Larry Masinter writes:
And the other encodings haven't transformed character sets but
have just caused some terminology changes in order to make
distinctions that didn't exist before.
How about:
NOTE: The term "character set" as used originally in MIME arose
with the use of US-ASCII and other 7bit and 8bit specifications
which employ a simple mapping from single octets to single
characters. The advent of multi-octet coded character sets and
switching techniques has made the situation more complex. For
example, some communities have adopted the term "character encoding"
for what MIME calls a "character set", while using the phrase "coded
character set" to denote an abstract mapping from integers (not
octets) to characters.
At the time of specifying the first RFCs of MIME, we were fully
aware that taking a simplistic 7bit or 8bit view was not
adequate and that we needed support for multi-octet schemes.
These were under development as RFC1342 etc were published.
Quite true, but completely irrelevant. The point is that the use of the term
"character set" was based on usage found in X3.4 and similar documents, and
that MIME usage did not reflect the terminology from documents describing
multi-octet character encodings. This was an intentional choice we made at the
time. It was far from clear back then what terminology and/or character
encodings would eventually come into widespread use.
All this says nothing at all about what efforts were underway, either then or
now, to define and register multi-octet character encodings for MIME use. We're
all well aware that efforts to label, register, and otherwise deal with
multi-octet character encodings in the Internet context predate MIME and were
the initial rationale for starting work on MIME.
And even though such efforts were underway before MIME, need I remind you that
at the time the ISO 10646 draft was a very different document specifying a very
different character set? Things have changed, and all this note is there to
indicate that MIME terminology is now different from the terminology used in
other specifications.
Ned