Mactel - The Unholy Alliance is Born

Like many of the Mac zealotry, my good friend, Jeff Porten, is reeling (and then reeling again) over Apple's recent announcement to use Intel chips in all of their machines by 2007.  His words are thought provoking, as is evidenced by what you're currently reading.  So much so, that I'm going to use the basic structure of what he wrote to offer my own thoughts on what this means to the technology world.

Jeff's First Point:  Intel Chips Don't Perform as Well as (Apple's) IBM Chips

I must admit, the best part of all of this will be watching studies like this one fall to pieces.  Anyone who has used any OS for any length of time, but especially Windows, knows that the speed with which a machine performs a given task is dependent on a thousand factors; the chip being one of the least important.  On my current (Windows) machine, editing a Photoshop image can be anywhere from near-instantaneous (fresh boot, nothing else running, empty cache, lots of virtual memory) to mind-numbingly slow (just finished editing a monster Excel workbook, 10MB Powerpoint file cut & paste into the clipboard and never removed, just visited a website/ran an app that leaks virtual memory).  When everything's running on the same box, I predict that these supposed "side-by-side" comparisons will quietly fade away.

Jeff's Second Point:  Macs are Macs Because of the OS, not the CPU

I would tend to agree with this statement, with the huge caveat that a lot of people have a lot of work to do to make this true.  The OS makes frequent calls to the CPU, so if the CPU changes, somebody has got to modify that OS so that the same call returns the same result every single time.  If they don't, applications are going to start breaking left and right.

It's very reminiscent of the Y2K era.  Literally millions of hours of work went into remediating software to handle the 19xx date issue, and when nothing broke on January 1, 2000, most of the media reported (and most of the people believed) that the whole thing was overblown hype, as if those millions of hours of work hadn't actually occurred (or accomplished anything).

If the people porting OS X to Intel do their jobs properly, OS X users will not be affected.  That's a mighty big "if," though.  To summarize, a direct response to Jeff's words:

The CPU is safely protected beneath about a dozen layers of abstraction and unless you're writing software, for most purposes you shouldn't need to care about this one way or the other. When you board an airplane, do you ask yourself whether the engines are made by Pratt & Whitney? Or is your sole concern that you don't take a sudden unscheduled stop in Kansas?

Jeff:  No, I don't care who made the engine, but I most certainly care that the cockpit controls communicate with the engine properly.  If not, my unscheduled stop in Kansas might not be at the airport.

Jeff's Third Point:  Will You be Able to Install Windows on Apple Hardware?

I must admit, when I first read the news, this is what I thought Apple was driving at. 

Large corporations tend to develop standard software architectures, for the simple reason that having a standard improves support, reduces errors, and increases overall stability in the environment.  They are usually less concerned with the hardware standard.  Typically, the brand of hardware sitting on a user's desk is a function of the company's depreciation schedules and the bulk discount they were able to obtain from the vendor, more than it is about how well the hardware performs.  This is because desktop hardware has become fairly commoditized.  Sure, there are horror stories about Dell, HP/Compaq, Gateway, and the others, but basically, any machine that can run the standard software architecture is fairly easily rolled into the environment if a deal is struck.

This could be a big boon for Apple.  If their new Intel-based machines are viewed as equivalent to Dells, Compaqs, etc., they might see a huge up-tick in hardware sales, especially where "sexy" hardware is of value (think customer-facing sales stations, retail kiosks, senior executive offices, or even trading floors).  Apple could enter a commodity marketplace and steal market share by providing that intangible "wow factor" that their hardware has always possessed to a much wider audience than they do today.

Of course, this all involves the major assumption that the new Macs will run on one of the existing Intel chips.  It's entirely possible that Intel designs and builds a completely separate chip for Apple, and that the two CPU's will be no more similar than the existing IBM and Intel chips, even though they're made by the same company.  People seem to be assuming that this isn't the case (and my hypothesis about this being a pure hardware play for Apple would suggest it isn't), but I haven't found anything written to confirm or refute it.  If you have, drop me a line.

Jeff  takes a different tack to the question, pondering whether a Mac running Windows will dual-boot, and how this may act as the fruit of knowledge for all those poor Windows users out there, who toil away endlessly with an inferior OS out of sheer ignorance that anything better exists in the world.  To this point, I will only say that not every technology decision taken in the world revolves around how to increase usage of Macintosh computers, and that any corporate IT department that has standardized on Windows would likely either disable dual-boot functionality (as they disable most things that violate their standard), or would look at that functionality as a reason to avoid the hardware altogether.

