And Oranges


Most people, I suspect, can tell the difference between a $500 bicycle and a $1000 bicycle — most don’t think the differences are worth the extra money, but they can perceive the differences (lighter-weight frames, visibly superior components, etc.).

I doubt, though, that most non-expert-cyclists are able to distinguish between, say, a $1000 bike and a $5000 bike. This holds true for many realms; it’s generally pretty easy for most people to perceive the difference between the mediocre and the good, but it often requires expertise or finely-honed sensibility to perceive the difference between the good and the great.

Furthermore, in some cases, most people generally prefer the good to the great. A high-end racing bicycle is a terrible bike for casual transportation around town — the seats are uncomfortable and the pedals require special footwear. Cameras are another good example. Most recreational photographers really do just want to point and shoot; they don’t buy point-and-shoot cameras just because they’re cheaper than SLRs, they buy them because they’re actually what they want.

“Better” is a complicated adjective, because it is subjective. Whether product A is better than Product B depends entirely upon what criteria matter to you. A $500 bike is better than a $5000 bike if you’re looking for something to ride around your college campus. A $300 Digital Elph is better than an $800 Digital Rebel XT if you want a camera that’s simply easy to use and fits in your pocket.

But just because evaluating “better-ness” is subjective does not mean it is arbitrary. For most products, most people appreciate this. Even though they can’t perceive the differences between the good and great — or if they do perceive the differences, don’t care about or don’t appreciate them — they don’t dispute that such differences actually exist, or that connoisseurs judge products in their area of expertise by what are sometimes very different criteria.

This is as it should be, and perhaps it all sounds rather obvious.

The point of all this is that in some cases, some people seem unwilling to concede that any criteria other than the ones they themselves deem important actually matter, or even exist.

That’s dogmatism,1 and the nature of dogma is such that it pretty much kills any reasonable discussion or debate.


Which brings us to Mark Pilgrim’s announcement two weeks ago that he’s switching from Mac OS X to Ubuntu Linux running on a Lenovo ThinkCentre.

It goes without saying that this is a rather curious direction in which to switch — the trend is the other way around, with far more people switching to the Mac than from it. It is even more curious if you know anything at all about Mark Pilgrim’s background and personal computer history. Pilgrim is not some Mac OS X-era carpetbagger. He’s a real Mac user, having been using Apple computers for 22 years. In the early ’90s, he developed a slew of cool games, hacks, and utilities for the Mac — and, notably, released them as open source under the GPL; there was very little open source software for the Mac in the System 7 era.

Pilgrim is an expert Mac OS X user; before taking his current job at IBM, he worked as an Apple-certified Mac OS X trainer. When Mac OS X 10.3 (a.k.a. Panther) shipped, he published this wonderfully exhaustive guide to what was new from 10.2. His weblog, Dive Into Mark — which prior to his taking an 18-month hiatus in October 2004 was extraordinarily popular and deservedly so, and was in fact an enormous influence on Daring Fireball — was never a “Mac weblog” or “Apple weblog”, but did contain a lot of insightful, informative stuff about Mac OS X and Apple. And even while his weblog was on hiatus, he was still publishing deeply investigative grist for Mac nerds.

All of which is to say that I found it very surprising that, when I linked to his announcement and further explanation, it elicited a wave of email more or less along the lines of “who gives a crap what computer this Pilgrim guy buys?”

Not so unexpected, but disappointing nonetheless, is that nearly all of those contributing to the lengthy comment threads on Pilgrim’s weblog seem to entirely and utterly miss the point of his argument — which argument is complex and in many ways subtle, which, in turn, is probably why most people chose not to think about it.

Interpolation on the Influence of Yours Truly on the Tone of the Comment Thread for Pilgrim’s ‘When the Bough Breaks’: It seems clear to me and deserves to be mentioned here in the open that part of the digressive “missing the point” nature of the comments on “When the Bough Breaks” are my fault. When I Link-Listed it, I quoted a paragraph that, I thought, captured a key aspect of Pilgrim’s argument:

I’m creating things now that I want to be able to read, hear, watch, search, and filter 50 years from now. Despite all their emphasis on content creators, Apple has made it clear that they do not share this goal. Openness is not a cargo cult. Some get it, some don’t. Apple doesn’t.

