ASCII-EBCDIC is on my Time Machine list (you know, "Kill Hitler; avoid the ASCII-EBCDIC dichotomy; no null-terminated strings in C." - you'll have your own items).
Note that UTF-8 is not a code page; it's an encoding for all of Unicode. To convert z/OS to use Unicode at this point would require an unimaginably huge investment, and that wouldn't solve the problem of all the programs that are EBCDIC now. As Ze'ev points out, never gonna happen. Just be glad that the basic character set is fixed in EBCDIC code pages. Imagine if A-Z moved around!
