Question: Is this really true?
Surely this would not work if the Hebrew text was encoded in Unicode? cf. The filter seems to work only at the byte level.
It doesn't appear to be an implementation of the bidi algorithm!
What was intended by this suggested application?
Other languages are written RTL, such as Arabic and Farsi. Can the Reverse each line filter be of any practical use for these?
Or was this just a mistaken idea in the help file? i.e. a pipe dream [pun intended]
