jump to navigation

A Depressingly Good Plot August 17, 2009

Posted by gordonwatts in computers.

Long time readers of this blog will known that I love graphics – and think we need to use visual data representation better in particle physics than we currently do (heck, one of the main reasons I started working with DeepTalk). A friend of mine here in Marseille pointed out this fantastic one from Slate:

image What you are looking at is a static version of the real thing. Seriously, go to the article, and click on the green play button. I’ll wait. You’ll be depressed. You might recognize maps like this from the election. Each box is every single county in the USA. Here the blue circles represent the # of jobs gained each month. The red ones the # lost. And the size of the circle is proportional to the actual number. As you might imagine as the animation makes it way into 2008/2009 the map turns rather red.

There is a lot of data being shown here. But in about 20 seconds you can get a good idea of what is happening to the US job market in the last 2.5 years. So, while a depressing plot, what a great way to show the time evolution data! It has very high information density.

In particle physics most of our plots are static and simple histograms – very low information density. The problem is making high density plots like this is very work intensive. And we’d need the tools do do it quickly. Do such tool kits exist?



1. Anonymous - August 18, 2009

I don’t know about this particular graph, but it is similar to Gapminder. There’s an interesting TED Talk about it by Hans Rosling. I’m not aware of any other similar tools, if you find them please blog about it.

2. Anonymous - August 18, 2009

I wonder if anyone has a job left in Detroit.
Seriously, and perhaps more pleasantly, perhaps Viewpoints (a NASA project) may be of some help. I believe some Babar folks are thinking of possibly using it for presentations. It can be found at http://astrophysics.arc.nasa.gov/~pgazis/viewpoints.htm

3. Isaac - August 22, 2009

Versions 6 and 7 of Mathematica have made it a choice tool for the rapid production of information- dense graphics. Especially when you just want to view the information, but are not interested in toiling away over micro-editing the plot for publication.

One advantage is the ease-of-use of the Import/Export capabilities, since even an excellent graphical tool is useless if it is a pain to load in the data.

Another advantage is that, in programming termnology, Mathematica treats graphics as literals of a basic type (graphics). This means that almost anything which can be done to nubers, symbols, or strings can also be done to graphics, which means that generating lists of graphics and displaying them dynamically and/or embedding them in one another is done using the same kind of constructs (Lists, Tables, For Loops) as one would use for calculations with numbers.

The final, and perhaps most important, point is that Mathematica has smart and stylish options for the default settings, which saves programmer effort (as when generating dynamic graphics (e.g. 20 plots which link together in time).

Check out the Wolfram Blog or the Wolfram Demonstrations Project.

P.S. I know Mathematica would never replace specialized tools like ROOT, and I only half recommend it for publication level graphics (when the defaults are inadequate, it can be a pain), but where I really recommend it is for scientists who want to make quick plots for their own purposes.

4. John Terning - September 11, 2009

Hans Rosling’s Gapminder software was aquired by Google and parts of it are availabe as free widgets, see for example:

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: