Apple Computer: Software Only in Five Years
If you have a child or if you’re just very averse to pain (like myself) you are probably familiar with the two theories of band-aid removal. The first theory is that you do it all at once, just rip the thing and suffer a brief, but intense, wave of pain. The other theory (popular in my household) maintains that you remove the band-aid slowly. This results in smaller amounts of discomfort but the sting lasts much longer. Apple is obviously an adherent of the slow removal school of thought.
How can we be sure Apple prefers extended unpleasantness to quick shot of pain? The answer lies in their history. While Steve was busy spinning like a Chinese acrobat at the World Wide Developers Conference he was leaving out a few salient facts. Steve talked up two transitions: from 680x0 Motorola chips to PowerPC chips and from Classic to OS X. What he left out was the transition from Apple II to the Mac and the impact each transition had on market share. Each transition brutalized Apple’s market share generally cutting their piece of the computing pie in half. Apple’s market share will be similarly impacted with this move, after all is it a reasonable expectation to think that people will invest in a PowerPC based Mac with the foreknowledge that Intel Macs are on the horizon?
So the Mac piece of the computer market is going to suffer, and suffer horribly. The hope is that once the transition is over Apple may start selling machines once again and pick up just a little bit of the share Apple will undoubtedly lose. That is not impossible, as Mr. Jobs noted the Mac was currently making headway. Yet the time for the Mac’s resurgence (if ever) will be at least a few years down the road. Or, to sum it up in one tidy sentence, there is absolutely no near term upside.
Some people will reflexively argue that Apple had no choice. IBM wasn’t coming through, Freescale left Apple twisting in the wind and the roadmap looked much better on the Intel side of things. At this point every Mac user sincerely needs to send Apple a letter asking them to raise Steve Job’s salary so he can afford to take a course on cartography at the local JuCo. Steve obviously has trouble reading roadmaps, how often has Steve said the roadmap looked great from Motorola or IBM? Too many times for anyone to have any faith left in his abilities at prognosticating the future of chips. Unfortunately that is exactly what one expects from a CEO, particularly a purported visionary like Steve Jobs, the ability to make sound long-term decisions. Changing from classic to X and from PowerPC to Intel in less than a decade is asking a lot of developers and customers and does not belie the “genius” status held by Steve Jobs.
At this point it is simple (as it usually is with hindsight) to say that it would’ve been infinitely better to make the jump to OS X and Intel chips simultaneously. Sure it would’ve have been more painful during the change but it would’ve have been over by now. Apple sales would have started growing and the momentum could have been sustained. Those that argue that that move would have been technically impossible need only look at the arguments made against the move to Intel in the past few weeks.
When all is said and done this move is just another step on the path that leads to the end for Apple as a computer manufacturer. With the change in architecture the difference between Macs and Wintel boxes is, more or less from a consumer’s point of view, purely aesthetic. While people will opt for aesthetics all things being equal they tend not to buy computers purely for the looks (witness the cube debacle). The end of Apple as a manufacturer doesn’t necessarily mean the end of Apple as a company. Many pundits have long encouraged Apple to jettison the computer aspect of their business and concentrate on being the best software company on earth. In five years we’ll find out if that is one more transition Apple can make (oddly that transition will be the easiest one to date). Those who are fond of noting that Apple’s profits (those will be sparse for some time) come primarily from hardware sales should realize that the main reason that the hardware sells is because of the OS, not the other way around. Can a company be profitable solely on the sales of OS software and associated applications? Sure. Can Apple? That remains to be seen.
del.icio.us
Digg
Facebook
Slashdot
StumbleUpon
Comments
So what you’re basically saying is that because Macs will run on Intel processors, consumers won’t be able to tell the difference between a Dell running XP and an Macintosh running OS X?
I disagree.
The only thing that changed last night was the logotype on the chips. If anything, it will make life easier since Apple doesn’t have to worry about performance anymore.
