Hey Nick!

 NB> Was the second one with fold or just "set trimblanks" disabled?

It was 'fold -s -w 69'.  The first is nano with "set trimblanks" and "set fill 69".

 NB> Which one do you think is doing it incorrectly?

Neither, they are true to what they're called up to do.  Offhand I suspect that fold is more DOSsie and it is usually one character shy of how others define the length of a string.  For example FTN standards list the subject field at 72 characters when it is really 71.  The 72nd character is \0, which is ye' olde string delimiter and in the case of MSGs it is a field delimiter.  Offhand I'd go with fold's specifiers and have .nanorc "set fill 68" which is one less than fold but produces the same end result as 'fold -s -w 69'.

Life is good,
Maurice

-o   -o    o-  -o
 (\   (\  /)    (\
 ^^   ^^  ^^    ^^
... Man by¨ on myrg¨e his magan leof; sceal ¨eah anra gehwylc o¨rum swican.
    Man in joy is dear to his kin; but they will have to part.
--- GNU bash, version 5.2.37(1)-release (x86_64-pc-linux-gnu)