Does This Brushed Metal Stuff Really Even Matter?

Friday, 22 October 2004

I’ve been thinking about brushed metal windows and Apple’s inconsistent use thereof at least since Panther shipped a year ago. So, why wait until now to write about it? Well, because I just wasn’t sure it actually mattered. And I’m still unsure.

There should be no dispute that Apple has no consistent policy on the use of brushed metal, and third party developers have followed Apple’s example, rather than following the HIG. The best counterargument to what I’ve written is not that there actually exists some inscrutable policy governing Apple’s use of brushed metal windows, but that it just isn’t important.

Mike Manzano crafted a well-reasoned response to both “Brushed Metal and the HIG” and “Today’s Theme: Themes”. The crux of his argument:

The difference in interface that we’re talking about is a matter of the colors and shading of the widgets alone. We’re not talking about widgets that look the same and function differently. We’re not talking about widgets existing where you don’t expect them to be. We’re not even talking about widgets that are of different sizes — Aqua itself has widgets of multiple sizes.

This is precisely at the heart of my uncertainty as to whether these brushed metal issues really matter. Maybe it is just aesthetics — merely the color of pixels, rather than how the interface truly works — and I’m looking for serious implications that aren’t there, simply because I happen to think brushed metal windows look stupid.

I.e. that “consistency” is essential only when it actually affects usability, and that consistency for its own sake is not important.

This isn’t a retraction, per se, but an admission of vacillation. I just don’t know.

Postscript Offering an Actual Usability Gripe Against Brushed Metal

I’ve received more email this week from brushed metal aficionados than I have from gloating Red Sox fans. One common justification for defending brushed metal — beyond just liking the way it looks — is that it makes it easier to pick out windows from metal apps using Exposé, or when just a few pixels of a background window are visible behind the front window.

This argument doesn’t scale, however. The logical conclusion of this line of thinking is that every app should use its own custom window style (a la GarageBand).

And in fact, the issue of background windows is exactly where it’s easiest to make an actual usability case against brushed metal.

With regular windows (on Mac OS X 10.3), the active window has a title bar that looks different than windows in the background. Background window title bars are lighter, flatter, and have a subtle pinstripe. The active window title bar has a gradient effect, is darker, and more three-dimensional.

The texture of brushed metal windows is exactly the same when they’re in the foreground and background.

This is not to say it can’t be discerned when a metal window is in the background. The title text is lighter, and the close/minimize/zoom buttons are dimmed (as is the scroll bar, if there is one). The point is simply that it’s much harder to tell at a glance whether an unobscured brushed metal window is active or not.

This in turn makes it easy to mistakenly invoke the wrong keyboard shortcut. Here’s one I’ve run into personally. Let’s say I have a Safari window and an iChat window open side-by-side, with no overlap. I think the Safari window is active, but it’s actually the iChat window. I type Cmd-T to open a new tab in Safari, but instead, I get the Font Panel in iChat.

The active window is an essential concept in the Macintosh user experience. The behavioral distinctions between an active and background window are important enough that they should also be significantly distinctive visually. I’d even argue that the active and background states of normal Aqua windows should be more distinctive, but they at least offer much more distinction than brushed metal ones. (The window styles in every version of the old Mac OS were better in this regard than those in any version of Mac OS X.)

The point of this postscript is not to proclaim brushed metal’s indistinctive active/background states as a singular reason to abandon it, nor do I hold out any hope that it’ll change any minds regarding the current state of affairs. No, the point is simply that if there is a case to be made for brushed metal, usability certainly isn’t it.