This in not related to the topic. But it did cause me to notice something listed in my copy of System/370 Extended Architecture Reference Summary, GX20-0157-2, dated February 1989.
"Move-Inverse Facility The Move Inverse instruction is provided on the 4381 and ES/4381 Processors; it is not provided on the 3081, 3083, 3084, and 3090 Processor Complexes" "Confidentially doc, I am the wabbit." Bugs Bunny Sent with Proton Mail secure email. On Monday, January 5th, 2026 at 5:27 PM, Paul Gilmartin <[email protected]> wrote: > On Mon, 5 Jan 2026 19:49:26 +0000, Denis wrote: > > > How about that: > > https://www.ibm.com/docs/en/cobol-zos/6.5.0?topic=literals-utf-8 > > Its available in earlier releases. > > Denis. > > Somewhere in there I see: > convenient to represent general Unicode code points > in the literal using only EBCDIC characters > > "convenient" I suppose, compared to binary. > > I'm imagining composing a program on a reasonable > terminal, not encumbered by a 256-character stet, > followed by a preprocessor which would convert > everything outside strings to EBCDIC and treat > UTF-8 strings properly. > > Better, Extended ASCII support for COBOL as z/OS > provides for XL C/C++. > > Is there anything comparable to FLOWASM for COBOL? > > > On Monday, January 5, 2026 at 08:10:15 PM GMT+1, Phil Smith III wrote: > > > > Right, that's why I was joking. Maybe: > > PIC UTF8(6) VALUE U+0445U+043EU+0440U+043EU+0448U+043E. > > > > Still fugly but unambiguous and it'd work. > > > -- > gil > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
