Jan 20, 2005 5:00 AM PT
One of the oddest facts about the Wintel versus Unix (including Linux and the Mac) debate is that the Wintel proponents practically brag about never having actually used Unix while the most committed Mac/Unix advocates have generally used both. If in reality the Unix products really are better, faster and cheaper, why is it that we're always on the defensive, having to justify our choices against a Wintel default?
Review, for example, the information collected under the Macintosh Justification heading by Macintouch.com. There are seven subsections, each containing a number of success stories, purchase and use rationales, and more or less personal analyses of the costs and consequences that go with a Mac decision.
Read this stuff quickly and you'll see that most of the contributers obviously have experience with both Wintel and the Mac but prefer the latter because Wintel consistently costs more and achieves less.
Look more carefully, however, and you'll see that the underlying reason in almost every case where the Mac lost out to Wintel doesn't have anything to do with rational arguments based on cost, performance or functionality. Instead, Wintel proponents are shown as consistently fudging such arguments as rationales for decisions already made and getting non-IT executives to justify signing off on those decisions mainly by overweighting the argument that Apple's market share has long since entered the insignificance of single digits.
In effect, Wingots seem to shout "We're winning" loud enough to convince the people around them that it's true, and indeed you could fill a good-size book shelf simply by quoting analysts and commentators who've publicly written Apple off over the years. However, almost all of the older ones also enthusiastically welcomed OS/2 as the anti-Unix, went on to join their younger colleagues in heralding the Itanium as the anti-RISC, and were united across generations in their shared enthusiasm for the security and performance benefits offered by each new Wintel generation.
Revenue Dollars Measured
It's true, of course, that Apple's share of new desktop sales runs only a bit over 3 percent right now, but the conclusion these commentators draw from this is wrong because the 3 percent is a measure of revenue dollars to PC sellers, not hours of usage by customers.
In reality, Apple's declining relative market share measured in dollars has been due more to the expense of Wintel product churn than to a fall-off of interest among Mac users. Over the longer term, Apple's unit sales have consistently increased; what caused the decline in Apple's annual share of market dollars has been growth in revenue to the PC sellers.
In other words, it's Wintel's rapid upgrade cycle that's been getting progressively more and more out of line with norms for industrial or retail electronics products, and therefore not falling interest in the Mac, that's behind the numbers. Think about this for a minute: If PCs remained usable as long as Macs do, industrywide total revenues (aka customer costs) would be nearly two-thirds lower.
The other argument you see people make all the time is that consultants and IT staff make money on failure, not success, and are therefore motivated to push Wintel. I think that's right, but insufficient as an explanation. After all, it's ultimately the non-technical customer who decides and all the MCSEs in the world couldn't keep a GM or KPMG using Windows if end users in those organizations didn't go along. So the question really is this: What keeps those people from actively investigating Mac OS X and other Unix variants like Linux on x86 or Solaris on SPARC?
Part of the answer really is the shouting coming from Microsoft and its supporters. Consider my deafened friend Dr. Bernie (not his real name). Bernie is your all-around good guy. He's very bright, deeply committed to his family first and his cultural values second, and fundamentally as honest as anyone I've met. On the other hand, he's got a crippling blind spot: a passionate loyalty to Microsoft that no amount of either rational argument or economic coercion is likely to break.
Until the NT bubble came along, Bernie had been doing pretty well for himself -- at the top of his profession and a solid middle class property owner with no mortgage and a well-stocked retirement fund. Unfortunately, his enthusiasm for all things Microsoft drove him to buy into bubble companies, and now he's working himself into an early grave trying to recoup before age catches up with him.
As part of that, he's exhausting his Rolodex to get outside projects while antagonizing his biggest supporters because he never has time to focus fully on anything -- in effect fooling himself by mimicking Microsoft in his ability to sell the next project in time to cover up shoddy execution on the current one.
The last time I was in his office, we were supposed to be discussing a price quotation he'd obtained on a new server for a course-ware project, but got sidetracked while he rebooted his latest laptop -- it hadn't recovered from a spam call that came in while he was showing me that he could now get free long distance calls.
In preparation, I'd done nothing more than print out a line from an e-mail dated a year or so earlier -- when the school's Blackboard server ran Linux:
Please be advised that the Blackboard server will off-line for several hours on October 26 for processor and memory upgrades.
In contrast, the present machine -- which runs Windows 2003/XP Server and some application written (I think) in BASIC that only works with IE and a Windows client -- happened to be accessible during our meeting, but even Bernie admitted that it hasn't exactly been setting records for stability.
That's due, he said, to the incompetence of the people running it -- and he wasn't in the least appreciative when I pointed out that he had previously argued that the Linux package should be replaced with a more expensive Windows solution mainly because it's so much easier to find Windows support expertise.
As part of spreading Microsoft wherever he can, Bernie now has four Wintel machines for personal use: his latest laptop, a 14-months-old office desktop, and two older machines at home. As a result, he wastes six to eight hours every week just keeping it all running -- time he desperately needs for his real work.
Pointing out to him that the Linux administrator could accurately predict his next service shutdown several weeks in advance while the Wintel people can barely limp through a day without a reboot was like throwing water on a duck -- it generated vague resentment but no behavioral change.
Later I tried talking to him about using a Powerbook to avoid Windows futz time but got nowhere, and two weeks ago I sent him a reference to the Macintouch justification page.
And do you know what my efforts have gotten me so far? Right: I'm no longer his friend. Apparently I have an agenda here; some diabolical personal reason for wanting to him to buy a Mac. And it's true, I do have a reason: He'd be a lot happier and more productive using a Mac because it would let him focus on the things he's good at -- an impressive list that doesn't include Windows administration.
Unfortunately, it's impossible to get there from where he is. Instead, I'm going to hear, someday soon, that the stress killed him, either directly or in traffic. That's a bitter lesson, but one we all need to learn: Sometimes rationality and goodwill just aren't enough, and you have to walk away because you can't talk to the willfully deaf.
Paul Murphy, a LinuxInsider columnist, wrote and published The Unix Guide to Defenestration. Murphy is a 20-year veteran of the IT consulting industry, specializing in Unix and Unix-related management issues. .