Finder Arrow Keys¶
383 words
Usually I try to not comment on the Mac OS X Finder. Just because it’s such a depressing topic and, given our experiences of the past years, any such comments should be considered a waste of time. But sometimes you just feel you should laugh at the damn thing for its extensive brokenness.
The problem is a very simple one: Keyboard navigation only makes sense if it is somewhat predictable. In proper (i.e. spatially arranged) Finder windows, most notably the desktop that just isn’t the case. With an icon selected it usually is very hard to guess where pressing an arrow key will move the selection. In particular pressing the left arrow key and then the left arrow key often doesn’t return you to the selection you initially had, but a completely different one.
Let me illustrate this with a ladybird:
Doesn’t make much sense in my book. I wonder whether this bug is just cause by negligence or – one of my pet peeves – by programmers trying to be clever. By the latter I mean that programmers started off full of good intentions but – as they mostly do – implemented them to something that just doesn’t work well. In this case they could have attempted to ‘improve’ keyboard navigation in a way that pressing a horizontal arrow key will move the selection in a mostly horizontal way. I.e. you could imagine situations where there are two icons that are at the same ‘height’ on the screen and 100 pixels apart horizontally. And there might be an icon that’s technically between them but, say, 1000 pixels further up. So it’s quite out of the way and going there isn’t really a ‘horizontal’ movement.
But usually such ‘improvements’ just cause confusion: Invariably there’ll be situations where pressing a key will do something different from what the user expects. My motto for software usually is ‘don’t be smart’, as such smartness usually goes wrong. And in this case it even breaks the whole idea of the left and right (or similarly up and down) arrow keys being inverse to each other in their actions.
At the end of the day I don’t care too much, though, whether this problem (#4375636, for those who care) is caused by some little error or by poorly implemented ‘good intentions’.
