Random Rants, by Thomas Andrew Olson

Thursday, June 09, 2005

Can MacIntel increase market demand?

Everyone else in the blogosphere has been waxing poetic on the virtues of Apple's Intel switch in terms of Apple's present business models, and the juggling act required to keep everyone happy going forward. There is little chatter, however, about the potential for growing actual market share - so I'll go with that topic.

On this issue, I get the sinking feeling in my gut whereby "Steve giveth, and Steve taketh away." Yes, Apple is moving to Intel processors, and from a long term tech roadmap position, that is justified. But Apple will not enable complete porting of OS X to run on just any Intel processor, specifically in legacy PCs currently running Windows. So if you want to run OS X, you'll still need to buy a Mac, and Bill Gates gets to continue sleeping like the proverbial baby at night. A Macworld FAQ on the topic seems to cover all the salient points. Andy Ihnatko has checked in, claiming that porting OS X to run on a Dell, for example, would be "suicide" for the company. He is probably correct, at least for now, in that Apple's present business model only thrives when you buy that $2000 iMac with OS X pre-installed, as opposed to a $129 Tiger CD for your current Dell. (However, he also suggests that the idea of running legacy Windows apps - natively - on a MacIntel is not farfetched. Others have discussed this as well.)

So...is it time for a new business model? While the present model keeps Apple solvent and innovative at the same pace, it does absolutely nothing to increase overall market share, which is the one thing that will sustain you for the long haul. I've always believed that the history of this business was driven by early adopters with big budgets, i.e., large Fortune 1000 corporations - the consumer desktop market would not have grown so quickly had not those consumers been already "trained" on office PCs. This is one of the important ways in which Microsoft won the OS wars. Ergo, any significant shift in market share for Apple will depend on a two-front battle, in both consumer and corporate desktops.

Personally, I'm disappointed by Apple's present strategy. I have long dreamt for the day when I could migrate an entire company's PC line to MacOS, while offering significant cost savings by not having to change much hardware up front, except for those machines already at the end of productive life. It would seem that right now, the timing is near perfect for exactly that sort of move - MSFT has been on the ropes for a couple of years now over security issues, issues that cost consumers and corporations billions annually in lost productivity. If you could give people a "switch" option that didn't involve massive hardware purchases, I'm willing to bet they would take it. Working in this business for 21 years, I'm here to tell you there is a lot of frustration and anger out there in the trenches, more than any marketing research pollster will ever tell you, without charging you $3,600. But it's a great key to new sales, and major shifts in technology.

However, there are many sides to this argument, pro and con.

If Mac OS X was available for any Pentium, not just MacIntels, it would potentially allow Apple to make incredible gains in OS market share virtually overnight. But that would be done at a price of other problems, and a potential loss in hardware sales, which currently drives the income stream for future development. Apple has never been a "software" or a "hardware" company. It is a technology company. By "building the whole widget", they have managed to avoid the unsupported third-party hardware glitch issues that have plagued the Wintel world forever. And who wants to be like them? Your only sales pitch at that point is "security", and unless there's another superworm in the pipeline that will paralyse the Fortune 1000 for a week, you'll need to offer more than that to convince them to make the move. It's got to be about costs, and ease of transition.

It is also doubtful that an OS X build would really work on just any Pentium. Hence there would be a need for huge levels of customizing, which could potentially send development and rampup costs (and timelines) through the roof - not to mention the loss of goodwill if a large deployment went badly. Better to customize OS X on the new Pentium M currently in the pipeline, ship 2007 MacIntels with those chips, to stay ahead of the pack, wait a few more years for those processors to filter into the rest of the installed PC base, then revisit the "cloning" issue, in, say, 2010. Meantime, the current profit model remains sound for the foreseeable future, and buys time to work out the bumps in the next "transition", if Apple wants one. (But I have another idea, and you won't have to wait until 2010 to make it happen.)

Of course, there are some who suggest that Apple is damned if they do, damned if they don't. If they don't open up to potential "cloning" by allowing the OS to run on an older Dell, they lose. They could also lose market share today under the present model, as significant numbers may hold off buying new Macs until the MacIntels arrive, under the false perception that the G5 is "obsolete".

