|From: Martin Duerst | | |At 11:52 04/07/28 +0200, Biot Olivier wrote: | |>The SMS service allows transport of text messages; however |the character set |>used in such an SMS message (including the message text |itself) is part of |>the binary data and must not be specified with the charset parameter. | |Thanks for the clarification. Apart from the change suggested by Ned, |I would change "character set" to "character encoding". I agree. |I would also be interested to know (doesn't necessarily need to show |up in the registration) what kind of encodings are supported. For |example, does SMS allow message text to be encoded in an Unicode- |based encoding? Are all the encodings supported registered as |IANA 'charset's, or are there others? According to 3GPP TS 23.038 (alphabets and language-specific information), the GSM SMS data coding allows GSM 7-bit alphabet (which allows to stuff up to 160 characters in 140 bytes), 8-bit data and 16-bit Unicode UCS-2 character encoding. The used data coding is encoded in binary fashion in the GSM SMS "data coding scheme", and has no relationship with (read: does not use) the "charset" tokens registered at IANA (although the used character encodings are identical). Best regards, Olivier