But in what I intended as an aside, I then commented on Pilgrim’s griping about having lost metadata in both iPhoto and iTunes to database corruption, which he suggested might have been somehow mitigated had those database file formats been publicly documented (which they aren’t). I wrote:

It’s obvious he’s been thinking about this for a long time, but I don’t agree with all of it. For example, he’s been bitten by metadata database corruption in both iPhoto and iTunes; I think the solution here is a good backup strategy, not open file formats for the binary metadata stores. Knowing the format of the file won’t help if the corruption is severe; a good back-up will always work.

I agree, though, that closed-format metadata stores are not good long-term archival formats.

I only mentioned backups to refute that one single point; not to refute Pilgrim’s entire argument.2 The problem is that a big chunk of the traffic to his post came from my link, and the comments, especially the early ones, are dominated by people telling him that his problems with Mac OS X could be solved by backing up. I think that’s the fault of the misplaced emphasis in what I wrote. (Pilgrim thinks so, too.) End interpolation.

So why is Pilgrim switching? For the benefit of those of you who haven’t read his piece yet and are too lazy to pause here and read it yourselves — which you really ought to do, if for no other reason than that he’s a good writer — I’ll list several of his cited reasons:

  • He’s a Firefox and Emacs junkie, and doesn’t need a Mac to use them. (I’ll add that Firefox, in my opinion, works better, or at least fits in better, on Windows and on Linux than it does on Mac OS X.)

  • He’s unhappy about the Apple software he does use — iTunes, iPhoto, and iMovie — because of the aforementioned corruption problems with the first two and because iMovie doesn’t let you export edit decision lists.3

  • He doesn’t like the iTunes Music Store’s DRM, and because he doesn’t like ITMS DRM, he doesn’t like the increasing number of ITMS tie-ins in each new revision of iTunes.

  • He’s irritated by QuickTime Pro’s $30 price tag, which most people pay just so they can watch movies in full-screen mode.

  • He doesn’t like .Mac. (He’s not alone.)

But the crux of his argument is the afore-quoted passage regarding openness and long-term data preservation. That’s good, because the points mentioned above really don’t make for much of a case, on their own.

Arguing over these points, however, which is what most of the commenters on his site have done, is missing the point. Yes, they’re all niggles, but Pilgrim isn’t claiming otherwise; he is simply pointing out the obvious, that a lot of little things about Mac OS X suck. These niggles don’t constitute the full case for his switch; they’re just factors that made his decision easier.

There’s an unbecoming tendency for some Mac users to contort their worldview in such a way so as to construe that Mac OS X is better than every other OS in every single way, or that its overall superiority ought to be obvious to everyone. This actually was true, or very nearly so, in the System 6 era of the late ’80s, but it certainly hasn’t been true since then; sticking to this notion just makes you look like a small-minded jackass. (Not to mention that many of the people I’m describing weren’t even out of diapers when System 6 was current.)

There are ways that Windows is better than the Mac, and many of them are rather obvious. E.g. if you need AutoCAD, Windows is better. If you want to play Half Life 2, Windows is better. For most of the past 20 years, PCs offered more raw computing performance than Macs. These are facts, but that hasn’t stopped some Mac users from trying to deny them.

The ways that Mac OS X is better are in many ways more subjective, and for many people, imperceptible. Or perceptible but they just don’t care. Many Mac users just can’t get it through their heads that most people don’t give a rip about the quality of the design of the user interfaces they use; that for most people “good enough” is good enough. Mac OS X is better than Windows if you care about user interface design; if you don’t, though, it’s just different, not better.

