4 ms·

Hey now, Hadley does really great work, IMO he's the main reason R is still worth using. I started using R about 10 years ago, but after learning Python I came

Hey now, Hadley does really great work, IMO he's the main reason R is still worth using. I started using R about 10 years ago, but after learning Python I came to really dislike R for how ugly it is. But I still kept using it for plotting because ggplot is so damn good. In the past year I've started to use R more again and have started to kinda like it. It's still unpleasant for general purpose programming, but for manipulating and plotting dataframes, dplyr/tidyr/ggplot is very clean and way nicer than pandas/matplotlib. So yeah, I think R has improved a lot since I first learned it, mainly due to Hadley, and is actually a pretty good language within its particular domain. But if I had to use it for general-purpose stuff I'd probably come to hate it again.

crispyambulance 9y ago
Yeah, I've been using R for about 7 years now. I find it frequently unpleasant and the documentation often makes me want to stomp on puppies, but in the end, it NEVER let me down. Hadley's Tidyverse (and RStudio) has breathed new life into R, but I wish that stuff was available in other languages. NOTHING comes close to ggplot, for example, for composing sophisticated plots with compact but expressive code. dplyr for data wrangling fits like a glove. After you've wrestled your data to submission, you then have a vast amount statistical analysis you can do out of the box using R libraries and little effort. I don't want to like R, but I can't avoid it.
hadley 9y ago
You should also check out http://purrr.tidyverse.org http://purrr.tidyverse.org and http://github.com/r-lib http://github.com/r-lib - we're working towards providing tools that make general purpose programming more pleasant. (Not because R should ever be your primary general purpose language, but many data analyses often involve some general purpose components)

crispyambulance 9y ago

Yeah, I've been using R for about 7 years now. I find it frequently unpleasant and the documentation often makes me want to stomp on puppies, but in the end, it NEVER let me down. Hadley's Tidyverse (and RStudio) has breathed new life into R, but I wish that stuff was available in other languages. NOTHING comes close to ggplot, for example, for composing sophisticated plots with compact but expressive code. dplyr for data wrangling fits like a glove. After you've wrestled your data to submission, you then have a vast amount statistical analysis you can do out of the box using R libraries and little effort. I don't want to like R, but I can't avoid it.

hadley 9y ago

You should also check out http://purrr.tidyverse.org http://purrr.tidyverse.org and http://github.com/r-lib http://github.com/r-lib - we're working towards providing tools that make general purpose programming more pleasant. (Not because R should ever be your primary general purpose language, but many data analyses often involve some general purpose components)