Skip navigation

Tag Archives: engineering

Apple, for some reason, has a reputation of making usable products, and of using very good engineering. Of course, I’m out to point out how neither couldn’t be further from the truth.

Engineering for dummies

I am not an engineer. But I do have a minimal understanding of what engineering is all about. Which is apparently more than can be said about all the idiots raving about the quality of Apple’s engineering.

Specially for them, this little excerpt of that difficultly consultable source of information, Wikipedia:

Engineering is the discipline, art and profession of acquiring and applying technical, scientific, and mathematical knowledge to design and implement materials, structures, machines, devices, systems, and processes that safely realize a desired objective or invention.

Now our idiotic Apple apologists seem to not know about the last portion of the definition. Engineering aims at realizing a desired objective. It’s not gratuitous aimless technological feats.

The irony is that Apple is extremely bad at engineering. At times they pull off industrial design decently, and even that, much less frequently than most people think. The rest of this post is dedicated to exemplifying this.

A few cheap shots

Before getting into the more complex arguments, I have to address, with spiteful delight, a few of the most idiotic arguments people give to defend the idea Apple is good at engineering.

The best one, and certainly the most ridiculous one, is the infamous “straight solder line” episode. According to some sources, Steve Jobs insisted all solder lines on the Apple ][ be perfectly straight. This is seen by some as attention to detail.

Sorry folks, that’s just plain idiocy. It has no added value whatsoever. It doesn’t enhance functionality, or provide extra reliability. It is just misspent effort.

Then of course, fast forward to 2010, and you have the iPhone 4 with it’s revolutionary antenna. Indeed, it is quite revolutionary to dare put out a phone with an antenna which is faulty by design in 2010. That’s not even attention to meaningless detail, it’s just sloppy – this kind of thing should be identified during testing.

UPDATE: Come on, serious engineers do not discard the warnings given by other engineers and partners.

With that out of the way, let’s get on to the more serious stuff – and please note I’m talking about computers in the rest of this post.

Computers designed not to be used: hardware

Apple & our favourite rodent

Apple brought us the mouse. Of course, contrary to what uninformed people sometimes believe, they didn’t invent it nor were they the first to put it on the market, but they were definitely the ones to popularize it.

Ironically, they have consistently made the worst mice on the market ever since. Raneko’s excellent mice gallery hereunder will help me illustrate the point. It lists Apple mice chronologically from right to left.

One can vaguely excuse the discomfort induced by most of the early mice it made – awful cubical boxes that actually hurt your hands seen on the right hand side of the picture – by the fact that they were the first mice around. Everybody needs experience to learn. Nobody was doing better at the time.

In 1993, they finally moved in the right direction with something a bit more round, vaguely more comfortable, the Apple Desktop Bus Mouse II. However, by that time, Logitech already had several much more comfortable products out there. Oh yeah, and they included multiple buttons too.

You’d think Apple would learn something from competitors, wouldn’t you? Wrong. They then go out of their way to design the pretty, but ridiculously uncomfortable “hockey puck” mouse for iMacs in 1998. By this time, very ergonomic mice are available for PCs. Good engineering usually does not include going from slightly decent to extremely bad, but that is what Apple did.

In the 2000’s, Apple ups the ante with the awful Mighty Mouse. Sure, it looks nice, but is uncomfortable to use, the long awaited arrival of extra buttons is countered by the fact they are nearly impossible to use. And in the day and age everybody was moving away from mechanical balls under mice because they clog up so easily, Apple goes and puts that on top of the mouse as a scrolling device. Massive failure.

Not to worry, in 2009 Apple shows it still has it’s mojo by producing a stunningly gorgeous mouse, the Magic Mouse, which is still ill-conceived. It’s a pity really. It sports lots of nifty features never seen before on a desktop mouse, but forgets to do what other manufacturers have figured out how to do over twenty years ago: be comfortable.

And that’s the key to it all really. Apple has, for thirty years, consistently failed at making a mouse which is comfortable to use, respecting basic ergonomics of human hands. This has been unacceptable for twenty years. And yet these guys are engineering geniuses? Sorry, but a mouse’s first feature is to be comfortable in your hand when controlling your computer for any period of time.

But Apple makes computers that are supposed to be pretty to look at, not used.

Apple & the keys to your computer

That also shows in their keyboards.

In this domain, their track record is decent. Their keyboards have been more or less acceptable, given the era, most of the time.

But then, in 2009, Apple decides numeric pads, a standard feature on most desktop keyboards since the IBM PC went on the market, are not useful anymore.  This is the time where more and more laptops include numeric pads. Any heavy computer user appreciates a numeric pad, even if, granted, one can live without it.

But better than that, they made the keys perfectly flat. Not concave at all.

Casual computer users won’t mind. But anyone who does extensive work on computers should be complaining. Because flat keys are a huge pain for touch typing. This is not something people have just discovered either. Keyboards have had concave tops for decades for precisely this reason. A good engineer never discards lessons from the past without good reason.