It is easier, or more comforting, for Windows nerds who do not perceive anything significantly better about the Mac’s UI to chalk up Mac users’ strong preference to their being irrational cultists than it is to admit that they (the Mac users) perhaps have a more acute sense of UI design aesthetics. Contrast this to the markets for, say, bicycles or cameras. As mentioned at the outset, lay people are, as a general rule, perfectly happy to concede that these high-end products are better in ways that they don’t perceive or understand or care about.

The difference is that with cameras and bicycles, the differences between high-end and consumer-end products are almost always measurable. Expensive bikes weigh less and have sturdier frames. High-end cameras have faster lenses and digital sensors with higher ISO sensitivity and can shoot more frames per second. Lay people may not understand or know about these differences, or why these differences are worth thousands of extra dollars to serious cyclists and photographers, but they could understand them if they wanted to sit down and learn about them. It’s just that they don’t care so they don’t bother.

I.e. the key to understanding high-end bikes and cameras is knowledge, not perception. (It’s a matter of taste which high-end bike or camera you buy, but not a matter of taste whether you want a $2000 DSLR or a $200 point-and-shoot.)

Whereas if we concede that the Mac is a computer with a high-end user interface, what it is that makes its UI “high-end” is a matter of taste and perception. There do exist ways that you can scientifically measure and test the usability of software — i.e. that by Fitts’s Law you can prove that the Mac menu bar is more usable than the Windows-style menu bar within each window — but that’s not why Mac users are Mac users. No one goes out and subjects themselves to serious usability testing to determine which OS they should use. Mac users prefer the Mac because they just prefer it. You can explain the reasons why you prefer the Mac, but can you explain why you care about those reasons? You either feel it or you don’t.

Hence the difficult situation faced by small-minded Windows users who do not get the appeal of the Mac; to admit that Mac users’ strong preferences are reasonable would be to admit that they (the Windows users) are unable to perceive something that Mac users can. That to concede that Mac users are reasonable wouldn’t just imply that Mac OS X is in at least certain ways much better than Windows, but that Mac users in certain ways have a more refined sense of taste than Windows users, which in turn cuts way too close to implying that in certain ways Mac users are smarter, which is where things turn ugly because those certain ways are, to Mac users, the ways that really matter, and any chance at reasonable discourse evaporates because both sides feel deeply insulted by the other.

The openness issues that Pilgrim is talking about are like that — except that maybe even fewer people perceive or deeply care about these issues than about user interface design. And the way that his arguments are being dismissed, by most of the responses I’ve seen, is similar, rhetorically, to the infuriating manner in which many Windows partisans have long dismissed the Mac. To wit: denying that the cited factors exist or matter, and thereby ascribing the avowed preference to irrational zealotry rather than to refined or acute sensibilities.

Telling Pilgrim that he’s making a mistake because Ubuntu doesn’t have as refined or cohesive a UI as Mac OS X is like telling someone who is switching from a Chevy Tahoe to a Toyota Prius that he’s not going to have as much cargo room. He knows it.


The point here isn’t that everyone ought to agree that the Mac has a “better” user interface than Windows or that the Mac isn’t “open” enough compared to Linux; the point is that everyone ought to be able to have a reasonable discussion about these issues and their relative merits, and do so with the understanding that no one ought to expect to be able to outright prove their case one way or the other.

It’s not a binary situation, where Windows and Linux have bad UIs and Mac OS X has a good one, and but Linux is “open” and Mac OS X and Windows are not. Nor is there just a single continuum upon which to gauge these choices.

It’s often said that you shouldn’t compare apples and oranges — generally used figuratively, but even looking at it literally you can see that it’s not true. You need to compare apples and oranges when you’re deciding what to pack in your lunch. What you can’t do is compare apples and oranges and somehow conclusively prove that one is better than the other.

Or ask yourself this: what would you rather read: a well-plotted but poorly-written potboiler or a well-written novel with a rather nondescript story line? A quick look at the best-seller lists tells you how most people would answer. The point is that you don’t choose one novel over another because it is somehow universally “better”, but rather because it is somehow more appealing, better for you, as an individual, based on the innumerable inscrutable tastes and desires and opinions that make you the unique snowflake that you are.

