From the “blogosphere”? Hardly.

January 27, 2010
By

(This article was first published on What You're Doing Is Rather Desperate » R, and kindly contributed to R-bloggers)

I generally skip over “From the Blogosphere”, a (mostly) weekly-summary of one or two blog posts in Nature’s “Authors” section (here is the latest). Why? Well, I’ve always suspected that the title is rather misleading. Now, I have the hard numbers to prove it.

My feed reader contains an archive of 128 articles, dating back to May 10 2007. I used them to create this CSV file with 3 fields: Date, Blog and Source. The “Blog” field contains the name of the primary blog mentioned in “From the Blogosphere”. On rare occasions where 2 blogs are mentioned, 2 entries were added to the CSV file. I skimmed each article rather quickly, so I may have missed blogs that were mentioned in passing.

Next, a short R script:

library(ggplot2)
# read file, remove rows with no blog or source
posts <- na.omit(posts)
# plot by blog name
png(filename="blogs.png",width=480,height=640)
qplot(posts$Blog,data=posts,xlab="Source",ylab="Number of mentions",geom="bar",main="By Blog") + geom_bar(fill="blue") + coord_flip() dev.off() # and plot by source name png(filename="sources.png",width=640,height=480) qplot(posts$Source,data=posts,xlab="Source",ylab="Number of mentions",geom="bar",main="By Source") + geom_bar(fill="blue")
dev.off()


Result:

As I suspected, “From the Blogosphere” should really be called something else. “From the NPG stable”, perhaps? Only one blog post outside of Nature or Nature Network received a notable mention: Egon’s review of Nature Chemistry.

Given NPG’s excellent blog tracking effort, Nature Blogs, I suggest that “From the Blogosphere” be retired. Or at least, renamed to something less misleading.

Posted in humour, R, statistics Tagged: blogosphere, nature, nature network