Of HIG, Apple, and User-Centric Design

[Dan Shafer]

When I said a few hundred posts ago that an unclickable checkbox was a 
violation of Apple's Human Interface Guidelines (HIG), what i *really* 
meant was that it violates what I'd consider to be *fundamental* UI 
design. Apple's HIG are the best single codification of UI design I 
know about, because they've been doing that (codifying) and studying 
the issue for longer than any mainstream software or hardware 
manufacturer.

I don't even know if Microsoft has set of codified UI guidelines; if 
they do, they clearly don't enforce it, even on their own products. And 
that's OK if that's how they want to proceed.

It is interesting that the Web browser has drastically altered peoples' 
expectations of the UI. In fact, it has lowered their expectations to a 
least common denominator that's not all that broad. And that may 
ultimately turn out to be a good thing.

I spent a good many years engaged primarily in GUI design (in fact, I 
own the domain gui.com) and am perhaps more sensitive to the issue than 
most users, even. But whether you choose Apple's HIG as your baseline 
(someone said they're no longer sure Apple's guidelines are all that 
good, but I challenge anyone to come up with a more coherent, 
consistent, and clear set of guidelines) or someone else's, the most 
important single rule of UI design that I have come up with in 15 years 
is simple:

Never surprise the user.

A surprised user is a confused user is a distressed user is a user who 
is not going to be a user one second longer than s/he has to be.

A checkbox I can't uncheck is a surprise. I never (or at least almost 
never) encounter such beasts. Put one into your program at the very 
high risk that I will be sufficiently uncomfortable that I will go 
away. And, what is perhaps more sinister and important, I will probably 
never tell you why I went away.