The reason this Pilgrim situation is so hideously complex is that all modern operating systems are complex. It takes a lot of work and investigation and expertise just to understand and form opinions about one of them, on its own; comparing one against another can’t be done by reducing the comparison to some single metric because they’re different in so many different ways. It’s easy to choose between two things that differ from each other in just one way — and it’s easy to explain your decision. Not so when choosing between things that differ in hundreds or maybe even thousands of ways.

Assuming you’ve made your choice with an open mind, it’s really unlikely that your decision came down to just one factor, or even just a handful. The interesting decision isn’t really about which choice you ultimately make, but in which factors you use to make that decision, and how much relative weight to ascribe to them.

It might strike you as crazy that when Pilgrim first decided to switch, he hadn’t even yet decided what open source operating system to switch to. I.e. how could it be anything other than a rash decision if he didn’t even know exactly what he was switching to? But the particulars of the OS he’s switching to don’t really matter, because what’s changed isn’t his opinion about Mac OS X or any particular distribution of Linux, but rather the amount of importance he’s decided to place on the openness of the OS he’s using. To place more value on openness, less on UI refinement, and thus to conclude that he’d be happier switching to Linux. That he has a long list of petty annoyances about Mac OS X and that desktop Linux continues to improve in terms of UI refinement — these things of course factor in as well.

Everything factors in is what I’m trying to say. What’s more important? Openness of data and control over one’s environment? Or aesthetic appeal and usability? It’s a gross error to think that this is some sort of binary choice, that on the one side you have openness, and on the other usability. Many aspects of Mac OS X are open. Ubuntu’s UI is steadily improving. The hard part is deciding how much importance to give to each factor you care about. How much openness are you willing to give up for a system with a better interface? How awkward of a UI are you willing to tolerate for a system that offers you the source code to everything?

Now factor in your habits and what you already know. I, for example, am far less likely to switch away from Mac OS X than I would be if I weren’t already an expert user. This isn’t blind loyalty or zealotry; expertise has value, and in my case it would take a lot of time to become equally proficient on another system.

So how do you evaluate so many different factors and reach a conclusion? I don’t know. (Paging Malcolm Gladwell.) You can’t just test drive an Accord and a Camry and conclude that one is good and the other is bad and drive off with the good one. There’s no way to compile a list that conclusively shows one is absolutely better. At some point it comes down to a feeling. Informed by thinking, after measuring the facts, yes. But a gut decision nonetheless.


You’re doing yourself a disservice if you dismiss an argument like Pilgrim’s simply because you believe he’s an open source/open format ideologue; ideologues aren’t necessarily irrational zealots. (And even irrational zealots or fanatics aren’t necessarily wrong; cf. Henry Kissinger’s quip: “Even a paranoid has some real enemies.”) An ideology is an organized system of beliefs; just because you don’t share them doesn’t mean they aren’t valid.

Ideological conviction doesn’t necessarily imply a rigid, quick-to-judge closed mind (even though, admittedly, that is often the case). You can be an ideologue with an open, honest mind — to believe otherwise is to say that someone with an open mind can never reach an uncompromising conclusion.

If your reaction to Pilgrim’s announcement was a snap judgment that he’s lost it, or that he’s being an asshole who’s just looking for attention as the guy who switched away from the Mac just at the time when it (the Mac) seems poised to become more popular than ever, or that he’s an open source fanatic who just can’t be reasoned with or trusted — are you sure that the zealotry at play is his?

Thought experiment: Let’s say that Microsoft puts together a miraculous fourth-quarter comeback and that Windows Vista rocks. Not just rocks compared to the way it currently appears as though Vista is actually going to turn out, but rocks, period. As in looks better than Mac OS X. More elegant than Mac OS X. Noticeably faster and snappier than Mac OS X. (That one’s actually quite likely.) I.e. “better” than Mac OS X, in glaringly obvious ways.

How would this make you feel?

