darn, that’s what happens when you get distracted before hitting “Submit reply”. sorry if that seemed like just a rehash.
EBCDIC seems to be mostly IBM-propriety stuff. everyone else moved on to ASCII as a more universally-accepted standard, but IBM hardware stuck to its roots (at least, when last I heard). if Big Blue has moved onto another coding basis since then, I haven’t heard about it.
this does cause some interesting collisions when you try and swap data from one system to another. EBCDIC tends to “stack” numerics, cramming two digits into the same number of bytes normally used by one character (i.e., a 15-character number would only occupy 8 bytes, with the leftover half-byte used for the sign ["+/-"] that notes whether figure is positive or negative). this makes for REALLY interesting translations if you don’t have a routine worked in that unpacks the numerics and turns them into “normal” numeric equivalents. otherwise you get all kinds of gibberish. trust me on this one. i’ve torn my hair numerous times when i’ve tried loading data that no one informed me was EBCDIC-based.
lachesis