Hello,
Am 21.04.2025 um 00:06 schrieb Reeve <rfritchman@xxxxxxxxx>:
I think part of the push for browser apps comes from users who grew up using PC's instead of adding machines and 3270/5250's.
At least in part this seems so, yes.
But the 132x27 limitation gets under my skin and having a broad palette of colors and field attributes would be enormously helpful.
A very valid point. Looking beyond one's own nose, having a freely definable terminal size is very convenient for the user, but I'm 100% sure this won't happen for 5250.
3270 was AFAIK less restricted but I think this involves to assemble the screens "by hand" into the direct 3270 data stream. Not sure if CICS BMS macros could handle that?
Have you seem some GUIs of real world applications out there? Sometimes a screen filled with an incomprehensible mess of static text, editable fields, dozens of buttons, etc. There is a saying: Less is more. :-) I believe that restricting
Then there's the other question: is the i the right platform?
Heresy! ;-)
I actually have asked the same question some time ago. As so often: It depends.
For me, the absolutely outstanding feature of IBM i is how easy it is to create text based applications from nearly scratch. Ever tried to achieve something similar on e. g. Linux, with libncurses? It's a mess!
I have not too much first hand experience, but dynamically assembling web pages on IBM i ("modernizing applications") isn't that straightforward, unless one has previously established a large library of reusable code. But the same stands true for e. g. Linux. With a slight difference: Linux is ubiquitous. Everyone being interested in computers sooner or later experiments with it. That happened for decades. Some of those people learned to program and published what they've done. This created the bigger part of the OpenSource ecosystem we now take for granted. Some of that can be integrated into PASE more or less easily.
Is IBM i the better web server? I don't know. But "better" is not always what cuts the tie.
The database and hardware reliability are key factors in maintaining the course. But there are plenty of other cloud- and locally-based platform options, some of which have attractive features. There are no great decisions here, only the least bad.
Some members of the general Midrange list hinted that IBM i is a platform slowly fading into oblivion.
Ever heard of some company moving away from SQL Server, MariaDB, Postgres, Oracle, or whatever to IBM i as a database server platform in the last three decades? I haven't. Note, I'm not talking about mergers or similar forced fusions of companies.
IBM i survives because it has actual users generating enough revenue for IBM to invest developer resources, in turn keeping the admins happy. Data always was on the 400 when it was a hot thing, probably migrated over from S/36, and has been migrated to newer hardware as it became available. Applications were expanded and rewritten as newer features became available over time. Applications which helped to run businesses, and generate income.
Rewriting all of that from scratch for some other platform — maybe the "better" web server" — might be doable. But is it a good idea? Is the potential cost cut of the money IBM asks recurrent worth the effort and cost of migration? Is it worth to deliberately throw a spanner in the works?
IBM i survives to this day because the answer is at least sometimes found to be "probably not". As you say, it's not a great decision, but it's the least bad.
:wq! PoC
midrange.com
