London Underground UI Fail

I mentioned this on Twitter the other day, but I was so incensed I just had to scribble out a blog post too. It’s a classic case of replacing a traditional (in this case paper-based) “interface” with an electronic one, and in the process failing to translate the fundamental functionality that made the original system useful.

How can you screw-up creating an electronic version of an existing sign?

Here’s the offending item:Underground line status
It’s a replacement for a combination of magnetic labels and hand scrawled messages that convey the current state of each of the London underground lines. They’re generally one of “good service”, “minor delays”, “severe delays”, or “oh no, you’ll never get to work”.

For the majority of people, their only interaction with this system is walking past, early in the morning, when they’re not at their sharpest, and trying to figure out if their tube line is working. As such, if there was a magnetic label stuck on your line, you knew things weren’t good. Given the lines themselves are colour coded, it’s easy to determine which line you’re interested in. Bear in mind that people aren’t going to get closer than about 2 metres to the board.

And herein lies the problem: by replacing the analogue board with a digital version, it’s no longer possible to ascertain by the physical presence of a label or writing, what state the line is in.

Let me say that again. It’s virtually impossible to tell, even from relatively close, if there are delays. Take a look at that screenshot (dodgy quality as it is) and see if you can tell which line has “minor delays”. Yeah, not easy is it? Now try doing that as you’re briskly walking past at 7.46am.

Underground line status (detail)

Underground line status (detail)

It’s not helped by the fact that the state descriptions are all roughly the same length and form, “good service” could just as easily be “severe delays” when it’s in 36pt and seen from 2m away. There have been many studies on the legibility of type, most in the context of road signage, and of course, there are the seminal texts (e.g. The Elements of Typographic Style) on how typography can be applied for good, not evil. Shame no-one responsible for this application seems to have read them.

I’m surprised that they’ve relied solely on text and haven’t used any other means of distinguishing the significant information. Some other things that could’ve been used:

  • Colour (of the text itself, the background, or some indicator. Need some inspiration? How about taking a look at something way out there, like, say, traffic lights)
  • Size
  • Weight (using font weight to indicate relative importance)
  • Icons (or some other kind of visual indicators, nothing flashy)

I’m sure there are a whole load of advantages to having this digital system; most notably that it’s connected, and can be automatically and centrally updated. If only they’d spent any time on thinking about how people were actually going to use it!

Harry Beck would be turning in his grave.

This entry was posted in Rant, Usability and tagged , , , . Bookmark the permalink. Post a comment or leave a trackback: Trackback URL.
  • Marc

    Excellent post. I see similar things happening all the time. Generic committees create checklists, procurement takes these and finds the cheapest contractor. Managers know more about management procedures and career options than the workings of business they are responsible for. Users have no way of knowing who is responsible and who to address. It’s very discouraging to see these collective achievements being lost forever without anybody noticing. Thanks for documenting this case before it’s value is lost forever.

blog comments powered by Disqus
  • Follow me on Twitter Follow me on Twitter @voyce

  • Categories

  • Archives