Anyone could make this screen look better. But it takes more than that to make it a good design.

I had an instant, negative reaction to the proposal. Not because of the design itself, which demonstrates a solid grasp of the information cab riders want and the context in which they’re consuming the information, but because the designers understood the design problem all wrong.

If the sole objective of Taxi TV is “providing useful information to riders,” then Antenna’s design is commendable. But that’s not the only–or even the primary–objective of Taxi TV. A 2007 Times article about NYC’s Taxi TV contracts confirms as much: Taxi TV is about advertising, not providing useful information. (Sidenote: WABC’s Taxi TV provides a pinpoint description of NYC taxi passengers, as only a marketer could—“upscale, affluent and captive”.)

It’s unpalatable, but Taxi TV is an advertising platform first and foremost, which means that Antenna’s proposal is a failure. The client—advertisers—would never agree to having their content off by default, hidden behind an “Entertainment” button. That design wouldn’t make it past the first client review (if even that far).

This is a perfect illustration of the tension designers face between providing the end user with the best possible experience and not getting fired by your client for doing so. For Taxi TV, the best user experience is clearly at odds with the business objectives. A good designer reconciles that issue instead of pretending it doesn’t exist.

How I Would Improve Taxi TV

I’ll put my money where my mouth is and offer a better proposal. Here are the design principles I’d follow, ones that I feel would result in a realistic and implementable design for Taxi TV’s interface.

Design to keep the screen on

From the advertisers’ perspective, nothing is worse than what I regularly do: hop in the cab, get instantly annoyed at the screen blaring at me, and immediately hit the OFF button on the screen. Of course, Taxi TV should allow users to turn the system off, but the content and the UI should do all it can to convince users that Taxi TV is worth keeping on. To do this, start the cab ride with a dashboard or overview of the things a passenger can do: see a map, get fare information, or watch TV. Don’t just give the user a choice, show them what they could experience with a small, live updating map like the one Antenna designed, a live (but muted) TV screen, and live updating fare information. Entice passengers to drill down into any of these three options.

Shut up and enjoy the ride

Leave the sound off until the passenger turns it on. This allows ads to be seen while decreasing the chance that a user immediately and instinctively turns the TV off when the ride starts. I would do my best to convince the client that muted ads are better than no ads at all. This could even be a useful design constraint for advertisers: make your ads compelling enough that passengers want to hear them. Or, to take it a step further, the NYC TLC could have structured the entire advertising model so that advertisers only pay for passenger interactions (as opposed to muted impressions).

Don’t do anything unless the user requests it

Taxi TV’s posture should be passive and reactive. Let the passenger approach Taxi TV and provide only what the passenger requests. Ultimately, this will feel like a less annoying experience to the passenger and makes them less likely to form a negative view of Taxi TV with repeated experiences. As a result, you have a passenger audience that is less hardened and more likely to see Taxi TV as a useful resource than an obtrusive annoyance.

Something happened on March 17, but you’d never know it by looking at Google Analytics.

What if Google Analytics let you place a simple marker on your timeline to help you remember events? It needn’t be any more complex than the dividend marker or news flags used to denote important events in Google Finance:

 

Two types of milestones in Google Finance.

A simple visual indication is all the work required of a Google Analytics milestone. I can manipulate the date ranges and data views as needed, on my own—no “milestone comparer” or “milestone manager” functionality necessary.

The idea came up at the time when I was completely bereft of ideas. I’d been working on my own music for a while and was quite lost, actually. And I really appreciated someone coming along and saying, “Here’s a specific problem – solve it.” The thing from the agency said, “We want a piece of music that is inspiring, universal, blah-blah, da-da-da, optimistic, futuristic, sentimental, emotional,” this whole list of adjectives, and then at the bottom it said “and it must be 3¼ seconds long.” I thought this was so funny and an amazing thought to actually try to make a little piece of music.

http://en.wikipedia.org/wiki/Brian_Eno#The_Microsoft_Sound

Note the subtle use of color to indicate pitch speed.

(via kottke)

Despite my frustration with needing to re-learn some CSS and Javascript, I’m glad I took the time to design and code the site myself. It was a nice reminder of the effort it takes to build a vision, an effort that I sometimes take for granted at my day job. I would never claim to be a graphic designer or a web developer, but spending time in both sets of shoes is, I think, an essential exercise for anyone in the user experience field.

Anyway, enjoy! Buy a book! Comments and criticism are appreciated.

Psychology postscript! My description of (and thinking throughout) the design process for this site triggered an association with my Myers Briggs personality type. The design effort for brooklynbookbinder.com definitely fits the description:

They regularly have to face the difficulties of bringing dreams into reality. The time and effort it takes is always more than what their intuition initially suggested. They are determined, perseverant, inspired and often see things just around the corner, into the near or far future.

Why didn’t I see this coming months ago? :p

There’s also a new hover interaction for names. It’s well designed visually–any time a designer puts several different pieces of functionality into small space but makes it appear uncluttered, she’s done something right. What one can do within this hover interaction, though, seems very limited. It’s prominent, striking functionality that I don’t anticipate using.

My favorite changes are also the most subtle. I love the clean lines and balanced padding around “More Actions” (but can we do something with that Refresh link? Ug-lay!).

GMail now uses a more muted yellow for messaging users about non-critical system activity. The previous orangish background was too loud for communicating system confirmation messages. It always gave me the feeling that something went kind of wrong, but not red wrong. The new yellow is softer and communicates the proper tone: “something just happened, but nothing is wrong.”

Conclusion: A solid if unspectacular update to the world’s best webmail client. sans comic sans gives the new GMail a B+, an 8.8 out of 10, and a check-plus-minus.

Error messaging on Basecamp (I think this is their 404 error, but I could be wrong).

Source: Playoff Prospectus: ALCS Preview, baseballprospectus.com

Which A’s player has the highest VORP? Who are the sluggers on each team? Are the players with high on-base percentages hitting at the top of the lineups? A very determined user can answer these questions with the data display given. HTML tables improve readability of sports statistics and increase the chances that people will discover trends and outliers within the data.

This is an easy solution that improves both the aesthetics and the usability of a data display. Now I want to know why Sean Casey is batting 3rd and Carlos Guillen 5th.