In addition, by "opening up" OS X to run on legacy Pentiums, Apple would have to re-engage the rough and tumble OS war with Microsoft - a marketing war MSFT clearly won, and won't want to re-engage, unless forced to. It will then become a formidable enemy (again), with an almost limitless bank account. To be fair, it will be the first time such a battle would be engaged on a level field, i.e., the same processor platform. I would expect Apple to do well in that regard, as OS X is still the better OS. But Apple would take a lot of lumps in the process that might hurt them badly. MSFT could play hardball by discontinuing the Mac product line altogether - no more Office. (Actually, all they would have to do is threaten corporate America with such a move, and any notions of switching OS'es on the fly will largely disappear - the "iron fist in the velvet glove" approach.) So for this to happen, there would have to be a lot of "what ifs" to deal with, with the attendant Plans A, B, C, and D fully developed. The corporate demand would have to be overwhelming and rock solid to be worth the risk.

So there needs to be a new strategy, one that can allow Apple to sell its own wares well, near term, yet still carve out new inroads in the corporate cubicle farms to significantly increase market share, long term, which would, in turn, feed back into the consumer market. So it would be a "semi-dissociation" of the OS from the box. But I firmly believe a form of "cloning" will be necessary, in the near term, as large corporate IT departments are facing a lot of fiscal challenges these days, and will be loath to throw out thousands of perfectly good PCs, viruses or no. Your goal here is to get them onboard now - then when the normal PC replacement cycle comes around again, a couple of years later, they are primed to buy "the whole widget" from you - thousands at a time. What you lose in immediate box sales you gain in spades down the road.

My strategy for large corporations would be this: Offer a "custom", one-size-fits-all OS X build on a CD for legacy, but late-model installed-base Pentiums, but at a higher price, to cover the additional development costs. This build would "expire" in three years. For older boxes at the end of the life cycle, pitch new MacIntels (a little of something is better than a lot of nothing). Sell them on support, "migration services", training the staff, training the trainers, project management, the full monty. As their Windows licenses will still be good, there will be no issues with running legacy Winapps in a separate window, as has already been projected for the new MacIntels. This gets you in the door, without detracting from new box sales. I would assume that native MacIntels will integrate a lot better than legacy Dells with OS X loads, but the differences will be balanced to encourage full hardware migration at a later time, while still providing the enjoyment of OS X to recovering Windows users immediately. This fits in with my theory of "as above, so below". As corporate users get accustomed to the new system, they will be more open to giving it a try at home, especially as MacMini's get cheaper, and will run their legacy apps and peripherals, so the market penetration builds on itself.

I think the time is ripe to encourage corporate America to make the move away from Microsoft - the only resistors I see are Redmond itself, and IT managers whose entire careers have been built on solving the never-ending Windows problems. But that's a strategy issue for another column. If Apple really wants to regain acceptance in the larger workplace, and get its market share back into double digits, thus ensuring long-term viability, it needs a "mixed" deployment strategy around its Pentium builds, which must include some form of limited "cloning" in the near term.

Wednesday, June 08, 2005

Update from yesterday, re OS X on legacy Pentiums

I took the liberty of hyperbole concerning an Intel-ported version of OS X potentially running on a legacy Winblows Pentium pee-cee, and shouldn't Bill of Borg be crying in his beer right now.

Well...I spoke too soon. I should have known that logic wouldn't extend beyond Steve's immediate self interest. It won't be as easy as all that - if ever. I quote from a poster to the Thinksecret mailing list:

"There is not going to be an Intel version of OS X that you could install on a currently-existing Windows laptop. Period. Not going to happen. Ever. And anyone who tells you "it'll be hacked and work fine" knows very little about
computer architecture, and even less about how usable "hacked" Mac OS X is likely to be (got a slightly different FireWire chip to the one that Apple uses? Boom - no booting for j00, haxx0r!).

