diff --git a/SC2031.md b/SC2031.md
index 50c28cf..218bfe7 100644
--- a/SC2031.md
+++ b/SC2031.md
@@ -18,7 +18,15 @@ n=0
 while read i; do (( n+=i )); done < <(printf "%s\n" {1..10})
 echo $n
 ```
-In `sh`, a temp file (better if fifo or fd) can be used instead of process substitution. And if it's acceptable to do it with waiting, try Here Documents.
+In `sh`, temporary files, FIFOs or file descriptors can be used instead. When the output of the command can be stored to a variable before entering the loop, here documents are a preferable alternative:
+```sh
+n=0
+SUMMANDS="$(printf '%s\n%s\n%s\n%s\n%s\n%s\n%s\n%s\n%s\n%s\n' 1 2 3 4 5 6 7 8 9 10)"
+while read i; do n=$(( n + i )); done <<SUMMANDS_INPUT
+$SUMMANDS
+SUMMANDS_INPUT
+echo $n
+```
 
 
 With Bash 4.2+ you can also use `shopt -s lastpipe` which will change the pipe behaviour to be similar to Ksh and Zsh (see Rationale below) [as long as job control is not active](https://www.gnu.org/software/bash/manual/html_node/The-Shopt-Builtin.html) (e.g. inside a script):