Subj : testing utf-8 via TB
To   : August Abolins
From : Richard Menedetter
Date : Sun Apr 07 2019 14:41:10

Hi August!

06 Apr 2019 19:22, from August Abolins -> Malte Tychsen:

 >>  -áAA> @CHRS: UTF-8 2
 >>  -áAA> @TZUTC: 0300
 >>  -áAA> Daudz baltu dieni+|Ñu Laimi+|Ña dodi,
 >> Looks good.
 AA> You're kidding, right?  :)

He probably is not.
Do you understand the concept of the charset kludge?
It identifies what charset was used when writing the message.
If you identify it as UTF-8, and the viewer can display that, then it will look 
OK.

The problem arises when you have configured a default charset for outgoing 
messages.
You write an answer that is using UTF-8 characters, but your default charset 
(eg- LATIN-1 or IBMPC) does not contain the characters that were in the 
original message.
So it will look garbled.

 AA> ..'cuz your message arrived
 AA> in "CHRS: LATIN-1 2"  ..and consequently, munged up my original.

Yes ... that happens when you cannot translate characters from the original 
message into the charset you have configured for outgoing messages.

 AA> TB seems to be much more pliable with foreign chars (easier to switch
 AA> and change things). ..or the unicode setting in TB just makes
 AA> everything work.

It is simply using UTF-8 encoding, and marks the content appropriately.
But most (99%) of fido systems will not be able to read the special non ascii 
characters.

CU, Ricsi

.... A pessimist is someone who has had to listen to too many optimists.
--- GoldED+/LNX
 * Origin: It's okay to be ugly... but aren't you overdoing it? (2:310/31)

.