Subject: Re: java to rexx
From: Martin Trübner (martin@pi-sysprog.de)
Date: Fri Nov 06 2009 - 14:01:38 EST
Tony,
let me answer each piece of your stmt.
> The end would be the ability to compile COBOL into objects that do not
> need LE and it's associated overhead
You are saying that LE is overhead. This is not true. It is a runtime
and it simplyfies a lot of tasks. It may not be optimum in some areas-
but naming them is better than this general stmt.
> or that must be written in assembler.
Which is what? IBM tries to get customers of of HLASM and they try to
offer every interface in such a way that higher languages can use it.
> The thought would be that shops with little or no assembler skill
> could actually use zCobol for things that currently require assembler.
The though of a shop that has little to no assembler-skills writing an
ICHRTX00 or a VENDOR-exit "pre-open" scares me.
> The primary source for each module would continue to be COBOL.
and then they use only the implemented subset- who is going to do the
maintenace and development (for what money?)
> Things like hooks into vendor products were LE can't be used. Maybe
> even operating system hooks?
Shivering and shaking c...akpeg
i0 z 35ki
5
klp o sjkasjk
--
Martin
Pi_cap_CPU - all you ever need around MWLC/SCRT/CMT in z/VSE
more at http://www.picapcpu.de
This archive was generated by hypermail 2b25 : Sun Nov 22 2009 - 02:20:06 EST