Using the location field in Twitter accounts, we try to approximate the geographic location of up to 5,000 users and map them for you. This is an inexact procedure and so some points may be mapped incorrectly.

This chart breaks down the hourly Twitter activity of pennjillette's friends.

This chart details pennjillette's Twitter activity. Using up to 400 of their most recent actions, we chart when they tweet, @contact tweet (ones that begin with someone's name), and retweet. Compare to the above chart to see how their activity compares to their friends.

To help make sense of the "biography" field of each Twitter user, we've assembled this word cloud which shows you the most frequently occuring words.

Similar to the above word cloud, here we show you the relative frequency of words used in the "location" field of users pennjillette follows.

Social Authority is our rating of a user's influence and engagement on Twitter. It ranges from 1 to 100, where higher scores indicate a person with greater influential activity.

Based on an analysis of users' profile data, when applicable and available, we guess their gender.

We've analyzed users pennjillette follows and assembled this chart to show you the number of people who follow those users. This is a useful way to pinpoint those users with the most followers (and therefore clout).

Here we highlight the following counts of users pennjillette follows. Generally, this is a less useful metric than the follower count (above). That said, we've found that people who follow a large number of users are often less "valuable" than those who follow fewer – this is because those people who follow a lot are less likely to comprehensively read their Twitter stream.

The age of an account is an interesting way to measure value. Users who have been on Twitter for a while are early adopters, and generally more valuable than newbies. As well, there is a higher percentage of spam accounts in newer user segments (as Twitter has not yet cleared them out).

If a person hasn't tweeted recently, then their account may be dormant. These data are based on the date that Followerwonk last checked the user, so the dates may not be entirely precise, but they should give you a good idea of who is tweeting more recently than others. Protected accounts are excluded from this analysis.

If a user has a low total tweet count, they may not be a "real" user. Perhaps they tried Twitter and gave it up? On the other hand, those with a very large number of tweets might be avoided (they could be spammers, overwrought, or just too verbose!)

Please note that Twitter only allows users to select from a few languages. As well, many Twitter users who tweet primarily in a foreign language may still have their official language set to English.

How many URLs do these folks tweet as a percent of their total tweets? Note: this chart is restricted to a smaller sample of Twitter users for whom we compute extended metrics.

Retweets are an excellent measure of engagement. The more someone retweets, the more he is consuming other peoples tweets. That's generally a good thing (because it means they are more likely to RT you). Here, we breakdown this audience by the amount that they retweet (as a percentage of their timeline). Note: this chart is restricted to a smaller sample of Twitter users for whom we compute extended metrics.

Similar to the above chart, we analyze how often this audience @contacts another person (a tweet that begins with an @mention). This score is the amount they @contact someone as a percent of their timeline. Note: this chart is restricted to a smaller sample of Twitter users for whom we compute extended metrics.