The ColorBrewer website interface has been updated, cleaning up the appearance and adding a few new features. One feature available in the web interface not available in the Brewer color tables for IDL is the ability to filter on colorblind-safe, print friendly, and photocopy-able color schemes.

I recently came across this post on Visuale collecting some great resources about using color effectively. I have posted about some of these before (such as the Brewer color tables) and meant to post about others (such as How NOT to lie with visualization, but this puts some great references covering the topic of color all in one place.

50 Great Examples of Data Visualization shows some great visualization eye-candy. Link via Datavisualization blog, which I’m finding to be the data visualization blog I read the most along with FlowingData.

I saw this post about a Top 10 list of programming fonts that might be of interest to other programmers. The comments offer a few more good suggests, too. I checked them out, but still use anti-aliased 12pt Monaco.

9-12% of men have some form of color blindness. When producing visualizations, it can be useful to simulate what the colors used in your visualizations look like to people with various types of color blindness. The two that I use are Color Oracle and Sim Daltonism.

Color Oracle is available for Windows, Mac, and Linux. I have only used the Mac version, but it works well: select the color blindness type you want to simulate from the menubar and it temporarily shows a simulated version of your display. It can simulate deuteranopia, protanopia, tritanopia.

Sim Daltonism is for the Mac only. It can simulate more types of color blindness than Color Oracle. It works by creating a new window which shows the area around the mouse with the selected type of color blindness. This is more useful for interactive visualizations.

There are also online tools available if you don’t want to install any software: Vischeck can convert images or URLs.

I mentioned horizon graphs previously, but didn’t give the code for producing them. Here’s the VIS_HORIZON routine (docs), but you will need VIS_LOADCT and the Brewer color table file to run the example program i.e. .run vis_horizon. UPDATE: you also need VIS_SRC_ROOT.

I’ve found horizon graphs are a good choice as an alternative to using OPLOT many times. The different line plot series quickly become crowded where the horizon graphs don’t interact as much and still maintain a fairly good ability to be compared to each other.

Horizon graphs divide a plot into bands that are collapsed into a smaller space and distinguished by color. In the below plot, dark blue represents the largest positive change, medium blue a moderate positive change, light blue a small positive change, and red is divided the same way for negative changes.

A more detailed explanation is available at Panopticon site. In particular, follow the sequence of diagrams halfway down the right side of the page and their captions.

Found via Stephen Few’s blog.

Another great interview on FLOSS Weekly, this time interviewing Ben Fry and Casey Reas about Processing. Also, I listened to, but forgot to mention, the interview of David Buck and Chris Cason of POV-Ray last February.

By the way, here is a fairly simple visualization I recently did as an exercise of using Processing.js, a port of Processing to JavaScript. I created the background image in IDL and then used Processing.js to give the interactive annotations. I just bought “Visualizing Data: Exploring and Explaining Data with the Processing Environment” by Ben Fry and hope to do more with it in the future.

Unearthed Outdoors is making the 250m True Marble global dataset available for free. GeoTIFF (9.6 GB) and PNG (6.67 GB) versions available. The dataset consists of 32 tiles of 21600 by 216000 pixels each, making a full 172800 by 86400 pixel image.

True Marble by Unearthed Outdoors, LLC is licensed under a Creative Commons Attribution 3.0 United States License.