The second point is particularly important with the pie chart because mousing over a wedge shows a quantitative breakdown of the comments on a particular subreddit: Pie charts generally donвЂ™t get a lot of love, but I use one to display total karma (and total comment length, and number of comments) by subreddit because it does a decent job of displaying the ratios and is simpler, more immediately understandable graph than the others. Like with the scatter plot, it is clear that the most commonly element вЂ“ this time the League of Legends subreddit instead of short posts вЂ“ occurs quite often and others less frequently, but the actual ratio between them is not clear. Still, while the distribution of commentвЂ™s length is apparent in the histogram, the distribution of comment length in each subreddit is difficult to discern. By removing the time component of the graph, the distribution of comment length becomes much easier to see вЂ“ the vast majority of comments IвЂ™ve made are quite short, even more so than the scatter plot shows. Instead, I use a totally different graph type to display the data:ДЇlot also does not include histograms, but sorting and grouping the comments then displaying them with the stacking plugin is simpler than creating a fisheye zoom effect within flot. Adding a heat map or fisheye zoom to the scatter plot could fix this problem, but neither are implemented in flot. Unfortunately, while it is easy to see that most of my comments are not very long, it isnвЂ™t very clear exactly how many shorter comments there are since the points cluster together closely at the bottom of the graph. I can see that IвЂ™ve spent most of my time talking about League of Legends, that I commented a lot in last July, and that IвЂ™m posting less frequently now. Mousing over a circle displays the comment it represents on the right panel.

On this scatter plot, every one of my comments is represented by a small circle plotted so its length position along the y-axis represents its number of (non-quotation) characters. Since redditвЂ™s user page only displays 20 comments at a time, it is very difficult to get a sense about how time has been spent on the site. The most obvious way to graph a set of data points is with a scatter plot. Using redditвЂ™s API, the program downloads a userвЂ™s comments and graphs them with flot. IвЂ™ve spent the last few days working on a visualizer for reddit comments.

Reddit visualizer program