RFC - e-mail in tough environments

Sergey Matveev stargravesm at gmail.com
Thu Dec 10 13:37:14 UTC 2009


On Thu, Dec 10, 2009 at 01:15:24PM +0000, Peter Lewis wrote:

> >     - *Line break of e-mails* Should be around 72 characters. Nobody
> >       will kill you if it is 70 or 74 or even 76. But a lot of people
> >       will get angry if you do not have a line break at all.
> Can I ask what the reason behind this is?

That is because of email systems was created when most users have
standart terminals without any GUI. The standart terminal's workspace
size is 80x25 (24?) characters. No more, no less. It is standart.

As mk@ wrote, email was used mainly by technical people in places like

Modern email clients like Mutt, Gnus, Mailx, Pine, etc of course use
terminal too. And it is really VERY ugly to read messages with lines
longer that 80 characters. 72 is some kind of "protective"-layer between
text itself and standart terminal's width of 80 characters. You know --
line numbers in Vi(m), line borders of email client, maybe too long
words and so on.

Line printer's standart width was 80 characters too.

> I really don't like static word wrapping, as much as I don't like static width web pages. I read emails on a variety of devices and window sizes, and the presumption that ~72 characters is appropriate for wherever I'm reading it seems rather arbitrary.

Web was never forced to be used on standart terminals -- it was GUI
oriented, so of course it must be dynamic anyway.

> A good client should have dynamic word wrapping IMO.

If you are using system that is 30 (40?) years old -- you should know
it's rules. Even RFCs will tell you about 80/72 characters issues.

If it will use dynamic wrapping, then noone can guarantee you that you
read EXACTLY what I send you. In HTML you have to use <pre>-tag for
what, all email always was used oriented (repeating again) on terminals
and so on, that is why it should not do anything with body (more or

Happy hacking, Sergey Matveev                   () ASCII Ribbon Campaign
FSF Associate member #5968 | FSFE Fellow #1390  /\ Keep it simple!

More information about the Discussion mailing list