The R tutorial was written to do things in non-R ways and to only work with one of the R interpreters. I had not looked at it before tonight because I didn't realize there was an intent to package it other than exclusively with the sparkr interpreter.
An R tutorial notebook should be both (a) Interpreter-agnostic, and (b) Familiar for people who are coming from the R-world. E.g., we make html using HTML objects, we make images by calling image(), we make tables by calling data.frame() or table(), if we call an interactive visualization outside of a knitr-like environment, we expect it to open a new window for display, we don't use for-loops for vectorizable operations, etc. etc.