Jeff's Fourth Point:  Will You be Able to Install OS X on Intel Hardware?

Given the context of the previous point, this question basically boils down to whether this could be viewed as a software play for Apple.  In other words, will some of the 97.7% (or is it 84%?  So confusing...) of desktop users out there that don't use OS X today be more likely to try it if they don't have to shell out a couple of grand for a shiny, new Macintosh?  I think the answer to this question is definitely yes, but I wonder what that experience will be like.  Given the superior graphical quality of OS X today, I always assumed the hardware it was running on had a lot to do with it.  If that's true, then Windows will undoubtedly look better on the Windows-running Macs described above, but OS X might also look significantly worse when running on a PC.  Again, I don't know here, but I'm happy to hear from those who do.

Again, Jeff takes the "What does this mean for Macs" view, and says thus:

Turns out, for 90% of business functions (which means, let's face it, Microsoft Office and Outlook), there's Mac software that does the same job or better. What forces many people to stick to Windows OS is that single application that doesn't exist on Mac; granted, there are 100,000 such applications, but there are maybe a dozen prime suspects, and it's very rare for a single user to run more than one or two.

Here, I'd refer him to my essay on the Microsoft monopoly and why it persists:  "Windows is not number one because of its quality; it's number one because of its ubiquity."  The Mac software that imitates MSOffice and MSOutlook doesn't do the same job or better, because it's not used by 97.7% of the people one communicates with.  It's this network effect that forces many people (and corporate IT departments) to stick to the Windows OS, not the "single application that doesn't exist on Mac."

If software interoperability encourages more people to try OS X, they'll switch because of its strengths (look and feel, usability, superior audio/video/graphical tools, short learning curve, "wow factor"), not because it can reasonably plug its weaknesses.

Jeff's Fifth Point:  Upgrade Before the Change

In his second article, Jeff recommends that people upgrade their hardware before the switch occurs, and gives two reasons for it:  1) the new hardware will be Version 1.0, and will likely be buggy, and 2) custom software you've had written might not work on the new machines.

I think both of these points are valid, although if the OS port is done properly, then the custom software should work just as quickly as the packaged stuff.  If each individual app has to be remediated (i.e., the port isn't a pure emulation), then this is going to be a long haul for everyone, including the package vendors (some of whom might decide to write-off the relatively small Mac market share, rather than incur the costs of an additional development cycle).

The other point I'd add here is one of vendor support.  While not buying the first version of anything is sound advice, so is not buying the last version of anything.  Let's say you run that business that Jeff describes - the one that paid to build the app from scratch.  You just bought a new cadre of machines, while you can still get the IBM chips on which your custom app has been tested.  Now, you need to enhance that app, and you require technical assistance from Apple.  At some point, you're going to be told by a service rep that your problem lies in your outmoded hardware, or that he can't provide advice because he has not been trained on the old machines.  Now, if Apple knows what's good for it (and it often does), this day will be far in the future.  Still, it's going to happen sooner than the equivalent message regarding the Intel chips, so you need to consider how much of a sunset you want to build into your vendor support model.

Jeff's Sixth Point:  Windows Users/Consultants are Idiots until Proven Otherwise

As a "Windows guy," I'd be remiss if I didn't mention Jeff's admitted prejudice in this area:

I've been consulting for twelve years and . . . I've never met an incompetent Macintosh consultant, although of course there are variations in skill. . . . When I meet a Windows consultant, a web designer, or a PHP programmer, I assume he's an idiot until I see his work.

Well, I consulted for twelve years too.  I've met plenty of idiots, most of whom specialize in Microsoft technologies.  And Jeff is clearly one of the best in the business, and his specialty is Mac.  I second his admonition:  HIRE HIM NOW.

But Jeff - give the rest of us a break, huh?  If the world seems that black and white to you, I'd suggest you haven't met enough Windows consultants or enough Mac consultants.  There are plenty of idiots out there - enough for everyone to share...

Final Thoughts:  Somewhere, a Place for Us

When all is said and done, the biggest effect of this news might be the softening of the clear battle lines between Mac and Windows users.  Right now, the whole thing has a West Side Story feel to it - rival gangs fighting turf wars that really don't matter at all, except to  the guys doing the fighting, of course.

Maybe Intel and Steve Jobs will be the Tony and Maria of desktop computing.  Maybe by working together, they'll show their respective users that the other side isn't the enemy all the time.  And maybe this time, the whole thing can resolve itself without someone getting shot.

[This essay is part of The Red and the Blue discussion: Apple Switches to Intel.]