As one of the creators put it, easy tools for creating data visualizations (Excel, Tableau) aren’t always very expressive. Tools like D3 are very expressive — but you have to write lots and lots of code. It’s hard.
Enter Lyra. Lyra puts a GUI on top of D3, and when you’re done, generates the code. It also puts out a flat JSON file in a format known as Vega. The cool thing about that? Vega is highly reusable. You can reuse a visualization simply by feeding it a different dataset. Cool.
Alexander Chen has created this amazing live data visualisation using MTA data. He calls the project Conductor.
The app tracks live trains as they leave stations on the NYC subway, applying rules to their paths that allows them to cross. Each time they cross they play chord. Creating an orchestra of audio art.
7 billion world
7 Billion World displays 7 billion people together on a single webpage. Developed by Worldometers - which themselves were originally posted in the good year of 2005 -, the web page itself is generated through some small programming code, yet is claimed to be 1 mile (1.6km) high and 800 feet (250m) wide, which is both horizontally and vertically scrollable.
Agnes Chavez creates algorithmic drawings from data, like this drawing of a forest. See more here.
Business on the right, party on the left.
WTFVIZ collects examples of data visualization that R DOIN IT RONG.
D3Plus: An extension to the D3 library that allows fast and easy creation of popular visualizations.
Recently, I attended a workshop by Alex Simoes and Dave Landry on D3Plus, an extension for D3.js they wrote that makes creating D3 visualizations quickly and easily. I was really blown away with the power and simplicity of this extension: I highly recommend it. You can find the library here and here. I pushed some sample code to a repository here. I’ve got my own tutorial in the works; stay tuned.
It would be badass if D3.js got easier to use.
This is one part of my Table2Chart project. The aim of it is to visualize html tables automatically. It decides which chart type is the best choice and for drawing I am using Chartjs and some additional pull requests. It’s open source => start to code :)
Wonderful data journalism. Look how southern California calls Colorado (where many residents from CA migrated to in the 90s and aughts), New England calls Florida (I live here and I don’t know what that’s about, but that might be because I don’t have my AARP card yet); and, most touchingly, how Louisiana and Alabama call Michigan. Many African-Americans migrated from the south to Detroit in the 40’s and 50’s, paving the way for Motown City.
The Connected States of America
Are our borders really the edges of our communities? The “internet guy” in me says “of course not” but that doesn’t really take into account how much of our day-to-day interaction takes place in geographical meatspace. But on the other hand, many of America’s state borders are very arbitrary delineations of latitude or since-bridged rivers, so how meaningful are they in 2013, really?
What would our borders and communities look like if we looked at other data, like phone calls? At Krulwich Wonders…, Robert Krulwich has taken a look at a couple of alternate “neighborhoods”.
The photo above was assembled from anonymous mobile phone data by MIT’s Xiaoji Chen, and it which regions call each other the most often. Anyone who’s been to my neck of the woods in Austin knows that Texans don’t call people in Oklahoma much (or College Station, for that matter), and the NorCal/SoCal split shows that the differences there go beyond suntans and dotcoms. And people in the Plains apparently just want to call anyone they can that doesn’t live in the Plains.
“What’s it like out there? Just grass here.”
Check out the rest of Robert’s post for more phone fun, plus a little look at how (not) far our money travels (and what that says about us).
THIS IS GOOD GOOD WORK.
Chicago Public Schools (CPS) has marked 129 elementary schools for possible closure based on utilization and performance. Closing a school is a very disruptive decision, affecting student outcomes, parental confidence, and neighborhood stability. In addition to showing the data being used by CPS, this website aims to help school leaders, parents, and communities learn more about the schools being considered, and what options will be available in case of closures.
Back at the Hacks/Hackers Media Party in Buenos Aires, I announced the creation of Code Sprints—funding opportunities to build open-sourced tools for journalism. We used Code Sprints to fund a collaboration between WNYC in New York and KPCC in Southern California to build a parser for…
Source/Open News is getting *really* interesting.
Vega lets you automatically generate visualizations by editing a JSON file with the data.
Chicago Transit Authority president Forrest Claypool had some biting words for the Chicago Sun-Times — on its own pages.
On Thursday, the CTA chief penned a letter to the editor, chastising the newspaper’s article on CTA crime that ran on Tuesday.
Read this, it’s amazing.
A collection of online data journalism resources compiled by the Tow Center.
#NICAR13 is already paying off. Thanks to a few sessions from a few brilliant people, I decided to try using R to evaluate bail data. Unfortunately, all these boxplots show me is that our data was screwed up when they changed the field type in 2004, which is just another fun problem to solve…