[Date Prev][Date Next][Subject Prev][Subject Next][ Date Index][ Subject Index]

Re: Replacing foreign characters



Reply to note from Richard Giering 
Thu, 20 Sep 2001 10:52:08 -0500

> A significant number of XYers also use NB. It appears to use
> still a different set in that if a message is composed in NB
> and then copied (with the windows clipboard) into a mail
> composition window, the sent message will be received with
> other than the characters desired. I expirmented by sending an
> E-Mail to a friend in Germany using the XY and the NB
> characters for Umlaut letters. The XY set arrived readable
> while the NB set did not.

There are two distinct issues here. One is the Windows character
set vs. the ASCII charset. The other is corruption of e-mail during
transport over the Internet due to the fact that messages travel
across different networking technologies. Many mail domains use
variations on the ASCII character set, or use charsets such as
EBCDIC which contain most but not all of the US-ASCII characters.
As a result, messages that contain characters outside the range
ASCII 32-126 may not be delivered accurately. (Even within that
range, "white space" characters, i.e., tabs and Ascii-32 spaces,
particularly at the beginning and end of lines, are not handled
uniformly by all mail systems.) That's why messages containing
accented letters and other special characters need to be encoded
(usually with the quoted-printable scheme, e.g., =AE and =AF for the
guillemets that enclose XyWrite embedded commands), and why binary
file attachments also have to be encoded (usually in base64).

NB for Windows uses the Windows charset (a/k/a ANSI or Latin-1 or
ISO 8859-1) -- no different from any other Windows app (well, the
command line uses ASCII, but that's a whole 'nother kettle of fish).
That an e-mail message "composed in NB" may not be delivered
accurately to someone in Germany (or anywhere else) has nothing to
do with NB or the Windows charset. It's to do with the
unreliability of sending raw (unencoded) characters that are not
pure ASCII 32-126.

The solution is simple: Whenever your e-mail message contains
accented characters (or anything other than ASCII 32-126), either
instruct your mail program to encode the message body (i.e., check
the quoted-printable box) or, if you must, send the message as an
attachment.

(A corollary, by the way, is that mail messages, especially those
sent to a mailing list such as this one, should favor plain text
over typographical fillips that don't travel well across systems.
Use hyphens, not em dashes; vertical quotes, not symmetrical open
and close quotes; (C) instead of the copyright symbol; and so
forth.)

--
Carl Distefano
cld@xxxxxxxx
http://users.datarealm.com/xywwweb/