Earlier quoted context omitted.
Thank you for weighing in, John Nagle! I understand that constraint, and it seems reasonable - but in that case, why not use a length prefix? That should be even more efficient than having to scan for a line containing a single period and nothing else.
Because many paths of that era were not binary-transparent. There was CR-LF to LF conversion, the possibility of ASCII/EBCDIC translation, and other transformations. Hence such horrors as MIME delimiters: -----=_Part_2827761_1947716067.1716352583971 Content-Type: multipart/alternative; boundary="----=_Part_2827760_1372041171.1716352583736" We still have the mess that is the required and standardized behavior of HTML5…
That's genuinely horrifying, but it explains why SMTP made a decision that I would otherwise categorize as "insane" - it was forced to by constraints of the time. Thank you for explaining!