Kenneth Whistler wrote on 1998-12-01 03:24 UTC:
> This means that the previous version of 8859-7 and
> the about-to-be-ballotted 8859-7:1998 will have at least four new
> differences in characters:
>
> U+20AC 0xA4 EURO SIGN
> U+XXXX 0xA5 DRACHMA SIGN
> U+037A 0xAA GREEK YPOGEGRAMMENI
> U+037E 0xAE GREEK QUESTION MARK
>
> And this will imply that there will need to be two MIME *charsets* -- one
> for the old version and one for the new version.
This is a point that I never understood. Why is it practical or even
necessary to have different MIME identifiers for upwards compatible
extensions of character sets? If you have the new character set, then
you can use it for displaying mail in both the old and the new subset.
Fine. If you have only the old 8859-7, then you will see some
replacement character being used for the above four characters, which is
still much better behaviour then just getting a warning message that
charset "ISO-8859-7:1999" is unknown and getting displayed everything in
CP1252 as a fallback, which is what most mail systems do these days.
Some people believe that there are applications out there who negotiate
about some mutually agreed common character set and then transform their
data to match the lowest common denominator set. This is a committee
dream from the days of OSI protocols. It has never worked in real life
and just creates new opportunities for protocol M3 (misunderstandings,
malfunctions, mischief).
Markus
-- Markus G. Kuhn, Computer Laboratory, University of Cambridge, UK Email: mkuhn at acm.org, WWW: <http://www.cl.cam.ac.uk/~mgk25/>
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:43 EDT