Hello mark!

26 May 17 06:26, you wrote to Maurice Kinal:

 ml> see what i mean about no character set control line? without it there's no
 ml> way for any system to know what character set to use and whether or not to
 ml> convert it to another...

 I do not agree. Using different character encodings is not limited to
 fidonet. Over the years some nice tools/libraries have been developed to
 define the possibly used encoding. The only thing is that the use of
 these methods have not been implemented in Fidonet software.

 I tried a simple message reader that uses a tweaked version of one of these
 tools, and I can assure you that the results are far better than honoring
 the characterset controlline. The encodings advertised are not always the
 ones that are really used.

 I presume that Maurice does something similar. Where I do not agree with
 Maurice is that, as he has no need for the characterset controlline, he does
 not add one to his messages. It would be a nice guesture if he did, as a
 service to backward compatibillity.

 Omitting the characterset controlline puts Maurices messages in the same
 class as messages origination from software that predates the standard.

 Most readers then make a fixed assumption on the encoding. Applying that
 assumption on UTF-8 encoded messages is not a good idea.

 It would be nice if coders of actively maintained message editors would
 consider to include the guessing tools in their software and ignore the
 the CRHS: kludge.

 But to be honest, that can only be a pipe dream.

Kees

--- GoldED+/LNX 1.1.5