LOL. Sounds like a bad time to be had in the red light district.

I’m not saying you should chase perfection, just make something more suited for now.

I try to remind myself that those weeks of interface “improvements” could have gone into actual features or improved functionality or new products that probably would have been of more value to my customers

There are plenty of apps out there with too many features and an unsavory user experience. Surely there’s a balance.

Cheap UI shot in passing: Garage Band.

I don’t follow.

The problem is that the search for UI perfection becomes an end in itself. I have literally spent months of my life tweaking, re-drawing, re-coding interface elements. Yes, the product is better looking for it (most of the time), but at what cost?

By “hacks”, I mean adjusting a design, or redesigning it or recoding it to make a certain look “work” because the control is broken or misfeatured in a particular OS version, because it’s inherently buggy or misdesigned, etc. It’s even more hackalicious if the work is probably going to be superseded in the very next product release, or more likely, OS version.

Perfectionism has a price. These days, I try to remind myself that those weeks of interface “improvements” could have gone into actual features or improved functionality or new products that probably would have been of more value to my customers.

As for customers liking new things, that’s true, but I also remember that I have been more impressed with iTunes newly colored icons than I have iTunes newly fashioned UI elements… ;)

Cheap UI shot in passing: Garage Band.

Just a note that Jeff Ganyard just released an enhanced version of my TunesWindow code that uses CGGradients to draw its textured areas, and has configurable colours, heights on the top and bottom borders, and so on. Get it here.

Not changing the UI is safer in the short term, but it’s also limiting. You might be preventing yourself from making dramatic improvements. If you’re concerned about messing things up, why come into work at all?

They had an interface that, quite simply, worked. And they’ve replaced it with one that… doesn’t.

As far as I can tell, pretty much everyone knows how to use iTunes. There are certainly differences in taste (you use the word “classy” above), but that’s outside of whether something works correctly or not.

For better or worse, consumers like things to feel new. If Microsoft continually updated their media player UI and Apple did not, I think iTunes would start to feel stale to many people.

workarounds, undocumented hacks and so forth, really worth the effort for our customers

As far as I can tell, nothing Daniel described involves hacks. You’ve always been allowed to draw your own controls.

BUT, I agree with patrick when he says that I prefer the original window. It is simple, yet still usable and classy.

(For instance, your initial buttons are FAR more useful and clean and attractive than your final buttons, IMO.)

As someone who has to make this sort of decision all day long, everrrrry day (sigh), I know this stuff is hard.

I have an idea which I call the “iTunes UI Sucks Theory”. It goes something like this:

As developers, we get bored with our own work. We create an interface for our product, polish it a little, stand back and think “cool”, or at least OK. Fast foward 3 months at 10 hours a day of looking at that same interface and, oh no, the interface now looks staid or even ugly. So, we suddenly decide the interface needs… improving.

When, in reality, it probably doesn’t!

The debacle that is the iTunes UI is exhibit A in this theory. They had an interface that, quite simply, worked. And they’ve replaced it with one that… doesn’t. Game developers have this problem too, which is why game developers should probably create their UI at the end of their projects. (Witness the Quake3Arena betas: UI boredom mixing with too much money is a potent, toxic brew.)

Above all, I think we have to ask ourselves is a month of work, with all the headaches involved in workarounds, undocumented hacks and so forth, really worth the effort for our customers?

The trick here is that every improvement can always be justified, by a developer. It looks better! People will use/buy it more! But really, as I’ve found to my cost too many times, a lot of interface work, especially technical, under-the-hood work, isn’t justified.

But again, very good blog.

I also enjoyed your Photoshop process. I am certainly not well-versed in that tool, but I’m finding that I can do just a little bit better each time I go back and dress up some of my graphics. I almost used the PolishedWindow as well, but I played with it just a couple days before my release and I couldn’t justify the last minute change. We always need to leave something for the next release, right?

Nice work overall on your product!