The only reason to buy a Windows machine today rather than a PowerPC Mac is the same as it's always been - if there's a specific application and/or hardware combination that Apple doesn't provide. For example, I wanted a
tablet. Apple doesn't make them. Hence, I'm typing this on a very pretty little (and very good) little Acer. Had Apple made a tablet, I'd have bought that instead in an instan[t]."

- Ian Betteridge

So Apple will protect its widget sales by insisting people buy "MacIntels". Blah. It doesn't mean that OS X builds for legacy PCs still isn't a great idea! And never underestimate the power of hackers. PearPC has been running the current Panther builds (albeit slowly) in emulation mode on Pentiums for some time now.

Apple is undergoing a resurgence in popularity - sales of the OS are going thru the roof. Given M$FT's vulnerability right now, it would be the perfect time to pitch the concept of "switching" desktop OS'es to corporate America - if they don't have to buy new hardware, it would be a much easier sell - and the Mac OS market share would be back in double digits. I'll be arguing more on this later. Comments?

Tuesday, June 07, 2005

Apple's Not in the Wilderness Anymore

This is something I always wanted: Being able to run Mac OS X on a Pentium.

Apparently, my moment arrived - and as usual, while I was out. Good thing I live for irony.

In his usual hyperbolic form, Steve Jobs, during his keynote at the WWDC Monday, formally announced that by 2007 all Apple products would ship with Pentium processors. This was a deal five years in the making. I had known for a long time that a Darwin build was running on a Pentium 4 somewhere in a basement at 1 Infinite Loop, but I felt it unlikely that they would port the whole farm over, given the 64-bit-code investment in the G5.

But this is an age of unlikely things, and while moving to Pentium a decade ago would have made little sense, it makes a lot of sense now, for a lot of reasons. Personally, I'm thrilled at the possibilities and opportunities.

Yes, the Mac purist techies will scream in rage at Jobs "sellout". But most users, frankly, could care less about the advantages of this processor over that one in the guts of their machines - they just want the fastest thing they can afford that will run the apps they want. It's all about momentum, supply chains, and consumer choice.

Apple has some great momentum right now, in the wake of iPods, iTunes, and MacMini's. It's share price has soared past its competitors, which (including Microsoft) have been fairly stagnant. Apple has proven itself time and again to be the most innovating force in the digital world. So there must be some compelling reasons for the company to shift hardware gears now. And there are many. Let me gloss over the ones I feel are most important:

The supply chain has always been a challenge for Apple - it seemed every time they switched to a newer PPC chip, new machines bought by eagerly anticipating users were almost instantly on back order. The G5 is screaming fast - I have a dual 2.0 Ghz at the office - but there is no new G5 laptop in the pipeline, due to excessive heat issues. Switching to Pentium gives Apple instant access to new laptop designs with cranked up horsepower, new tablet designs, and no shortages of anything again, ever. So Apple's machines will be priced even lower, and offer more raw power for the buck.

Even more importantly, Apple's overall "market share" will finally be dissociated from hardware sales. The "cloning" experiment a decade ago was too little too late, as it was tied to the "minority" processor, and Apple itself had to compete with its own clonemakers for PPC chips, so market share stagnated. Today, we have a company with a rock-solid, secure, 'nix-based OS, that will run on multiple processors, including the one which runs over 80% of the world's computers. It will compete for the first time on a truly level hardware field against the dominant OS, Windows. Microsoft has had that field to itself for a long time. (While Linux has made some inroads, it fails to excite users, is built mainly for high-end tinkerers, and has a paucity of application offerings.) Additionally, Microsoft is vulnerable right now, due to all the hidden costs (in the billions) to both consumers and corporate America of virus attacks, spyware, and other malicious problems. OS X-on-a-Pentium offers people a way out - change your OS without changing your hardware. Expect this to happen on the corporate server farm first, then watch it spread out to corporate desktops, and from there to consumer desktops.