Again, hardware made to be looked at rather than used a lot.

Quality first – sometimes

Apple computers and devices are on average more expensive than other brands. All the more so – and this is seldom taken into account – because it’s usually quite harder to upgrade a Mac than a PC, and if your screen dies, well, you’re basically good to buy another computer.

But Apple loyalists like to claim this is due to the high quality of the products.

Unfortunately for them, that’s just not true. Regardless the design flaws pointed out above (which should actually not be disregarded when paying a premium).

Quality control at Apple means massively delivering non-booting iMacs (or alternatively broken screens). A company that takes build quality seriously – while actually charging a premium for it – would not have that kind of issue. This kind of problem is expected of low-end devices, not top of the range machines.

Even more interesting is the failure rate of Apple laptops. While being bested by Sony, who basically builds premium laptops too, is acceptable, it is quite damning for Apple to be bested by Toshiba and Asus, two companies which are in the commodity laptop business. Much cheaper and more reliable.

What was that about quality again?

It’s the software stupid!

But of course, Apple loyalists will be quick to point out that I’m missing the point, because the Apple experience is all about the quality of their software, their beloved Mac OS X.

Now there’s no disputing the fact Mac OS X is a good OS. But it is far from immune from criticism, including in the engineering & usability department. Given Apple’s reputation, I find that quite ironic.

Usability issues

The Apple menu bar

The Apple menu bar is one of my favourites. It’s a typical case of difference between Windows and Mac OS, so people usually tell me it’s a question of getting used to it. Unfortunately, that’s not all it is.

Just a reminder for those unfamiliar with Mac OS: the Apple menu bar provides the functionality you get with Windows applications in a central unique location for all applications on your Mac: at the top of the screen. Check out the link above if you want to know more.

The idea behind placing all that up there is to conform to Fitts’s law. This basically states that

the time required to rapidly move to a target area is a function of the distance to and the size of the target.

The consequence being that the top of your screen having “infinite width”, it’s easier to acquire the top of your screen than the top of your window.

The problem is not that Fitts’s law is untrue. It’s that it is a very bad metric.

For better or for worse, computer interfaces have been for decades using the desktop metaphor. Here again, Apple deserves credit for popularizing the desktop metaphor (with the Macintosh). The basic idea is that your computer mimics, insofar as applicable, the way you work with your physical desktop.

Now the desktop metaphor is broken in many ways, and most operating systems wander away from it in various ways, usually for good reasons. For instance, spatial file-browsing just sucks for most people, so breaking the metaphor in that case is a good idea.

But the way the Apple menu bar breaks the metaphor is Very Bad™. Basically, it breaks away the application and it’s configuration (and options, etc.). Which is the equivalent of needing to rummage in your desk drawers to configure the stapler that is in your hands. It dissociates an object (the application) from it’s properties (configuration, options etc.). Which is a disruption of your thought process.

Furthermore, if Fitts’s law really mattered to Apple, most actions could be carried out by a context click (equivalent of the right-click in Windows), since no movement whatsoever is needed. But that’s not the case. Understandably so, since the actual right-click is still a despised feature only begrudgingly given to it’s users by Apple.

Adding insult to injury, and perfectly negating any benefits one might have found in Fitts’s law’s application, the Apple menu bar can also result in unnecessary mouse-clicks. Indeed, if another window than the one you want to access the menu of is in focus, you need first to bring that window in focus, and then only do you have the opportunity to access the menu. Talk about usability.

And of course, let’s not forget that with larger and larger screens, multiple screens and the like, reaching the very top left of your screen is getting to be more of a big deal than it was on the screens of thirty years ago.

Window controls

Another usability pet peeve of mine is the non-discoverability of the window controls in OS X.  You see, the close button, the “zoom” button (what comes closest to maximize in the Apple world) and the minimize button sport the red, green and orange colours respectively. No visual cue as to what they do. Just colours. So while red is about as universal as an X for signalling that what the user is about to do is dismiss or reject what he is doing (=close his application), green and orange tell me nothing at all about the current functionality of the button. To have the faintest idea, I need to hover to the button to get the visual cues which Windows (among others) permanently provides.

And just to make it more fun, the colours will change according to your theme. I guess that helps the colour-blind people who would otherwise be screwed… Not really user-friendly.

Ergonomics placement

Finally, another ergonomics issue which people dismiss as a question of habit when it is far from it.

For this example, I’ll start off with the windows way. You see, the Windows menu and the window controls are placed like they are for a reason. The Windows menu is at the bottom left of your screen, while window controls (minimize, maximize, close) are at the top right of your window.

This is purposefully thought out for right-handed people, which are, after all, the vast majority of people (sorry lefties). Why? Those areas, which are destined to be very frequently used, are the ones which require the less physical effort to reach with a mouse. It’s a simple rotation of your arm on your elbow axis which allows you to reach them. Your elbow doesn’t need to move. In Mac OS, those window controls are top left. Which means you necessarily need to move your whole arm (elbow included) to reach them.

