« Because it's Friday: Frosty, too frosty | Main | NYT uses R to map the 1% »

January 16, 2012


Feed You can follow this conversation by subscribing to the comment feed for this post.

Another place to look at is GitHub. On Dec 12, R ranked at #29 (I have a record on Google+ https://plus.google.com/u/0/109653178371807724268/posts/Bnt4523ELMm); today it ranks at #28 (https://github.com/languages/R). One step up in a month.

Good point, Xie Yihui. Thank you for that link. I like to follow popularity of programming languages on GitHub too. But I was tracking the enduring presence of FORTRAN, mostly for amusement purposes, now at #37 of approx 73.

Tracking GitHub R usage on a monthly basis is an excellent benchmark for comparison-validation w.r.t. the TIOBE index. The difference in rank, #28 versus #19, does not concern me. The rate of increase in popularity over time will be more significant.

I think Java is number 1, not ojbective C.

I'm not convinced by the argument that R is a domain-specific language and thus not likely to break into the top 10. I'll grant that R is going to be used mainly for data analysis, and hence domain-specific in that sense. But everyone has data.

R is a great substitute for a lot of what is done in Excel (an accounting-specific language).

You're correct, thanks Frank. Objective-C was the fastest riser. I'll update the post.

Pat, good points. I'd love to see R in the top 10 myself, and I agree as more programmers focus on data-centric applications its usage (and rank) will rise further.

The methodology is a little suspect, but the rankings are mostly plausible. Except for Logo. I have no idea how it jumped so many positions.

If R is ever packaged in a way it can compete for the run of the mill spread sheet user it could grab a lot of users. I don't know how much productivity would be lost by users tinkering writing the same tools over and over and trying to make the best graphical presentation rather than a very good one.

I have been trying to bend my mind around R for about 18 months to use for both data and models.
R can do things that most languages are used for as long as one is willing to use Java for human interfaces and C, C++, BASH shell scripts, HTML, system calls and such as needed. R makes those interfaces with other language and systems much easier than most language if it really is a language. I use it almost the same as would a Linux/Unix command line at times.

I can see doing point of sale, accounting and inventory control programs in R using either Java or HTML as human interface.


The comments to this entry are closed.

Search Revolutions Blog

Got comments or suggestions for the blog editor?
Email David Smith.
Follow revodavid on Twitter Follow David on Twitter: @revodavid