Parallel Sets

We are happy to announce the release of Parallel Sets 2.1. The new version fixes a number of bugs and introduces a few new features. The biggest changes are under hood, with a new database model that can now handle much more complex datasets (in terms of number of dimensions and categories), and the new streaming import can load in datasets with a much larger number of records. We have also added a way to automatically sort categories by name and size, a screenshot function, and more. Upgrading is strongly recommended.

We are looking for people who use Parallel Sets with real-world data and who would be interested in taking part in writing up a case study for the InfoVis Discovery Exhibition. You don't have to share your data, and the images can be anonymized, but it has to be real data (and you have to be able to describe what it is and what you learned in a way that is not too general).

After an initial commitment, an announcement, and a delay, we are proud to announce that the Parallel Sets application has been released! Mac OS X and Windows versions are ready to be downloaded, and the source code is available for your enlightenment and/or entertainment.

<

p>Parallel Sets (ParSets) is a visualization application for categorical data, like census and survey data, inventory, and many other kinds of data that can be summed up in a cross-tabulation. ParSets provide a simple, interactive way to explore and analyze such data.

I'm very sorry, but we have to delay the release. We need to make sure that we have the right to release the program, and this is taking a bit longer than expected. The software was written in part by students paid from the SRVAC, which is funded by DHS and DOE. It is unlikely that there will be a problem, but we don't have the go-ahead yet, either. I am hoping that we can get the green light this week.

We promised the release of the Parallel Sets program for the end of April, but we have to push it to May 7. The reason is quite simply lack of time: the semester is ending here at UNC Charlotte, and things are rather busy as a result. There is also the nuts-and-bolts work of making this work as an application for end-users that works well on at least two platforms (Windows and Mac OS X), and that can provide useful information when it doesn't.

A few days ago, Michael Blastland published a very interesting little article on the BBC Magazine website, in which he showed an image of Parallel Sets. He also mentioned my name and even though there was no link or email address, I have seen a spike in traffic on this website and my university page, and have gotten a barrage of emails from people who want to try it out. Here is some information on the technique and a plan for getting the tool out into the world.