Masataka Ohta writes:
I am sure that MIME works, and a lot of other schemes would work also,
like Otha-san's, or some ISO 10646 encoding
UTF-7 is just as bad as quoted printables.
'
Yes, but it works between two UTF-7 users.
Raw 10646, which includes bare NULL, won't pass any actual mail
transport.
Yes, but some 10646 encoding will, eg UTF-7.
That was why I created the mnemonic system some 4 years ago,
which has a smooth migration path from 7-bit ASCII thru 8-bit
character sets into 16-bit and even 32 bit encoding.
Mnemonic is a little worse than quoted printables in a sense that
they can't be viewed by those who don't support it.
no. Mnemonic is intended to be used with *no* support.
That is the whole point of it.
Of cause mnemonics can be used with mnemonic support software, too.
Mnemonic is a lot worse than quoted printables in encoding length
and understandability.
The encoding length of QP vs mnemonic is almost identical.
In the typical case, 8-bit charsets, both use 3 bytes per character.
Mnemonic is more understandable than QP; that would be part
of the main purpose of the mnemonic design.
Almost no one knows that "NASCULINE ORDIAL INDICATOR", whose shape is
superscript-o-with-underline, is 0xba of 8bit ISO 8859/1.
true. The mnemonic is "-o" for MASCULINE ORDINAL INDICATOR, so that
I think would be a little easier to remember than the QP hex code.
Also I do not think many users would know the official name of
that character, and some would not know how it is used. The mnemonic
are then designed so that the ASCII combination graphically
resembles the original character, and the use and meaning is not
indicated (in most cases).
keld