Am 2000-10-12 um 3:49 h UCT hat John Hudson geschrieben:
> I [...] have encountered an odd symbol [..]
> 	http://www.tiro.com/transfer/thing.gif

Am 2000-10-12 um 10:04 h UCT schrieb ich:
> 2311 Square Lozenge

Am 2000-10-12 um 11:30 h UCT hat Michael Everson geschrieben:
> John's sample looks lots bigger than the itty bitty SQUARE LOZENGE
> in the standard.

Thus sayeth the Scripture:
  Character images shown in the code charts are not prescriptive.
  In actual fonts, considerable variations are to be expected.
(chapter 14, verse 0)

Am 2000-10-12 um 3:49 h UCT hat John Hudson geschrieben:
> I'm trying to rationalise a strange set of legacy fonts

The character presented by John (glyph about as wide as an "n",
and not much higher, preferably square), was heavily used with
the TR440 computer (a German brand), about 20 years ago. It was
dubbed "Kissen" (=pillow), because of its glyph. It was used as
a command delimiter, and in this rôle, it was called "Fluchtsymbol"
(=escape symbol). In session transcripts, the Fluchtsymbol was
printed either as a square lozenge, or a lozenge (cf. U25CA),
whatever the printer would be able to produce.

So, if one of John's legacy character sets happens to be from the
TR440, a lozenge would do as well.

Depending on the character set of the input device, other characters
("$" or "#") would be used to represent the Fluchtsymbol; you
could ask the TR440's input spooler to treat the respective character
specially (i., e., encoding it as hex 35, internally) or to leave it
alone (i. e., encoding it as pillow (7C), hash (72), or dollar (73),
respectively). Because of this correspondence, users tended to equate
the three characters (pillow, hash, dollar). Because of ISO 646,
also "¤" ("Ricardi-Sonne" (Ricardi's Sun), aka "Filzlaus" (sort of
louse)) was equated to "$", and hence to the above.

Best wishes,
   Otto Stolz