In Novem­ber 2009, I did a mini-project together with Boris Müller and the boys from rau­reif. My task was to cre­ate a visu­al­iza­tion of the sur­vey results of an event. The par­tic­i­pants were asked to rate the events with respect to 9 ques­tions on a scale from 1–10. As we did not have much time (nor bud­get), we went for the first good-looking idea avail­able. What could that be? Right, a radial visu­al­iza­tion (be damned, cir­cles for non-circular data!). Any­ways, I pro­duced a quick funky mockup with ran­dom data: Each cir­cle sec­tor stands for one person’s rat­ings, and these are ordered by their aver­age rat­ing. For each sin­gle rat­ing, I draw a semi-transparent wedge, with dis­tance from cen­ter as well as color indi­cat­ing the rating’s value. Spe­cial treat­ment is pro­vided for the over­all event rat­ing (a more opaque, smaller wedge). For visual spice, a black spline con­nects all the aver­age val­ues of the ratings.

So, we agreed on it and shipped it. See­ing it with the real data, how­ever, made me won­der if I should have looked into typ­i­cal rat­ing sta­tis­tics a bit more :)

Well. Les­son learnt. It is a nice lit­tle visu­al­iza­tion nevertheless.

Which reminds me of an excel­lent arti­cle about how to pre­vent to uni­form votes already in the interface.

As a bonus, here is a lit­tle remake using pro­to­vis with again, ridicu­lously few lines of code: (more…)

Today was a good day, so I thought I would share its results imme­di­ately, instead of fine-tuning for­ever — who knows when I find the time anyways!

I built a lit­tle facet browser for the New York Times Arti­cle Search API - an impres­sively fast faceted search engine cov­er­ing over two mil­lion arti­cles. So, give it a spin!

Some caveats:

• Don’t look for the page nav­i­ga­tion — there is none. Pure lazi­ness, will update it soon.
• The ini­tial counts are based on a search for “the” (which I fig­ured would appear in all arti­cles). Unfor­tu­nately, only the top 15 or so val­ues per facet are returned, so you can­not click, e.g. the year 2008 in the begin­ning. Will fix.
• The API has a request limit of 5000 queries per day. So if your requests don’t work — come back tomor­row morning :)
• Unfor­tu­nately, the API seems to sup­port only one value per facet. So, all facets are single-select.(fixed, see comments).

The code is based on my totally revamped elas­tic lists pro­to­type. I used this project as a lit­tle sand­box exper­i­ment of how easy cus­tomiza­tion is pos­si­ble, and espe­cially how to make a switch from a fully client-based to a server–based fil­ter­ing model.

For the EU project MACE, I have been exper­i­ment­ing with hier­ar­chi­cal visu­al­iza­tions. Just the quick link for now, I hope I find the time to share some of the back­ground and find­ings later…

On a related note: 9 days left to hand in your papers and take part in a great con­fer­ence this autumn!