Comment by not_the_nsa | Hacker News Reader

Comment by not_the_nsa | Hacker News Reader

Parent Full thread

Not mentioned are the fantastic user communities, especially the culture of inclusiveness and openness fostered by RStudio ([code of conduct](https://github.com/tidyverse/dev-day-2019/blob/master/CODE_O...) and [rOpenSci](https://ropensci.org/community/). Basically the inverse of SnarkOverflow.

Especially rOpenSci's peer review process ([more here](https://devguide.ropensci.org/softwarereviewintro.html)) for R packages is fantastic.

I do most data engineering in R (RMarkdown workbooks), and most software engineering in Python/Django. It took three separate, dedicated attempts to get warm with R (pre-tidyverse, showing my age), now I'm interrupting work on an RShiny app to write this comment. The ecosystem around the tidyverse helps immensely to convert my colleague's workflows from Excel to R. Clarity and simplicity wins over purity here (you may now light your pitchforks). And NSE still breaks my brain.

I think the community that R users have managed to create is it's strength. Python has a community of software developers, while R has a community of people who use programming to solve their problems. It's a fundamental difference in mindset, one which really shapes the community.

What also helps is that R is so focused on data and statistics. It gives a focus to the users that really helps when it comes to finding help. Python is famously second best at everything, but that also means it's community is spread thinner over more subjects.

It's this same reason that running anything that depends on R in production is a PITA.

I run R in production and its absolutely fine and wasnt harder than pretty much any other thing in software development.

I didn't suggest it was hard. I found CRAN repos to be insecure and unreliable.

I am not a R user (using primarily julia and python), but can you expand on the insecure aspect of CRAN. Do you refer to (potentially) missing package signing (similar to [1])? I am not aware that python or julia support this either. Or is the software download over ftp/http instead of https?

[1] https://wiki.debian.org/SecureApt

I too got started pre-tidyverse. I've done some minor analysis the last two years and been blown away by how easy it is to get up and running with that code. Way easier than it used to be. I actually bumbled my way through building a simple report building system pre-1.0.0. It was horrible in comparison.

I've gotten some adoption of R Studio at two companies now. It's amazing for exploratory analysis and its cloud capabilities are wonderful.

Reply on news.ycombinator.com