in User Interface

Smart design vs Too-smart design

I love gmail as much as the next guy, but sometimes even the ui wizards at Google seem to have fallen for that most tempting of traps in user interface design – trying to be smart. Wait a second you say, isn’t ui design about being smart! ok, not to get bogged down in semantics here, but what I mean by this is when the overly eager designer tries to make one too many assumptions about a user’s behavior, essentially painting their design into a conceptual corner. For example, the designers at google decided that if someone posts a message to a list, and you are on that list, that you would not want to also be a recipient of that same message once it has been posted. Really? I actually like to see that my message went through.

But ok, let’s say that I’m ok with this. Unfortunately, the real problem with trying to be smart are all those pesky unexpected scenarios that you hadn’t consider. Ever tried sending an email to another of your addresses that is in turn forwarded back to the gmail address? Well, gmail sees that it was sent from your address and therefore decides (apparently) that this was someting you posted to a list and quietly does away with the message. Same with the otherwise brilliant gmail alias feature. Am I being overly picky? Are these realy non-issues? (How often do you need to test your own email address?) Possibly. At the same time, not showing you an email that you sent to yourself is non-standard behavior. For that reason, even if it is a good idea, it is something that the user should be able to decide that they want to do. The first time gmail does this, it should ask the user if this is their preference, and provide information on how this preference can be updated. This provides the best of both worlds, rather than mysteriously lost email because some designer decided that they know how their users behave…