[comp.protocols.misc] International Character Sets in FTAM, X.400, etc

fdc@WATSUN.CC.COLUMBIA.EDU (Frank da Cruz) (08/10/89)

We're trying to find out how ISO applications, such as FTAM (File Transfer
And Management) or X.400 (Message Handling System), cope with international
character sets.  We have a copy of ISO/DIS 8571/1-4 1986 (FTAM), and find no
specific references to character sets in it.  However 8571/4, which
describes FTAM's transfer syntax, says that ASN.1 (ISO 8824) is used to
specify data types, and it refers to the GraphicString Simple Type, using:

  [UNIVERSAL 22] -> ISO646String

as an example.  ISO 646 is, of course, roughly equivalent to 7-bit ASCII,
and therefore usually incapable of representing words in more than one
language.

For X.400, we have the 1984 CCITT Recommendations.  In particular, Section 2
of X.408 lists nine possible types of coded information: TLX (CCITT F.1),
IA5 Text (CCITT T.50), TTX (CCITT T.61), G3 Fax (T.4, T.30), TIF0 (T.73),
Videotex (T.100, T.101), Voice (further study), SFD (X.420, T.61), TIF1
(T.73).  IA5 is International Alphabet 5, equivalent to ISO 646, or 7-bit
ASCII.  T.61 is the character set used in international (i.e. European)
telegraphy, in which Roman letters may be combined with diacritics which are
transmitted separately.  There's no mention of any of the ISO-8859 Latin
alphabets, let alone Greek, Hebrew, Arabic, or Cyrillic, and certainly
nothing to indicate a concern for Japanese, Chinese, Korean, etc.

We also have a copy of ISO/DIS 9040, Virtual Terminal Service, 1986.
Unlike FTAM and X.400, this application defines its repertoire to include
a set of up to four coded character sets as defined in ISO 2022,
designatable by the escape sequences of ISO 2022 and ISO 2375 and the
International Register of Coded Character Sets.

Our questions are: Have X.400 and FTAM been updated to incorporate the ISO
8859 and other registered 8-bit single-byte or multibyte character sets?  Are
there any other applications besides Virtual Terminal Service which claim to
do so?  Are there any examples of these applications which actually work in
the real world?  Finally, does ASN.1 include mechanisms for designating
character sets?

These questions are in connection with our efforts to extend the Kermit file
transfer protocol to work with text files written in languages other than
English.

Thank you!  - Christine Gianone and Frank da Cruz, Columbia University