On 01.07.2024 06:54, Timothy Sipples wrote:
To address a few questions scattered in this thread...
According to the current TIOBE Index (June, 2024), an attempt to rank the
popularity of programming languages, Python is #1. But wherever you rank it,
Python is at least reasonably popular. Inclusion of popular programming
languages in the base z/OS operating system, with IBM support, is surely a
welcome development. Whether you personally choose to use Python or not is up
to you. And nothing is taken away. If you prefer another programming language,
no problem.
Adding support for Python is not bad, but it is good. Badmouthing other programming languages is
bad, not good.
The popularity communicated by TIOBE is measured by search engines being used to get information
about programming languages (and most likely problems with that language) [1]. The assumption is
that the more queries there are, the more popular they are. However, this assumption is not tested.
For, the more difficult a language, the more queries on solving problems become necessary... ;)
With no representative measurements and no definitions the world agrees upon about measuring the
importance of a language, indexes like TIOBE are taken as a surrogate. One should never forget that.
In the mainframe segment, REXX is much more important than Python. Yet, TIOBE would not be able to
reflect that fact as the mainframe population measured by the worldwide audience of
programming-related queries on some search engines is practically nil (there are millions more
programmers outside the mainframe world).
From a strategic point of view, it is good to add Python to the table, but stupid (and probably
quite damaging) to try to kill REXX in favor of Python before Python has replaced REXX, which is not
that easy and not to be expected soon.
Bringing mainframe REXX up to level 2 of the Rexx language specifications (after more than 30
years!) has unexplainably not happened. Ignoring much more modern versions of Rexx, like ooRexx, is
- to be polite - interesting. Anyone who has seriously looked into it would have realized how much
productivity gains it would immediately bring to the table.
Those shops that are on Linux on Z can check and assess the combination of ooRexx and its Java
bindings, making it possible to jump-start creating Z-ooRexx scripts on Linux on Z that interact
with the Z-related Java classes or interacting with DB2, or... It is quite easy! It is a
productivity explosion that becomes available at one's fingertips on Linux on Z. However, one would
have to look into it. Otherwise, this opportunity will not become visible and a reality.
ooRexx, being a modern version of REXX, would allow controlling the operation of not only mainframe
applications but also Python applications and any other imaginable application.
Again, supporting Python is important in today's world. Adding support for modern versions of REXX
has been long overdue. Given the importance of REXX on mainframes, this is something IBM should not
consider but actively do ASAP if they want to serve their customers as well as possible. It is not
"an either-or" decision, but an "as well as" decision.
---rony
[1] TIOBE definitions:
<https://www.tiobe.com/tiobe-index/programminglanguages_definition/>
--
__________________________________________________________________________________
Prof. Dr. Rony G. Flatscher
Department Wirtschaftsinformatik und Operations Management
Institut für Wirtschaftsinformatik und Gesellschaft
D2c 2.086
WU Wien
Welthandelsplatz 1
A-1020 Wien/Vienna, Austria/Europe
http://www.wu.ac.at
__________________________________________________________________________________
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
