It’s time to fix the window controls
This is awkward. We need to talk.
March 3rd was a strange day. It was one day before User Interface freeze for the upcoming Ubuntu Lucid long term support release. By the end of the day we were supposed to have the entire look and feel of the desktop settled on so people could start writing documentation and books.
This was the same day that Canonical finally released the new theme that had been under secretive development. It was bold, daring, light-inspired, and perhaps most popularly, not brown. Jono Bacon, Canonical’s community manager, broke the news. Mark Shuttleworth followed up on his own blog, thanking three members of the design team for leading the effort.
The most important change, however, wasn’t actually talked about. The designers don’t blog themselves, and Mark and Jono didn’t mention it directly. It had to be found in the screenshots, or experienced firsthand by alpha testers.
This is not ok
Soon, the community learned the change was intentional: a bug about the misplaced window controls was quickly marked invalid, and when the controls briefly reappeared on the right again the change was reverted. What’s disturbing is that Planet Ubuntu has been rather silent on the topic. No one’s posted a real defense of this change yet, or for that matter even claimed responsibility. It’s like there’s this collective unease about criticizing something that feels like it came directly from on high. So, instead, people are just silent. I’d certainly be if I worked for Canonical. Perhaps I should be, as I still hope to work for them.
If you read between the lines, you can tell that people aren’t too happy about it. The most flattering thing a developer’s said about the left-sided Window controls is that they “got used to them after a few days”. We’re quick to praise the theme (it’s gorgeous), but talking about this major sudden change to the window controls feels like taboo. That’s incredibly unhealthy for a community project. It’s like there’s this collective unease and everyone’s worrying if we’re about to release something embarrassing.
This experiment was a failure and we need to realize it
The alpha releases are great places for usability experiments. Sometimes, they don’t work out. Put a new user on today’s Ubuntu Lucid and they’ll think it’s fantastic, sleek, and absolutely gorgeous right up to the point where they have to close a window. That’s where our first impression becomes something awful.
Note: The new card backs pictured above are my doing and are now default (Mads Rosendahl drew them).
A brief summary of the complaints about the left side window controls
Some of these I noticed myself, a few are gathered from various comment threads on forums and blogs over the past week.
• Because the window title isn’t centered, the window controls being placed directly in front of it put it in a weird indented position
• The “slightly off left” location is inconsistent with Nautilus, Firefox, Thunderbird, Pidgin, Empathy, and every other tabbed program we have, which have close buttons for their tabs on the right.
• The left position is inconsistent with Windows, previous versions of Ubuntu, and even OSX – users have to relearn decades of muscle memory.
• Users who interact with both Windows and Ubuntu machines (or migrate from Windows) will have a much harder time than they did before.
• The buttons are too close to the file and edit menus, making catastrophic misclicks much more likely. Closing something on accident should be as rare as possible.
• Even without misclicking, a user will have to take more time to use the window control and avoid a misclick. This is an example of Fitt’s Law.
• The close position is also inconsistent with the power button in upper right. Currently, “close it down” is something you can always do from the upper right anywhere in the system: within a tab, within a window, and even for the whole computer. The new window controls break that entirely.
• The new position leaves a lot of empty, wasted space in the upper right of most windows. While strictly speaking the amount of unused space is the same, it looks much worse when it’s all clustered together. When the controls are on the right, the extra space can function as a buffer for the potentially destructive window controls.
• Similarly, the upper left of most windows now becomes much more crowded, creating a rather unpleasing contrast to the relatively empty upper right.
• In previous Ubuntus you could close windows on the left if you really wanted, by expanding the small circle menu that’s now gone entirely. File->Quit is also an option, which is now very close to the close box.
• Gnome upstream has them on the right, causing consistency and developmental problems when we deviate. This is particularly jarring with the adoption of future projects like Gnome shell and Gnome 3, which will change again how we interact with window controls.
• The current implementation breaks themes not designed for the new button order (which is currently every theme we ship, so even changing the theme back doesn’t help)
• A day before User Interface freeze of a long term support release is the worst possible time to suddenly spring this on everyone without explanation.
• It is very difficult to change them back as we don’t have any UI tool for doing this (the current method is manually editing gconf keys)
• The new position doesn’t actually do anything beneficial.
That last point is the most important. Other than “looking different” the change doesn’t do anything helpful. It’s a huge usability loss for an awful lot of people. Some people get used to it quickly. Others don’t, and like me end up getting physically angry when trying to use their computer. I can’t remember ever having my computer make me feel this way for a long time, and I’ve been running Ubuntu alphas for five years now.
Let’s admit we have a problem
Ivanka deserves credit for being the first member of the design team to at least talk about the controls:
Are we smoking crack to think that the learning curve or getting used to a new position is ever going to be worth any real or perceived benefit of new positions?
I’ll leave that question to the reader.