Hello Nicholas,

On Thursday July 16 2020 18:55, you wrote to me:

 NB>>> xlatimport cp437

 NB>>> Should fix this for you.

 MV>> Only for this message and others messages that use CP437 as the
 MV>> encoding.

And even more important, it is the same as correcting it in my own copy when my
name is misspelled in the telephone book. It only works for me, not for all the
other readers of the book.

 NB> Which, per our discussion about BBS softwares that don't use the CHRS
 NB> kludge, should fix most things for you.

Not just for me, for the enitre pool of readers.

 MV>>  The world of Fidonet is bigger than CP437. Much bigger.

 NB> The fact that the world of Fidonet is MUCH bigger (I laughed when I
 NB> read that) than anything is in your own delusions. ;)

Laugh if you want, but it is a fact that only a minority of messages written in
Fidonet is encoded in CP437. You may not be aware of it but the vast majority
of messages written in Fidonet is encoded in CP866. Then come Latin-1, CP850,
ASCII, UTF-8 and CP437.

 MV>> What you achieved is converting the Capital O with circumflex to
 MV>> UFT-8. I do not know if that was the plan...

 NB> No. Whatever came in here as cp437 (the umlaut o as was read properly
 NB> in CP437) got converted to that capital O with circumflex.

What came in here was character nr 147 decimal or 94 hex. Encoding unknown.

 NB> If that was wrong, then Golded's conversion tables are trash (oh
 NB> that's right, even Golded the end all be all can not convert 8bit to
 NB> UTF-8 properly,

There is nothing wrong with Golded's translation tables and it CAN correctly
convert 8bit to UTF-8 properly. But Golded is not clearvoyant. It has to be
correctly told what the source encoding is. Otherwise it is GIGO. And that is
what happened.

Thank you for your comments.

I say: When posting messages without a CHRS kludge, one can not expect
non-ASCII characters to be correctly displayed at the readers' end.


Cheers, Michiel

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