On Tue, 11 Nov 2008 16:40:50 -0500, David Andrews wrote: >On Tue, 2008-11-11 at 14:47 -0600, Rich Smrcina wrote: >> Why re-invent the wheel on z/OS? > >Because that's where the data is. (Or "are", for John Gilmore.) > Because that's where the SVCs is. I'd love to be able to ATTACH GIMSMP, GIMZIP, and IEWL from a process space in which I also could employ the GNU utilities. find(1) lacking -print0 and xargs lacking -0 are pathetic.
How wide is the ASCII gulf? What fraction of it is bridged by the C Enhanced ASCII support? What fraction of GNU utilities could simply be compiled and run in Enhanced ASCII mode? I know deficiencies in libraries: Curses and X11 are not supported in ASCII. If that were resolved, I'd mount all my foreign NFS filesystems XLAT(NO) and convert all my z/OS Unix filesystems to ASCII, and look back with a twinge of regret only at Rexx for its SYSCALL support. (And even Rexx might run with data conversion.) I haven't the talent nor resources to build NCurses and XFreeBSD on z/OS (in ASCII mode). In retrospect I rue IBM's decision to implement Unix System Services in EBCDIC rather than ASCII. Diachronicity: the Enhanced ASCII support didn't exist in the needed time frame. I would have preferred a proprietary kernel with shell and utilities all left to open source. ASCII. -- 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