If your answer is that it would depress you, or sicken you, or really in any way dampen your spirits — why? Wouldn’t this be good news for everyone, including Mac users? Either Apple would have to up the ante and improve Mac OS X in similar ways, or, Mac users could just switch. Either way the result is that we’d be able to use something better than what we’re using now.

One reasonable objection would be to point out that the Microsoft Corporation has a proven record of cheating and rule-bending and law-breaking and really just generally unsavory conduct at its executive levels, and that because of this track record it really ought to be depressing for Microsoft to “win” in this regard.

So let’s say instead it’s Ubuntu Linux.

How can you argue that this would be anything but a good thing?

I’m deeply suspicious of Mac users who claim to be perfectly happy with Mac OS X. Real Mac users, to me, are people with much higher standards, impossibly high standards, and who use Macs not because they’re great, but because they suck less than everything else. Pilgrim, to me, is a quintessential Mac user in that regard; and what he’s doing is wondering if maybe things might suck less somewhere else.

“Better” for Mark Pilgrim doesn’t mean better for you, and nowhere has Pilgrim implied that it does.

Admitting that he has a point, or several points, or that he may well be correct that he’s going to be more satisfied with Ubuntu than he was with Mac OS X, does not imply that Mac users are wrong or stupid or foolish.

And the truth is I’m not entirely sure he’s making the right decision, even for himself. Forget all the niggling details he cites, and focus only on his central beef — that Apple is a company that does not “get” openness, and that this deficiency is going to hinder Pilgrim’s long-term access to the data he’s creating. But if that’s the case, and Pilgrim has been using Apple computers for 22 years, why hasn’t it happened already? Openness isn’t binary, a choice between totally open and totally closed, it’s a continuum. The question isn’t “Does Apple get it?”, but “Does Apple get it enough?” But from the perspective of someone immersed in the free software culture, where everything operates near the extreme edge of the open/closed continuum, it’s easy to see how things begin to look binary — not open/closed, but totally-open/not-totally-open.

That, to me, is the most convincing counter-argument to Pilgrim’s. That if Apple’s lack of openness were a disaster in the making, it (the disaster) would have occurred already. And while it is easy to find ways to complain that Apple is not open enough — under-documented and undocumented security updates and system revisions, under-documented and undocumented file formats — it would be hard to argue with the premise that Apple today is more open than it has ever been before. (Exhibit A: the Web Kit project.)

But there are things that could be better, should be better, but aren’t, and it’s hard to ascribe these policies to anything other than management that is, at best, indifferent to issues related to openness. Cf. Tom Yager’s report in Macworld yesterday of being invited to Apple to discuss his InfoWorld column on the closing of the x86 version of the Darwin kernel; the gist of Apple’s response was to tell Yager that he shouldn’t have written about it because no one cares, which response is rather painfully tone-deaf PR-wise, in that the reason Yager wrote it in the first place is that he obviously thinks his readers do care.

Stories like that give rise to the sinking feeling that Apple’s executives aren’t merely indifferent to openness, but rather that their stance on openness is in fact highly calculated, and that the calculation is that Apple should be open only so far as necessary to be perceived as being open. I.e. that openness only pertains to marketing, and not to engineering. There’s no question that Apple is chock-a-block with engineers and engineering managers who care about openness, but that does little good without a mandate from the executive level.

I’m willing to admit that I don’t have a finely-tuned sense for this stuff. Pilgrim does, and that’s why I’m listening.

  1. I very much like the definition of “dogmatism” from Mac OS X’s New Oxford American Dictionary: “the tendency to lay down principles as incontrovertibly true, without consideration of evidence or the opinions of others”. ↩︎

  2. If I were more introspective, I might observe here that focusing on an easily-refuted side point in Pilgrim’s piece was likely a sort of defensive gesture so as to avoid looking squarely at some of the uncomfortable issues raised by Pilgrim’s main argument. ↩︎

  3. A standard format for describing the edits you make to video; given an EDL file and the original footage, you can recreate the final output you’ve edited together using any other software that supports EDL files. ↩︎