seems to have passed all tests (I'm skeptical there still may be problems)

  Specifically with the following two issues...
  1) hanging on R code blocks when we have multiple R sessions in play
  2) very occasionally the results of ruby code blocks are iffy

This whole session experience proves the general point that functional
code is easier to write and debug than code with state.

org-babel.org

@@ -2060,6 +2060,8 @@ of these tests may fail.
 | get from ruby session   | get-ruby-session-var       |     |           3 |           3 | pass |
 | set python session      | set-python-session-var     |     |         set |         set | pass |
 | get from python session | get-python-session-var     |     |           4 |           4 | pass |
+| set R session           | set-R-session-var          |     |         set |         set | pass |
+| get from R session      | get-R-session-var          |     |           5 |           5 | pass |
 #+TBLFM: $5='(if (= (length $3) 1) (progn (message (format "running %S" '(sbe $2 (n $3)))) (sbe $2 (n $3))) (sbe $2))::$6='(if (string= $4 $5) "pass" (format "expected %S but was %S" $4 $5))
 
 ** basic tests