A few days ago I had the classic "out of free space on drive C" error on my main computer. I wanted to find what was taking up space so I could cull the waste from my system. My friend Anson Horton pointed me to a tool on the web called SpaceMonger. Within a few minutes I had downloaded the application, installed it, and found the wasted space on my hard drive. Within five minutes I had freed up 33% of my drive.
SpaceMonger is based on a classic UI break through called treemaps. Back in school, we studied treemaps as part of an interface design class. SpaceMonger uses treemaps to give you a visual overview of the files on your hard drive.
Treemaps are a useful tool for visualizing statistics. It utilizes our ability to more quickly absorb statistics in a visual, rather than numeric form.
When dealing with he wetware in our head, the bottom line is that our visual tools can be more powerful than our more abstract grasp of symbols such as numbers. For instance, it is easy for us to look at 500 objects and see which one is largest. It takes us longer, however, to look at 500 numbers and pick out the ones that are largest.
In Figure 1, you can see SpaceMonger's view of my hard drive after I cleaned it up. Items colored pink are on the root of my drive. Those color light orange one level down from the root. Yellow colored items are two levels down from the root, and green items are three levels down from the root.
Figure 1: A treemap of the used space on my Inspiron 9400 running Vista.
After glancing at the image, it immediately becomes apparent that the largest file on my system is -- appropriately enough -- a video of Eric Lippert. For most people, it would take longer to garner the same information from an alphanumeric file listing that is several pages long.
Treemaps were created by Ben Schneiderman, a professor at the University of Maryland. He had his epiphany while trying to find a way to visualize a hard drive. The SpaceMonger program is therefore very much in the spirit of his work.
Schneiderman and other researchers have since used treemaps for a variety of tasks that involve the need to visualize large datasets. Examples include tracking sports teams and their players or tracking stock portfolios. In the latter example, size was used to show the value of a stock, and color to show whether it was increasing or decreasing in value. For instance, a bright shade of red might be appropriate for a bull stock that is growing quickly, while a dark blue color could represent a bear stock that was losing value.
SpaceMonger is very configurable. In Figure 2 I've changed the setting so that I can see both the parts of my hard drive that are being used, and those parts that are unused.
Figure 2: In this image you can see the 32.7% of my hard drive that I freed up using SpaceMonger.
If you zoom in on a particular area, you can often see a surprising amount of detail. Looking up at Figure 1, near the bottom center left, you can see a small directory called "Music." Figure 3 shows what this same area looks like when you zoom in on it.
Figure 3: Zooming in on the Music directory. Click on the screenshot to see the original image.
You can expand the screen area of the image shown in Figure 3, or you can zoom in on a subdirectory within the image. For instance, clicking on a particular directory will zoom in on it until it occupies most of the available screen real estate found inside SpaceMonger.
SpaceMonger is interesting because it hints at how our ability to readily create and display visual images can supercede more established technologies such as the alphanumeric data found in text documents or spreadsheets. The spread of podcasts and videos hints at the possibility that the printed word may not be as important to future generations as it is to us. Will there come a time when only a small but very powerful priesthood of technological experts will understand abstract symbols such as numbers and letters?