michaelgalloy.com

The Ebb and Flow of Movies graphic at The New York Times displays the box office sales (adjusted for inflation, of course) for popular movies over the last twenty years. I think this is excellent work, but I wanted to see long-term patterns that are easier to see when zooming out (as in this thumbnail).

Found via FlowingData.

The Parsons Institute for Information Mapping
InfoVis site (”1100+ examples of information visualization”) does have many great examples of InfoVis. You can search or browse through the images (although browsing is done by where the graphic comes from, not the content of the graphic). The site accepts submissions.

Found via SimpleComplexity.

I have updated the Brewer color tables that I posted last fall. The download includes mg_loadct and mg_xloadct, which are wrappers around loadct and xloadct that use the Brewer color tables instead of the default color tables.

The update fixes the qualitative color tables. These color tables are for labeling, so they contain 3 to 12 colors instead of a full set of 256 colors. The qualitative color tables each have multiple versions included, depending on how many colors are needed. For example, “Pastel1″ has versions with 3, 4, 5, 6, 7, 8, or 9 colors.

Continue reading “Updated Brewer color tables.”

I saw Jeffrey Heer present his work on animated transitions at IEEE Vis 07. I think this idea has a lot of promise, but I wonder when it’s worth the extra complexity to produce the animation. For most ad hoc analysis, this would not be worth it. But it might be useful for the big presentation that determines next year’s funding.

Found via FlowingData, the movie is a useful short introduction.

Google is planning to host a lot of scientific data. Below are the slides from Jon Trowbridge at Google (my favorite part is comparing the transfer rates of FedEx to Ethernet):

Depending on the API Google provides, it would be convenient to have IDL bindings capable of grabbing parts of this data.

Link via FlowingData.

I finally got a copy of Information Visualization by Colin Ware and started reading it. Information Visualization and Readings in Information Visualization edited by Stuart Card, Jock Mackinlay, and Ben Schneiderman are considered the classics in the field of Information Visualization (or InfoVis).

In Information Visualization, Ware evaluates visualizations using the scientific methods of psychology. Perception, vision, color, etc. are discussed and used to reach conclusions for creating visualizations.

The section on various flow visualization techniques was very useful since I’ve been thinking about line-integral convolutions (LIC) lately. There were also discussions of choosing colors, using texture maps, 3D shading, and a few plot types that have changed the way I will do some types of visualizations in the future.

In summary, I would say this is a useful book if you are serious about improving your visualizations. (Colin Ware has a new book coming out this April that looks interesting as well.)

Jon Udell interviewed Fernanda Viegas and Martin Wattenberg, creators of Many Eyes, recently about social visualization. I’ve posted about Many Eyes before. I also saw them at the IEEE Vis 2007 Conference last fall.

I read the following blogs regularly (via RSS) and they usually have interesting examples that make me think a bit about the visualizations I’m currently involved in. All have a fairly low volume of posts, but that fits well with their content.

1. Charts, Junk Charts, and Flowing Data all have a lot of great examples of practical visualization problems (with good and bad solutions).
2. Visual Business Intelligence is Stephen Few’s blog. I heard his talk at IEEE Vis 2007. Dealing with the business community, he has a lot to say about bringing visualization to “less sophisticated” users (than the scientific visualization commnunity).
3. EagerEyes is Robert Kosara’s blog, who I also saw at IEEE Vis 2007. He posts some interesting examples and evangelism of information visualization.
4. Ask E.T. is a forum discussing topics of interest to Edward Tufte.
5. Visuale contains some in-depth analysis of current information visualization topics.

I was hoping to speed up the code I mentioned last month implementing a line integral convolution (LIC) flow visualization, but I haven’t had time to improve it (and probably won’t for awhile). I would like to eventually write this as a DLM, but until then here’s my pure IDL draft. It contains a main-level program that is an example of using the main routine mg_lic.

Let me know if it helps or you make any improvements.

Google released a seriously cool charting API today. The data and options about its presentation (not as many as IDL, but quite a few) are encoded into the URL of the image. Of course, there are plenty of chartjunk options, but it seems quite possible to make some serviceable graphs. The one thing I wasn’t able to was turn off the axes in order to make a sparkline. The charts shrinks to the right size , but I couldn’t remove the axes totally (nor make them transparent, set their color to the background color, etc.).

UPDATE: With help from Jonathan in the comments, I can now make a true sparkline . Also, check out Jonathan’s cool Google Charting Tool where you can easily create the line plot that you want.