Yes, precisely. Still this is a very interesting research and you can see clusters of languages like the most popular ones. I just noticed that people often tend to look at a chart and jump to quick conclusions, so just wanted to add this "grain of salt" :)
And judging the relative popularity for less popular languages such as Common Lisp is even less precise if we base the research only on github and stackoverflow. So I would be very careful with this comparison of language popularity. Although it is better to have some chart that none :)
Check out my article "Measuring Language Popularity is Harder Than Many Think" http://smthngsmwhr.wordpress.com/2012/11/19/measuring-popula... and I must say that both github.com and stackoverflow.com are not just typical sites and getting data only from those sites is not enough. JavaScript is very popular at github.com and C# at stackoverflow.com Other sources give different results as well.
We may be more or less sure only in 10 or so most popular languages and judging relative online popularity may be hard for them.