The Lack of Affordance Problem

I've often said that the hallmark of good engineering is designing your error cases: of the cases where something goes wrong being pleasant and non-disastrous. The issue is that optimizing some known quality, even one as subjective as aesthetics, is a lot easier than optimizing unknowns, which is what designing failure cases requires. That's why it's hard, why it requires deep and broad domain knowledge, and why it requires knowing your users. It's hard, really, to point to software or hardware that does this well, but it's easy to point to bad examples, such as the OP. But consider things like undo, version control and versioned file systems; think about TCP's recovery abilities, about the fact that Google 2-factor auth gives you lock-out codes and application-specific passwords, the X~ and #X# files Emacs dumps all over your harddrive (they've saved me more than once), and the recovery page you get when Firefox crashes. Or consider the HTML5 parsing algorithm, character set detection, and Google's spelling suggestions. Each of these required identifying a class of problems and specifically designing in features that protected against them.

Please, when designing a project, think not just about your features, usability, and similar features. They're important, crucial even. But also spend time thinking about what problems users will have. What bugs are you likely to have, what misbehavior of your program will be most damaging? Implement a common format because that'll make debugging and workarounds easier, add a little diagnostic console or log in case your users have a problem you didn't expect, make backups of data you destructively modify, and so on. We all idolize Mac Airs now, for beautifully-designed and slick software. But there's something to be said for hardware and software that works through thick and thin, where there's some workaround or safety measure for every failure.

You're wrong. I'm way out here in the apple camp myself (see me wave among my iMacs and Airs? Hi.), but he's most definitely correct with his general comment that the lack of affordances is something Apple doesn't solve. They do this out of love for simplicity, and while they often find the right balance between that simplicity and raw hidden power, sometimes (as is the case with the external Superdrive) you do need control.

It is possible to have both control and looks (most Apple products are good in this regard), but in the case of the superdrive, they went just a notch too far and assumed that the 1% of cases when you need to force a disc out, doesn't exist.

The criticism of Apple's USB drive was insightful, and showed a definite flaw in the design of the unit. However, these two examples are just petty. First, any time you need your serial number you will be told how to get it, and clicking Apple -> About This Mac -> More Info is not an exceptional amount of work. Compare this with what you have to do on Windows:

http://support.microsoft.com/kb/558124

Also, though this is not obvious when the unit is purchased, the serial number is etched into the bottom of the casing, not painted as the rest of the text is. I only discovered this when the rest of the text rubbed off of my MacBook Pro. This means that it can't be printed in a different color, but I'd rather have the permanence of etching over paint that can be scratched off.

Finally, the USB extension cable cannot be bought alone - it only ships with an Apple keyboard. It's been designed for the electrical requirements of the keyboard and not all USB devices, so it's understandable that Apple limits its use to the device it's manufactured for.

Is it just me or does it seem that people are a bit more upfront criticizing Apple design now that Steve is gone?

The problem with the non-standard connector is this isn't an Apple defined standard. This is USB. And the problem with this design is I can use a non-Apple cable to extend the keyboard (because the keyboard has a USB plug on it), but I can't use the extender for something else (e.g. even an Apple mouse!).