3 Rules of Thumb When Designing Visualization

Posted to Visualization  |  Nathan Yau

Bernard Kerr, the lead designer for del.icio.us, gave an interesting talk (below) focused on remail (mentioned here) and tagorbitals. At the end, he offers three important lessons.

Reduce Multidimensional Data

After showing many thread arc versions, Kerr says that when you are dealing with multidimensional data, pick two variables; otherwise, you're going to end up with a big mess. He says this literally, but don't forget that you can also reduce dimensionality with super special and magical statistical methods.

Use Real Data

You won't know what you're really dealing with until you have the real data. You can spend lots of time guessing what the data are going to be, but it's the real data that will eventually drive your design. This goes for statistics too. Real data leads to real analysis.

Try Adobe Illustrator

Adobe Illustrator offers a javascript interface, so try that out before opening Processing or Flex Builder, and programming through the midnight hours. Illlustrator is of course also good for static mockups and brainstorming. My work flow usually starts with paper and pencil, to Illustrator, and then to the programming. Some people go straight to code, but that's never worked well for me.

What rules of thumb do you follow?

Here's the talk in full. It's pretty interesting, if you've got about 25 minutes to spare.

[via infosthetics]

8 Comments

  • Interesting…I start to wonder what’s going to look like in new del.icio.us interface

  • Interesting stuff.

    The key points I work from are
    – understand your own goals, and what you’re trying to communicate
    – understand what the needs and contexts of your audience are

    I believe that those are the most relevant considerations when designing good visualizations. He’s coming from an experimental place, so it’s all research and all about exploration, which makes the second point moot.

    With a clear idea of what your priorities are, the issue of multidimensional data can reduce itself. You start with figuring out how to represent the most important parts, and the rest, ideally, falls into place or becomes irrelevant.

    People interested in techniques for effective information visualization might enjoy my thesis.

    Best, Noah

  • Interesting stuff.

    The key points I work from are
    – understand your own goals, and what you’re trying to communicate
    – understand what the needs and contexts of your audience are

    I believe that those are the most relevant considerations when designing good visualizations. He’s coming from an experimental place, so it’s all research and all about exploration, which makes the second point moot.

    With a clear idea of what your priorities are, the issue of multidimensional data can reduce itself. You start with figuring out how to represent the most important parts, and the rest, ideally, falls into place or becomes irrelevant.

    People interested in techniques for effective information visualization might enjoy my thesis.

    Best, Noah

  • Great stuff. Any idea where I can find more videos like this?

  • all the talks from the event can be found at: http://www.vimeo.com/user378630

Favorites

Real Chart Rules to Follow

There are rules—usually for specific chart types meant to be read in a specific way—that you shouldn’t break. When they are, everyone loses. This is that small handful.

Top Brewery Road Trip, Routed Algorithmically

There are a lot of great craft breweries in the United States, but there is only so much time. This is the computed best way to get to the top rated breweries and how to maximize the beer tasting experience. Every journey begins with a single sip.

A Day in the Life of Americans

I wanted to see how daily patterns emerge at the individual level and how a person’s entire day plays out. So I simulated 1,000 of them.

Interactive: When Do Americans Leave For Work?

We don’t all start our work days at the same time, despite what morning rush hour might have you think.