I've made up my mind about the things that I listed yes - and a bunch of other use cases where Excel is better; but I'm not saying 'Excel is better' or 'R is better'. What I'm saying is that proper applications of Excel and R are (almost) orthogonal; the number of Excel users who should 'upgrade' to R is very small, compared to the total population. R is a statistics tool, it competes with SPSS and maybe some uses of SAS. For those uses, sure use R; use R when teaching statistics. But saying R is some natural, 'better' progression of Excel is myopic and not understanding the use case of Excel.
Not a month goes by that I don't try to switch some Excel sheet I have into R; I run into the problems with Excel every day. It's not like I don't know Excel isn't great. My point is that the solution to those problems isn't R. I don't know what is, but I know it's not R.
Like, the other day I tried to do a real estate investment analysis in R. I always do that in Excel; I build the model step by step, starting with some basics, then filling in the details of the case at hand as I go. All the time, I can focus on the numbers; adding an indicator or refinement is part of the natural workflow. In R, you always have to switch between 'code mode' and 'data mode'. In Excel, you don't have this difference. Which is at the same time also its weakness, of course.
Sharing R models is a pain in the ass. Others have to get the exact packages, you have to tell them what to look at out of all the variables, ... Shiny sucks for that, it's read-only. In Excel someone makes some changes and sends you back the sheet, boom done. In R you would set up a versioning repo, data is split over multiple files, packages need to be installed, scripts for reporting, ... Fine for big projects, not for the small analysis exercises that make up the bulk of the uses of Excel.
Sure you 'can' blog in R. Last week I didn't want to walk to the shed to get my hammer, so I used a brick to pound in a nail. Doesn't make it right. String manipulation in R is a joke. A function to concatenate strings? Please. Again, R is fine for statistics, but it's not a general purpose replacement of Excel.
Not a month goes by that I don't try to switch some Excel sheet I have into R; I run into the problems with Excel every day. It's not like I don't know Excel isn't great. My point is that the solution to those problems isn't R. I don't know what is, but I know it's not R.
Like, the other day I tried to do a real estate investment analysis in R. I always do that in Excel; I build the model step by step, starting with some basics, then filling in the details of the case at hand as I go. All the time, I can focus on the numbers; adding an indicator or refinement is part of the natural workflow. In R, you always have to switch between 'code mode' and 'data mode'. In Excel, you don't have this difference. Which is at the same time also its weakness, of course.
Sharing R models is a pain in the ass. Others have to get the exact packages, you have to tell them what to look at out of all the variables, ... Shiny sucks for that, it's read-only. In Excel someone makes some changes and sends you back the sheet, boom done. In R you would set up a versioning repo, data is split over multiple files, packages need to be installed, scripts for reporting, ... Fine for big projects, not for the small analysis exercises that make up the bulk of the uses of Excel.
Sure you 'can' blog in R. Last week I didn't want to walk to the shed to get my hammer, so I used a brick to pound in a nail. Doesn't make it right. String manipulation in R is a joke. A function to concatenate strings? Please. Again, R is fine for statistics, but it's not a general purpose replacement of Excel.