufacturers work. Phone and PDA makers typically license intellectual property rights to a core - ARM, MIPS and Parthus are your specialists here - then embroider it with the parts such as I/O, or DSP, or integrated communication stacks that can produce a working piece of kit. The emphasis on multiple processor vendors here is a dead giveaway.
So herald, the Son of Newton?
Naturally Apple doesn't say, but we can reasonably infer that it is taking baby steps to regain a foothold in a category it entered with the Newton. It also crashed out of this category in spectacular fashion with the Newton. The Newton's lasting legacy was the acronym 'PDA', although the bulk and (initially, at least) lack of utility of the Newton made the handheld market a no-go area for investors and developers for a good two years.
But times have changed, and basic economics suggest that standalone PDAs (without some kind of in-built communications to the Internet or to pervasive access networks such as GPRS or 3G) will soon be extinct. Or if not extinct, will have become commoditised to the extent that they're sold for what they're reallyworth, which is the same as today's databank handhelds - just a few bucks.
In fact the Apple vacancy is a template for a technical lead on a communicator (as these kinds of devices are called) project. Only it doesn't say so.
So why exactly why would Apple need to appoint a chip czar to design its own communicator? Diligent Register readers will note that there's no shortage of intermediaries happy to carry out such work on its behalf.
For example, Symbian licensee Sanyo last fall demoed a palm-size communicator within two months of inking the software deal, thanks to integration work from Texas Instruments.
And out of Folsom, Chipzilla is preparing the very same 'building' blocks for communicator OEMs based on XScale (nee StrongARM) cores. So if Apple really wanted to get into the smartphone/communicator business, it could do so very quickly, with very little R&D expense.
The answer is probably that Apple wants to tune its communicator to target its core businesses (education), and leverage long-nurtured investments such as Firewire and MPEG-4/QuickTime. No-one has yet developed an entirely appropriate platform - Apple thinks that there is a breach to be filled, and that it should be doing the filling.
This is a reasonable assumption, and Psion has followed a similar path with successive devices including the Series 5 (where it partnered with ARM and Cirrus Logic) and the Quartz project (where it's partnered with Parthus). ®
We were going to dwell on the multiple ironies here. Should Apple plump for an ARM core - that's the basic CPU in the majority of cellular phones in use today - it would be returning to the company it helped found.
Apple's investment in ARM has significantly bolstered its results in recent years, although as we note here the pot is nearly dry. And we could mention that Apple in the 80s poured millions into its own from-scratch chip, Aquarius, which was supposed to produce a rival to Sun's SPARC, but really gave engineers the chance to play with a genuine Cray. Any Aquarius veterans out there? Tell us what we've missed.