🚀 go-pugleaf

On Wed, 16 Nov 2005, Masanori HATA wrote:
> I have a question about the number of characters that each of
> multiple-line header field that contains one or more 'encoded-words' is
> limited to 76 in RFC 2047 (section 2) (or in its former version 1522).
> Why is it 76 characters? Why isn't it 78 characters?

An encoded-word is limited, by definition, to 75 characters.  With a 
single continuation whitespace character, the result is 76 characters.

I remember when this was discussed nearly 15 years ago.  There was 
testimony that certain email gateways could not be trusted not to break 
lines with a larger limit, and that limiting the lines to 76 octets was 
the only way to be sure that would not happen.  Line breaking by an email 
gateway would be disasterous if it happened to an encoded-word.  I don't 
remember the details, but I'm sure that the working group discussion 
archives are available someplace.

RFC 2822 was produced 10 years after MIME's decision was made.  The 78 
character recommendation in RFC 2822 is to avoid line break behavior on 
24x80 type displays, as opposed to worrying about email gateways that do 
line breaking.

It's also quite possible that many (most? all?) of the email gateways 
which caused the problems in the early 1990s are no longer an issue today.

-- Mark --

http://panda.com/mrc
Democracy is two wolves and a sheep deciding what to eat for lunch.
Liberty is a well-armed sheep contesting the vote.