3 ms·

Posit (Rstudio) have been carrying out layoffs for the last few months, judging by the twitter posts of well known R developers looking for work, but the actual

Posit (Rstudio) have been carrying out layoffs for the last few months, judging by the twitter posts of well known R developers looking for work, but the actual numbers seem to be a well held secret. I can't tell if they're struggling or this is part of their wider shift away from an R focus towards datascience in general?

prepend 3y ago
Wouldn’t a shift to data science require more R people, not fewer? I worry that they’re struggling so the “data science” bit is just marketing scrambling to reposition the good stuff it already does.
- nhinck3 3y ago
  They're pivoting away from R which is a shame.
  - ATMLOTTOBEER 3y ago
    I did a considerable amount of R dev during an internship. It had a lot of breadth surrounding classical stats that python didn’t have. Other killer features were rstudio itself, ggplot2 graphing lib, dplyr data munging lib, r shiny for interactive graphs. It seems like python has surpassed it in all areas. When it’s time, it’s time. I hope this person lands on his feet!
  - prepend 3y ago
    That doesn’t really make sense to me because there are a million “data science” consulting firms. The only thing that made them special is R. If they try to just be a python package company it’s going to be rough times.
    - brnt 3y ago
      What's the difference? I'm still not actually sure where R sits these days but it is certainly my impression its confined to an ever smaller niche of social scientists and non-CSy folks. As more and more fields becme programmer-native, that niche becomes naturally smaller. Most, if not all, quantitativepy trained students now (and have been for a while) are bilingual in Python for glue and analysis and C++ for compute. R is also not a great fit for HPC, at least, they people attempting it are as few and far between as Windows supercomputers these days. Rstudio is arguably Spyder, and they seem well positioned to blow it away if they let go of R. I've worked with a few numerical fields now (physics, medicine and neuroscience, some of the largest consumers of academic HPC) and I can count the number of R folks without hands. Even friends at consulting firms and fintech, and they're also not using R. Or am I missing something?

prepend 3y ago

Wouldn’t a shift to data science require more R people, not fewer? I worry that they’re struggling so the “data science” bit is just marketing scrambling to reposition the good stuff it already does.

nhinck3 3y ago
They're pivoting away from R which is a shame.
- ATMLOTTOBEER 3y ago
  I did a considerable amount of R dev during an internship. It had a lot of breadth surrounding classical stats that python didn’t have. Other killer features were rstudio itself, ggplot2 graphing lib, dplyr data munging lib, r shiny for interactive graphs. It seems like python has surpassed it in all areas. When it’s time, it’s time. I hope this person lands on his feet!
- prepend 3y ago
  That doesn’t really make sense to me because there are a million “data science” consulting firms. The only thing that made them special is R. If they try to just be a python package company it’s going to be rough times.
  - brnt 3y ago
    What's the difference? I'm still not actually sure where R sits these days but it is certainly my impression its confined to an ever smaller niche of social scientists and non-CSy folks. As more and more fields becme programmer-native, that niche becomes naturally smaller. Most, if not all, quantitativepy trained students now (and have been for a while) are bilingual in Python for glue and analysis and C++ for compute. R is also not a great fit for HPC, at least, they people attempting it are as few and far between as Windows supercomputers these days. Rstudio is arguably Spyder, and they seem well positioned to blow it away if they let go of R. I've worked with a few numerical fields now (physics, medicine and neuroscience, some of the largest consumers of academic HPC) and I can count the number of R folks without hands. Even friends at consulting firms and fintech, and they're also not using R. Or am I missing something?