Dashboard’s Damage
I have a love-hate relationship with Mac OS X 10.4’s Dashboard.
On the one hand, it’s nice to have the weather, a calculator, a calendar, my upcoming iCal events, the time in DC, a fast way to post to del.icio.us, a sticky note, a unit converter, a dictionary/thesaurus, and the Ruby core library reference no more than a keystroke or hot corner away.
On the other hand, I had all of these things before. While I appreciate Apple’s strict and clear Dashboard programming guide for what it suggests widgets should be, it isn’t strict enough.
Widgets should not be task-redundant. The world does not need hundreds of widgets that do nothing but fetch and display an RSS feed. This is Dashboard’s major offense: it encourages people away from well-rounded solutions and towards brittle and overly-simplistic ones. A weather widget is good; people need to glance at a weather report. A Google Maps widget is bad; why break off a subset of that exceptional site’s features into a constrained viewable area when the whole thing can be a bookmark away?
Widgets are defined in contrast to applications in the aforementioned programmer’s guide, but in reality many widgets are bite-sized chunks of web applications. Apple should add a new strong suggestion to that document: don’t duplicate any functionality that’s better served by a real website.
Dashboard is also in desperate need of a seamless widget update mechanism. Widgets downloaded from Apple’s directory should be upgraded automatically at a user-configurable interval; you should “subscribe” to a widget, in effect. Keeping up with a dozen widgets on top of my arsenal of applications is a bear.
The feature’s memory consumption is criminal. Thanks almost singlehandedly to Dashboard, Tiger does not feel faster than Panther on my amply-equipped PowerBook G4. This is the first downgrade in perceived performance from release-to-release since the earliest versions of OS X.
Overall, it’s the sort of gimmicky feature I wouldn’t have expected Apple to bring to market. I hope the next iteration of OS X focuses on more substantive improvements. In the meantime, I keep toying with the idea of turning the damn thing off.