Is R Making a Programming-World Comeback?

For the past few years, the narrative around the R programming language, which is used heavily in data science, has remained much the same: Although academia and specialized data-science firms used R pretty heavily, Python was rapidly eclipsing it as the language of choice for all things data-related.

However, the latest update of the TIOBE Index suggests something incredible: That news of R’s demise has been premature, and the language might be making a bit of a comeback. Specifically, R has jumped up to eighth place on the Index, up from 20th place a year ago.

What’s behind this surprising rise? “There are 2 trends that might boost the R language: 1) the days of commercial statistical languages and packages such as SAS, Stata and SPSS are over,” TIOBE wrote in a note accompanying the data. “Universities and research institutes embrace Python and R for their [statistical] analyses, 2) lots of statistics and data mining need to be done to find a vaccine for the COVID-19 virus.”

In order to generate its rankings, TIOBE leverages data from a variety of aggregators and search engines, including Google, Wikipedia, YouTube, and Amazon. For a language to rank, it must be Turing complete, have its own Wikipedia entry, and earn more than 5,000 hits for +”<language> programming” on Google. That methodology has attracted its share of critics over the years, who argue that the rankings are more a measure of these languages’ “buzz” (and SEO juice) than actual usage.  

That being said, TIOBE is a useful way to monitor languages that are potentially on the rise (and fall). And for a long time, it seemed that R was falling. Way back in 2018, for example, a KDnuggets poll of technologists who used both R and Python showed a slow decline in R usage in favor of Python. At around the same time, a separate survey from Burtch Works revealed that Python use among analytics professionals grew from 53 percent to 69 percent over that same two-year period, even as the R user-base shrunk by nearly a third.

“R has issues with scalability,” Enriko Aryanto, the CTO and a co-founder of the Redwood City, Calif.-based QuanticMind, a data platform for intelligent marketing, told Dice at the time. “It’s a single-threaded language that runs in RAM, so it’s memory-constrained, while Python has full support for multi-threading and doesn’t have memory issues. When choosing a language, it all comes down to choosing what’s best to solve your problem.”

Meanwhile, Python continued to slither its way deeper into the data-science arena. “Behind Python’s growth is a speedily-expanding community of data science professionals and hobbyists—and the tools and frameworks they use every day,” GitHub stated during its 2019 edition of the State of the Octoverse. “These include the many core data science packages powered by Python that are both lowering the barriers to data science work and proving foundational to projects in academia and companies alike.”

But while Python remains immensely popular, both in data science and programming as a whole, it seems you shouldn’t count out R just yet; it’s clearly still drawing attention and usage—whether or not COVID-19 has anything to do with it.

In the meantime, if you’re new to Python and want to learn its ways, check out, which offers lots of documentation, including a useful beginner’s guide to programming in it. Once you’ve learned some core concepts, focus on writing faster code (via Functions, Lists, and more), debugging, and other more advanced skills. Microsoft also has a video series, “Python for Beginners,” with 44 short videos (most under five minutes in length; none longer than 13 minutes); it recently added even more content, including “More Python for Beginners” (20 videos), which covers key concepts such as managing a file system and asynchronous operations, and “Even More Python for Beginners: Data Tools” (31 videos).

7 Responses to “Is R Making a Programming-World Comeback?”

  1. Okay, they guys saying Python doesn’t run in memory and is multi-threaded unlike R doesn’t know what he’s talking about. It’s trivial to run R in parallel, and where does he think Py runs? On disk? Also, look at’s benchmarks, data.table in R is way more scalable than pandas.

    The problem R faces is that there are a lot of very vocal, very ignorant python fan boys trash talking it all the time. Like, go back to chasing down indentation errors, my guy.

    • Dhen Phu

      so true… Python doesn’t multi-thread very well at all. It is slow by itself, and the only merits of it can be summed up by serious weaknesses, actually: it has a large fanboy base, who don’t know any other language, can’t really program, and just throw some shit against the wall to see what sticks. The other merit is that a lot of libraries (mostly C and C++ written, actually) have been wrapped in it. So in fact, you’re not looking at python-based software really, but just at python-wrapped software, written in actual languages, not the crap toy called python. Anyways… those who see python as the best thing after sliced bread, are usually not the people who’s opinion is worth anything in programming: they unmask themselves as the fanboys of one of the worst languages thrown together by a Dutch guy with no talent for programming at all. That being said, sometimes crap technology wins from good technology, because there are enough crap users who know nothing else. meh. It will go away sooner or later, because it will collapse under its own inconsistencies and flaws, which are so abundant it hardly stands a chance.

    • You entirely miss the point. With libraries any language can do anything. But no library can conceal the fundamental, profound ugliness and awkwardness of R code. R is basically a patched together, slipshod language with no underlying aesthetic sensibility. WTF is “alsist[[1]][2]”? That is a language I would never ask for a dance. RStudio is an elegant patch on a profoundly broken chassis.

  2. Charles Smith

    R and Python are becoming more and more interoperable thanks to reticulate. You can now do a project in R Studio using both languages. For example, data wrangle with R, then use Python for the machine learning part, then go back to R for ggplot2 visualizations. The best of both worlds.

  3. asiwel

    Heck, I like Python and have done a lot with it. But mostly it is interpreted and while it has excellent mathematical and numerical libraries, IMO they are not documented or developed the way, for instance, SAS is. It does make a good Notebook tool. I also like R – naturally since I grew up on FORTRAN – and definitely think it is a serious tool. You can do fMRI analysis and other difficult stuff with R. Mostly today, I use C# for scientific and commercial programming though. So any languages learned and forgotten, it is nice to have a safely managed, well-documented, frequently updated, sort of “tame” language to think in.

  4. Hazem Abdelazim

    After many years in programming since the early lisp / COBOL/ Fortran . I still believe R is much more elegant than python in terms of syntax, in fact the most elegant and fast (in coding cycle) language I have encountered. But I have to switch to python unhappily due to large amount of very good AI/ML courses in python only in th last period.