Hello August,

On Thursday July 23 2020 06:58, you wrote to me:

 MvdV>> So it looks OK to you when assuming CP437 is the default. What
 MvdV>> I see are three characters consisting of three stacked dashes.
 MvdV>> Looks OK to me too. But /is/ it OK? I don't know, I do not know
 MvdV>> what a tribble is or what it is supposed to look like, let
 MvdV>> alone a dissected trible.

 AA> Not familiar with the original StarTrek TV series at all?

Sure I know about captain Kirk, Spock, Doc, Scotty, Uhuru, Sulu and Tjechov.

But that was almost half a century ago in what now looks like a totally
different life. I didn't see all of it and I certainly do not remember all of
it. "Tribble" didn't ring a bell, so I probably missed it or just forgot.

 MvdV>> However... the author of the original message says it is supposed
 Mvdv>> to be a single dash and that is how it looks like when assuming
 MvdV>> CP850.

 AA> I thought the single dash comment was unusual too.  Stacked-dashes
 AA> just  made more sense based on the tribble joke.

Wikipedida is your friend, so now I know someting about tribbles. But I stil
can't tell if a single or triple dash would is what makes more sense in this
case. I just went with what the author of the message says.

 AA> And OXP seemed to present alt- 240 just fine.

Is it clearvoyant?

 MvdV>> Your assumption that CP437 is the default turns out to be wrong
 MvdV>> in this case.

 AA> In actuality, I can't be sure *what* default codepage OXP is using
 AA> when  *reading* the original incoming message that fails to provide
 AA> the CHARS  kludge.

In Golded it is configurable. Per area. But that is beside the point. The point
is that without a CHRS kludge, the software has no way of knowing what encodig
scheme was used when writing the message. Determining the encoding from the
message itself requires far more inteligence than can be expected from Fidonet
software.

The course of action is simple. Use sofwtare that can properly deal with the
CHRS kludge, /or/ write using ASCII only.

 AA> I wonder how many of the 3-digit and 4-digit alt-nnnn chars below
 AA> (from a  collection of math symbols which the 3-stacked-dashes is
 AA> found) will  translate properly going out:

The math characters in DOS CP 437 are not al that usefull for a mathematician
as the set it far to limited.

Other than that: you mention 3 and 4 digit alt-N charcters, but the 3 and 4
sequences can not be used in one and the same message. The 3 digit sequences
are DOS characters. Which ones depends on the active code page. The 4 digit
sequences are Windows characters. Which ones is also dependant on the OS
setting. You can have both in the windows command line because Windows (from
Win2000 onwards IIRC) uses Unicode internally. Typing Alt 0128 on my command
line gets me the Euro sign.

You can even have alt [hex nuumber] sequences to get all unicode characters on
the screen but that is not enabled by default. I wrote a Fidonews article about
that. Feel free to look it up.

 AA> Alt 0190 _ Three quarters     <==[does not render in OXP]

Of course not. When OXP is configured to use CP437 it can not display the 3/4
because there is no glyph for 3/4 in CP437. There is a glyph for 3/4 in the
Windows char sets (in my case AKA CP1252) but to disply that you may have to
configure OXP for CP1252. That may or may not be possible.

Cheers, Michiel

--- GoldED+/W32-MSVC 1.1.5-b20170303