Probably not a big deal, but here again, attention to detail is far from characterizing Apple, and in this case does characterize Windows.

Keep it simple… as long as it’s simple

Just for laughs, I’ll add a link to Bruce Tognazzini’s webpage. I disagree with him on a number of questions, but this particular page made me smile in agreement. For reference, the guy is one of the original Mac OS engineers and founded the Apple Human Interface Group. The most important take-away is that Apple’s “keep it simple” philosophy actually makes your life harder over time. But it’s alright as long as you don’t use your computer much…

Engineering issues

Then there are those nice little things in OS X that are just plain dumb. It actually doesn’t even boil down to engineering, but just to plain common sense.

See, that close button I was talking about, it doesn’t actually close the application, it just closes the window. While I can make sense out of that as all geeks can (because no, an application is not just the window), my non-geek friends can’t (because to them the application is just the window).

Again, this simply breaks the desktop metaphor. If I dismiss an object on my desk, it’s not supposed to be lying around on the desk afterwards. Otherwise I would have left it available (which is what minimizing is for in computing). Yet in Mac OS X, that’s what it does, it’s still there, hogging up space (system resources). And it’s even quite confusing to the casual user because it’s there hogging up resources, yet it’s not clearly visible.

And there is no one-click way to actually quit the application. If I want to do it, I have to use keyboard shortcuts or multiple clicks. Great.

Now let’s even concede the difference between closing the window and quitting the application to Apple (we shouldn’t, certainly not to an OS that claims to be so user-friendly, but still, just for the sake of argument). It’s poorly executed.

First, it’s not done consistently. There are applications which you actually quit (for good) by clicking on the close button. Just not all of them. See, those where you can only open a single window are actually quit, while those where multiple windows can be opened are left running (even if only one window is open). Good luck explaining that to a casual user. That’s bad engineering, and bad usability.

Second, if you’re going to do that kind of “close window but not quit application” nonsense in an operating system that is basically built on top of Unix, you should use the suspend function that has been around in Unix for decades, so that you waste no resources. But that is of course not how it is done. Bad engineering.

And then there’s a last unrelated engineering mess-up I find amusing. Mouse acceleration is exceedingly lousy on Macs. It’s one of the reasons I always feel weird when using one. It’s actually so bad that users have actually started a petition to request it be changed and there are several freeware applications out there to correct the problem…

Let me spell it out for you. Not only has Apple succeeded in mis-implementing the algorithm for mouse acceleration so that default behaviour is sucky, but they do not provide controls to fix this problem (even though such controls are fairly trivial to implement). And when it gets hilarious is when you notice this problem was introduced by Mac OS X – previous versions handled this correctly. Again, Apple manages to actually regress instead of progress. Congratulations!

Where does that leave us?

Am I trying to argue that Apple only makes badly engineered unusable lame products? Hardly. Apple makes good products. Definitely very desirable products, and the looks of most of them are absolutely beautiful – even though I am personally starting to tire of the “white iPod” look all over the place and welcome the change that is currently occurring.

Only an idiot would defend the viewpoint that Mac OS is a crappy OS or that the iMac is a bad computer.

However, Apple products have just as many quirks as other manufacturer’s products do, and the track record is actually getting spottier by the day. Furthermore, between the premium price and the much more limited software availability than on the Wintel platform, the products just exclude themselves out of many markets – it’s no surprise Apple doesn’t sell that much to enterprises these days.  And of course, the overarching impression that form is more important than functionality oozes ever more strongly out of the Cupertino-based products (re-read the hardware section if you’re still in denial about this).

As such, what are Apple products? Niche products. In the same way shiny sports cars are niche products. A few of the more affluent people with specific tastes will buy them. In the case of sports cars, because driving fast and “sporty” is enjoyable. In the case of Apple, because owning a beautiful computer is pleasing. And for the geekier because you have a Unix foundation.

But just like the sports car owner would be an idiot to mock the sedan/station wagon owner or deride him about engineering feats when his sports car can’t carry many people around or hurts his back, the Mac owner is in a very bad position to mock computers whose primary goal is not to be aesthetically pleasing, but to get the job done.

And just like sports cars are niche, so are Macs. With a market penetration of 5.1%. Because despite the hype, PCs are the true computers for the rest of us.

In a nutshell

Apple’s engineering and quality is certainly not above industry par. It is just as guilty as the next company of releasing shoddy products (and has a consistent track record of this in the mouse area). It routinely fails to apply basic principles, even going so far as to make things worse over time.

As such, it’s reputation of quality and good engineering is an absolute myth. Given the low standards of the IT industry, this doesn’t stop them from making desirable products, mainly thanks to their aesthetics. As such, there is a very good market for Apple products – albeit a niche one. There’s definitely nothing stupid about buying a Mac, but it’s definitely ridiculous to claim Macs are vastly superior to PCs in any domain other than aesthetics.

Advertisements