I wasn’t into computers when the Mac came out and I don’t doubt that it was a difficult transition. It was sort of an Osborne effect with the Mac competing against the Lisa, the Apple III and the Apple II, but I doubt that we would have Apple Computers today if they hadn’t produced the Mac. The Lisa wasn’t going to compete with IBM (at least not with price and software) and the Apple III was a sinking ship before it was even shipped. Jobs didn’t mention it yesterday, but never the less, it was the most important transition to date. We won’t know for sure if this is a great thing or if it death, but we wouldn’t even be debating this if Steve hadn’t jumped in head-first a few times. He isn’t infallible, but I will choose to trust him again. This is nothing like the mid 90’s; have a bit more faith in the Man.
JDL
This is right on the money! Who in their right mind would purchase an Apple computer knowing the next great thing has an Intel chip inside.
Apple’s laptops are the best in the business. I have yet to see any PC makers of laptop that come close. I think that is where their real potential is. As for the large powermac G5s, well for almost $3000 you have lots of other choices for a desktop computer for a lot cheaper. I’m a pro-apple guy just like the next guy, but I think their powermacs are overpriced, compared to competition price for a similar large scale power-machine in the market. If anything, I see an overhaul or redesign of this big expensive powermacs. I mean even Jobs himself said the nice compact (and half the price of powermacs) iMac G5s are the hot seller.
ANYTHING BUT WINTEL. . . . . Steve Jobs and Michael Dell are selling out to the marketing of Intel-only monopoly. For Jobs to snub AMD is strange in that AMD chips are better designed and more powerful. Similar attributes to Apple. Or I should say, the old Apple. Jobs pocketing a lot of money on the side to place an Apple enemy inside.
Tsk tsk to the pessimist. What’s with all the doom and gloom? The transition to Intel is a necessary one, both in terms of costs and capabilities. Macs will be cheaper and more powerful - is this a bad thing? The OS will be consistent, making the transition completely transparent for all save hardware designers and software engineers (who would rather be working with the world’s most popular chips to begin with). And to say that no one’s going to want to buy PowerPC-based machines now that this move has been announced is just plain silly. It’s virtually the same difference as buying a G4 or a G5 machine: you’ll either buy what you can afford now, with the ever-present risk that your hardware won’t be able to run the next generation of software (true for both Macs & PCs), or you’ll wait for the more powerful (and traditionally more expensive) option - except this time that option won’t be as pricey as before. Cheer up, it’s just a chip.
“With the change in architecture the difference between Macs and Wintel boxes is, more or less from a consumer’s point of view, purely aesthetic.”
This is an idiotic thing to say. I don’t give a damn that my 12” Powerbook has a non-Intel, non-x86 processor. I like it because it is one of the best laptop designs on the market at the moment, with a fantastic OS. That’s what I use. The processor is irrelevant.
Oh, and you know what? The “architecture” of the Intel Macs is still very different from Wintel boxes. You may or may not have noticed that the BSD architecture is somewhat different to Windows…
Thanks for the note Allan. I, of course, was referring to the architecture off the machine not the architecture of the OS. Sticking with the machine I will note that Phil Schiller said that the new Macintels will be able to run Windows. Therefore one can easily conculde the architecture will be similiar. He did note that OS X will only (for now) run on Macintels. With these facts in mind one can only conclude that the architecture is essentially the same with some artificial stuff implemented so only a Macintel will run OS X.
There is a great danger here, too much compatibility can be deadly. You note that Apple once sold a PowerPC machine with a 486DX (Performa DOS). Turns out people were getting used to Windows and not coming back to the Mac. Unsurprisingly Apple didn’t sell that model very long. For another example I ask you recall OS/2 from IBM. By most accounts that was a superior OS and yet it isn’t being widely used today.
Yet I suspect OS X will not share the same fate. If adoption of OS X is speedy enough the incentive for Apple will be to jettison hardware sales with their high per unit marginal cost and substantial development cost in favor of pure software which has a substantial development cost but a near per unit marginal cost.
Gentlemen and Ladies:
Read what Mr. Jobs said carefully. Apple will release hardware based on the Intel platform that will run OS/X. He did not say the basic x86 platform will run OS/X. If an Apple-branded oomputer runs OS/X and another OS (Windows XP, Linux, and/or FreeBSD), why would the enterprise NOT buy the apple product?
