Data Visualization As Generative Narrative » rupa

out of sight, out of mind

rupa — Fri, 13 Nov 2009 05:14:05 +0000

1. Overview
My project will map the frequency of words related to the wars in Afghanistan and Iraq in relation to the number of American casualties in those two conflicts.

2. Data
NY Times Article Search API, TimesTags API, http://icasualties.org/

3. Design Questions

The shape of a graph of casualties in Iraq and Afghanistan will obviously be one that increases over time – sharply at first, then more slowly as time passes. My question is how much and until what point the number of
articles regarding the wars in the two regions follows the same slope. I would probably track these numbers month by month. I would also like to compare Iraqi civilian casualties, but this will be much more difficult. It would also be interesting to see the first mentions of an insurgency, as well as terms relating to withdrawal from the conflict.

4. Prior Art / Precedents

Most of the data visualizations I’ve seen about the war in Iraq have plotted the casualty count data against date, where the deaths took place, where the soldiers were from, and how their number compared to estimated Iraqi civilian deaths. I have not seen any that address the coverage of the conflict in the media or how it changed over time. I feel that this pattern makes the stories told by other data visualizations about the war that much more powerful and moving.

*Edit: Lie by Lie is an amazing interactive timeline of the events leading to the war in Iraq and asks who knew what when: our leaders, journalists, commentators, etc.

Mark Lombardi

rupa — Fri, 06 Nov 2009 15:31:28 +0000

(The artist Jane mentioned in class today – I posted this to my blog the other day, probably should have posted it here too.)

Looking at stuff like this all day….

really makes me appreciate stuff like this.

By the late conspiracy theorist / artist Mark Lombardi. Check out this great piece about him on NPR here: The ‘Conspiracy’ Art of Mark Lombardi

Timeswire Data Viz

rupa — Tue, 06 Oct 2009 17:26:13 +0000

Data viz using the New York Times Newswire API (swf)

The Newswire API is returning the last 20 updates to the Times website. My data viz is analyzing that data based on section & time submitted. Each update is signified by the small light blue boxes. Its value along the y axis signifies what category it falls into (based on what section heading it’s on the same line as) and its x axis value signifies the time when it was submitted (in relation to the earliest and most recent updates).

Data viz mockup using same data

This mockup is a play on TalkLikeWarrenEllis.com, which randomly generates phrases based on his tweets (example: “Good day, sleek piranha of impending doom.”) albeit much less hilarious. The idea is that it would take headlines from the TimesWire and split them up around punctuation marks and randomly generate sentences based on those pieces.

Data Types: Relationships (Bobby, Nasser, Rupa)

rupa — Tue, 22 Sep 2009 01:58:27 +0000

Presentation Outline

Reading Response #2

rupa — Fri, 18 Sep 2009 06:34:35 +0000

Although much of these two readings seemed to be rather ‘obvious’, I think they outlined very clearly what an information designer’s primary concerns should be – things that are very easy to lose sight of when trying to bend Processing or AS3 to your will in order to the make the most striking data viz. I particularly appreciated the portion about ‘preattentive variables’ – once again, the ideas were not surprising or new, but it was helpful seeing them quantified in such striking examples. Variations in color and size seem to be the most popular way to categorize data in most visualizations – the graphic was a reminder that other options exist as well. This was also true for the Patterns section in the Tidwell reading – I was reminded of underused visualization approaches (such as tree tables) and introduced to entirely new ones (data brushing). The Communicating Numbers whitepaper in particular made me re-evaluate past projects of mine and understand more fully why they worked, didn’t work, or could have been better. These readings may not have been particularly exciting or novel, but I’m sure I’ll be referring back to them regularly.

Data viz inspiration – Time geography

rupa — Sat, 05 Sep 2009 17:13:55 +0000

Space-time paths of women of color in Portland, Oregon - the vertical dimension represents time.

I came upon the concept of “time geography” while working on my major studio final project first semester (“This Post-Geographic Life”). Geographer Torsten Hagerstrand developed the concept to stress the importance of time in daily human activities. Mei-Po Kwan applied Hagerstrand’s space-time path to feminist geography, showing the disparate effects on gender on one’s daily travel patterns.

*Edit: Here’s a less dry data visualization. Flags of the world : What They Really Stand For.