On Sun, 3 Jan 2010 15:25:02 -0800, Charles Mills wrote: > >I recall that DOS/360 had a series of file-to-file utilities (no device >independence so you needed a disk to tape utility, a card to tape utility, a >tape to disk utility, ...) > So that's where CMS got that idea. But z/OS device independence is eroding. Why are there TPUT/TGET/PUTLINE/GETLINE (whatever) rather than just doing QSAM I/O to SYSTSPRT and SYSTSIN? And I'm dismayed at the number of z/OS utilities that balk at "DD PATH=...", not because their QSAM calls couldn't perfectly well handle it, but merely because they're afraid to try.
>And anyone else ever write IV-Phase assembler? 5-character symbols, of which >only three (24 bit word machine) were significant, so NOTAG was the same > Someone should have introduced them to RADIX50. But I always disapprove of quiet truncation. >symbol as NOTUP. And it did not flag duplicates, it just silently redefined >the symbol. Nor did it flag undefined symbols -- just passed them to the >linker as externs. > They could have got that idea from C (or perhaps vice-versa). But I once wrote a very simple minded assembler that flagged duplicate definitions only if the values differed. And JCL and Rexx both leave an undefined symbol as it appeared in the source, with no warning. And POSIX shell expands the containing string as if the symbol weren't there at all. All irresponsible, IMO. -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
