I love pie, I hate pie charts

Reading through Tableau’s recent white paper, Which Chart Or Graph Is Right For You, I was heartened to see this explanation of pie carts: Pie charts should be used to show relative proportions – or percentages – of information. That’s it. Despite this narrow recommendation for when to use pies, they are made with abandon. As a result, they are the most commonly mis-used chart type. If you are trying to compare data, leave it to bars or stacked bars. Don’t ask your viewer to translate pie wedges into relevant data or compare one pie to another. Key points from your data will be missed and the viewer has to work too hard. AMEN! If you have to create charts or graphs, or interpret data for work or fun, you should review this white...

Time and money will be wasted

This weekend 80 to 100 coders will join others at the Seattle Hack The Commute Hackathon. The one thing I can guarantee that will happen is time and money will be wasted. Some might say I am being negative, but I firmly believe there is no software that can solve the commute and traffic problems that Seattle is experiencing. How much money? Well, 100 coders who command an average rate of $100 will blow through $160,000 in (free) labor across 2 8 hour days. Let’s be clear, the problem is not construction, the problem is too many people want to drive private and often single occupancy vehicles to and from work in the central business district.  The problem is also that there are not sufficient fast and convenient public transit options.  Software cannot solve these two problems.  We need to fund more transit options and create serious disincentives for people to drive SOVs into the CBD every day. Here is my free advice: make parking cost a minimum of $25/day in the CBD. That will modify behavior. Pass a transit package that will increase frequency and reliability of mass transit. Build more mass transit with dedicate, grade separated right-of-way. We don’t need more hackathons with largely homogeneous groups of coders and tech people trying to solve problems with software that actually require policy...

The Circular Timeline, My Data Visualization Nemesis

In a few recent presentations/trainings that I’ve given through my work at Resource Media on the topic of data visualization, I’ve pointed to one particular circular timeline visualization that drives me crazy. It’s the “sleep habits of geniuses” visualization that came out in the summer of 2014, and made the rounds on blogs and news sites.  I hate this visualization for the following reasons: It looks pretty but is not user friendly There is no significance to the color The viewer cannot easily determine the total number of hours of sleep for any one of the “geniuses” The area (and length) of each arc does not relate to the total number of hours, giving the visual impression that the people on the inner most part of the graphic have less sleep than the those on the outside of the graphic, even if they have the same amount of sleep. The visual analogy is of a clock face, but rather than showing only 12 hours, it shows 24 The white/black background is split 12 to 12 rather than 6 to 6 which would be more aligned with day/night. When I raised these critiques at the Seattle Tech4Good Meetup, one of my co-presenters, Ben Jones from Tableau pointed me at this Tableau Viz, which turns the data into more of a bar chart and adds some significance to the colors. I went a little less artsy and more utilitarian with my version of the genius data visualization. I was happy to find a specialist in data visualization and design, Alberto Cairo, voice some of the same concerns about another circular...

Thinking about Data Visualization

I’ve been getting deeper into data visualization, and will probably work on sharing more about it here on the site. But for starters, I wanted to share this quote I found in an article/book review about infographics (the author, Steven Heller, is no relation). This is the paragraph that I thought was best, and the emphasis is mine: Data visualization has been used as a visual shorthand in newspapers, magazines, and textbooks since the 19th century, if not earlier, Heller writes. But in the Internet age, infographics are more useful than ever. “Their popularity now has to do with the fact that we’re being bombarded by media and data, and there are so many different ways of addressing, analyzing, and serving that data,” Heller says. So often, this excessive information is conveyed sloppily, thoughtlessly, without enough attention to the reader’s experience. That’s what makes it so important to understand how deliberate infographic designers are about their process: many graphics look deceptively simple, but great visualizations aren’t whipped up in an instant; they’re planned impeccably, as these sketchbooks...

Share Vs. Rent

I am constantly surprised  (and a little dismayed) when an article that is ostensibly about sharing, and the sharing economy talks about renting. This quote,”You could start to equalize standards of living if you allow people who have a lot of stuff to comfortably rent out things to people who don’t,” from  Arun Sundararajan, and NYU researcher, is even more alarming. The article I lifted the quote from is on FastCoExist, “How The Sharing Economy Could Help the Poorest Among Us” How does it help the poorest among us for those who have stuff to charge the poor to use it. That sounds pretty much like the world we live in now. The paradigm shift (and I hope the quote was taken out of context) is when those who have stuff will lend it to those who don’t have stuff, ideally NOT for money. IE Share the stuff with people, not rent the stuff to people. Originally Posted on...