Enough with Themes Already
March 14, 2001
This goes far beyond changing colors and fonts. Themes can drastically change the appearance of an application, and even change the functionality. Proponents tout them as a way to help people bond with their software.
I have a proposal for what to do with most themes: get rid of 'em.
Themes began with media players such as Winamp and RealJukebox, and now seem to be spreading, kudzu-like, into other application categories and operating systems themselves. The ability to theme is an in-demand feature that can differentiate a piece of software. I guess bollixing up your media player so your friend can't even find the volume control is seen as hip and cool.
In an article in the October 2000 issue of Wired, Jef Raskin scoffs at this notion: "The role of software is to make the functions of a computer more accessible. That role is enhanced by uniformity."
You tell 'em, Jef! Consider driving a car. The statement, "I know how to drive a car," has meaning that "I know how to use a computer" currently lacks. Cars have standard controls that allow someone who has driven one model car to hop behind the wheel of a different model and drive away. The areas that are not standardized are non-critical ones. (Coincidentally, or perhaps not, a major non-standardized area in cars is the "media player" -- that is, the stereo controls.)
Meanwhile, knowing how to use a computer merely means that you have enough experience with computers that you can usually bluff your way through whatever situation arises. The "Start" button may be a good idea, but if the user doesn't know to click on it, it isn't much help. Just moving the toolbar from the bottom of the screen to the side would probably generate a help desk call from half the office workers in the country.
In the same Wired article, the owner of a Web site that stores raw bitmaps for use by themers is quoted as saying, "Think about cars in the era of the Model T. Everyone drove clumsy black boxes, and they loved it -- until someone decided to be original. Soon afterward, they had different colors, different shapes, dashboard options, convertible roofs, musical car horns, and bumper stickers."
This misrepresents how radical a change themes are. What he describes is the customization that graphical operating systems have always had -- the ability to change the background bitmap, window colors, what applications are installed.
Instead, imagine a car manufacturer describing its "innovative" new car, which allows you to switch the location of the pedals around. Now imagine your grandmother hopping in the car for a quick trip to the store, and you will get a feel for how disruptive themes are to a computer novice.
Some people say that computers have become so standardized that they are just a commodity, whose main functionality can be captured inside an "appliance," thus rendering the fully flexible PC obsolete.
But consider how much PC interfaces have changed. Microsoft radically changed the GUI in 1990, again in 1995, and now is talking about more changes in Windows XP. The Macintosh is going to redo its GUI with OS X. Linux has multiple GUIs. Until all this is standardized, the idea of being "able to use a computer" will have little meaning, and anyone who tries to capture the current GUI concepts in an appliance will soon find it obsolete.
Themes have their place. Skinz.org has over 15,000 downloads available -- anything that gets people that excited about computers can't be all bad. If someone wants to theme a game, more power to 'em. Themes can be useful for testing out new looks, and some good ideas about how to improve the user interface have emerged from themes. But theming key applications or the operating system itself is a step backwards.
Perhaps we are witnessing the last paroxysm of flexibility before a new, consistent user interface emerges from all the conflicting models out there. Maybe themes will get so out of control that users will revolt and demand a brave new world of consistent interfaces. But if all themes are going to do is make a virtue out of obscurity and delay progress, then I see them as a plague that must be stopped.
Adam Barr worked at Microsoft for over 10 years before leaving in April 2000. His book about his time there, "Proudly Serving My Corporate Masters," has just been published by iUniverse. He lives in Redmond, Washington.