As I think most people are coming to realize - the future is MacIntel - meaning, Mac OS running on Intel chips, exclusively on Apple designed hardware. No clones, no beige-box Mac, no bargain basement prices.
I think something that will be surprising in the next 2 years - is that the Mac faithfull will see only one change: that is faster clocked processors.
We already saw comparable system bus speeds and RAM speeds - just about the only comparable x86 to PowerPC metric.
I predict we will still see Apple-centric 8+ month hardware cycles, that have nothing to do with Intel releases. Business as usual.
There will also not be price cuts. Apple is now between a rock and hard place, their only garuntee to future chips is that Intel is a chip-only business. Unlike IBM who can kill product line at the drop of a hat.
This absolutely sounds like a partnership of necessity. One that only has marketshare problems for the future of Apple: due to consumer confusion, developer frustration, and expensive dual application support.
If it’s a faster, better, cheaper CPU, I don’t care if it’s made by Mattel.
What I an concerned about is transparent backward compatibility of future software. Will Photoshop 9 or 10, etc., work on my existing office full of G5s, or will they become obsolete if I want the latest version of software?
That’s what I want to know.
I’ve followed this story carefully, and one important point I think everyone is overlooking is this: the story Apple is currently telling is that they are dropping the PowerPC. In reality they are simply picking a specific time to exercise an option they’ve had in their back pocket for 5 years-- a processor-independent architecture. Right now they want to be buddies with Intel-- fine. In a few years they can be buddies with AMD. It makes perfect sense to go with the big name first, and then consider the AMD later. They can also keep coming out with PowerPC machines, should that line of processors be able to do something better than the Intel chips of the day. There’s nothing wrong with flexibility.
I am a Mac software developer-- in my case the effort to convert my apps to universal binaries will be trivial. While I realize that this will not be the case for all developers, I do think that once the Mac development community hits its stride developing universal binaries as a matter of course, then it won’t be much additional effort to produce them from then on.
As a guy who has programmed assembly language on every Apple processor from the 6502 forward, I can say with complete confidence that the processor is only one of many things that make a computer great.
So, to summarize:
1) The user experience will be the same or better. Even most old software will continue to run thanks to Rosetta.
2) Apple has more degrees of freedom when it comes to picking processors for various products.
3) Inconvenience is minimal for most developers.
4) Apple will continue to control the hardware and the OS that runs on it-- a chief competitive advantage.
And this is bad… how?
Oh, and to respond to Xooberant’s comment above: since producing universal binaries is a simple thing for (most) developers to do, I would be very surprised if new versions of most applications weren’t backwards-compatible until your G5s turn to dust.
This would apply doubly to workhorse business applications, and triply to applications that are also already cross-platform, such as Photoshop.
The issue here is not about the future roadmap. The issue is about the disenfrancised customers and loss of marketshare in the near term.
Example: Me! I spent close to $3000 to get a top of the line dual G5 which I expected to only get faster as developers optimized for G5, dual processors, 64 bit, etc. Now just over a year later, Steve drops this bombshell. Suddenly my 1 year old top of the line system is obselete. No further optimizations, and probably no forward compatibility with the new intel compiled apps.
Am I going to buy another piece of mac hardware? Hell no. I’m going to find any possible way to buy a Dell and run OS X on that. And mark my words, someone will find a way. Do I need my mac in a shiny metal box? No, not really. The point you make about the hardware being irrelivent is correct on that point. I could care less about a fancy box as long as it runs the os I want (which will still be OS X).
And in the near term I am taking every opportunity to discourage every potential mac owner I know from buying a mac until the transition is made, hence the drop in market share.
This article is one of the few that isn’t influienced by the reality distortion field. He’s right.
If an Apple-branded oomputer runs OS/X and another OS (Windows XP, Linux, and/or FreeBSD), why would the enterprise NOT buy the apple product?
Simple. Cost. The idea that the Intel chip will make Macs cheaper is largely unfounded, IMO. Macs cost what they do because Jobs believes philosophically in charging a premium for quality products. And IIRC, IBM charges less for the PPC chip than Intel is going to charge for theirs.