Of course, for this to fly, it would have to be very easy for Mac app developers to recompile their code for Pentium chips, and that, apparently, is the case in the form of a new compiler/translator called Rosetta. All the more reason that this move be announced at the WorldWide Developers Conference, as opposed to MacWorld Expo - without the developers onboard, you're nowhere. And to a great extent, if the early returns mean anything, they are, indeed, onboard. (But of course when you have an announced two year lead time, it's easier to gain acceptance.)

Personally, I can crank up my consulting practice quite a bit by offering "migration services" from Windows to OS X on-a-Pentium, recompiling their key apps on the fly. If there are some "Winblows only" apps that they must have no matter what, the legacy system can be dual-boot. There will be no need for "virtual OS" emulators any longer. (Looks as though Connectix sold out to MSFT just in time!)

It's also good for the consumer in general. Microsoft has had its OEMs, like Dell and HP, over a barrel forever in terms of licensing agreements. Now the 'softies have to play nice, or Michael Dell punches up Bill Gates on his speed-dial and tells him the next two million Dimensions are shipping with Tiger pre-installed. Hell, he should do it anyway - Tiger is half the price of XP Pro. Low-end PC prices will continue to drop, and support calls along with them. Virus and spyware protection will be far less of a worry at all levels, increasing overall productivity.

The only ones to lose by this are those who have carved lucrative careers for themselves in Windows land, living parasitically off those who have become utterly dependent on them for solving the ubiquitous Windows problems - problems that invariably crop up on them again a day, week, or month later. Huge IT departments with huge power bases and budgets have been made in corporate America on the sole basis of Window's complexity and insecurity. Wait until the boardrooms get word that their lives can be simplified and budgets shrunk, long term, by simply moving their OS - no hardware change required. I foresee a lot of MSCEs in OS X classes before too long, if they want to keep their jobs. But then, as OS X needs less hands-on support than Windows, and therefore less staff, expect a lot of resistance from the low/middle-end techies at first.

For the first time in over a decade, this industry is going to get shaken up. New alliances will be forged, new opportunities discovered. And in the end, the consumer wins, big time. Apple will continue to make Mac boxes, but if they want, they can concentrate on the high-end Porsche and Mercedes versions, with the biggest margins, while still extending their reach to those who can only afford (and/or already have) Kias.

Even so, there would still be a market for a MacMini at a lower price - and they wouldn't be backordered any longer.

Wednesday, June 01, 2005

I expect more from my Legends

I think this story is pathetic.

It seems like the "First Man to Walk on the Moon" has lived in a cave all these years. He has actively shunned publicity, preferring to live in relative obscurity in his native Ohio, where he lectures at university. He's trotted out by NASA every five years for the obligatory "anniversary of the moon landing" speech. He came out of hiding a couple of years back, commenting on the Columbia disaster, and he's made a public comment or two concerning upcoming policy initiatives; but that's a piss poor public record for someone who did something that no one else in all of human history has done, or, for that matter, ever will do. Only one human being could be the first to walk on a planet other than the one that bore and nurtured us, and that man will forever be Neil Armstrong.

He could have done so much more to promote space to young people, be a spokesman for the alt.space community, encourage science and engineering education, kept the public engaged and excited about the promise of the future. "The First Man On the Moon" is a permanent door-opener. Instead, he let the "Second Man on the Moon", his partner Buzz Aldrin, carry his water all these years.

At the last International Space Development Conference, many keynote speakers prefaced their remarks with the expression "...when Buzz Aldrin and his associate set foot on the Moon..." . They said it that way to honor Buzz for his tireless contributions to the space community all these years, and as a slam at Neil for sitting on the sidelines.

But no, he'd rather sue his barber for selling his cut hair to a collector. Well, what did he expect? He hasn't contributed much else.

We complain about NASA's failed policies today, but the seeds of those policies were sown back in 1967, when the crew decisions were made for the Apollo flights. The "First Man on the Moon" should have been a damn good pilot, to be sure, but NASA should have also realized they would need someone dynamic and inspirational, for reasons of PR and posterity. Buzz Aldrin has proven, time and again, that he always had the whole package.

I never thought I'd see myself type this, but on 20 July 1969, the wrong man stepped out of the LEM first.

"Buzz Aldrin's associate" says it all.