Why do companies still use this old shit?

Why do companies still use this old shit?

Attached: image006.gif (728x535, 18.12K)

Because it works.

Attached: krosh_happy.jpg (1392x783, 110.17K)

Yeah but it's glitchy and sucks

Mainframe doesn't break as your script does

just werks also who wants to spend 6 mil to change all pcs around

You suck

Attached: yoj_sad.jpg (1200x1027, 98.13K)

but it's archaic and ugly

I don't know what this is but it's based as hell.
You want a javascript app that breaks when some dumbass removes his left-pad script from npm instead?

I'm l1 help desk so I don't know nothing about that. So sure as long as it looks more like a product designed in the last 10 years.

It's IBM CCS and it's actually pretty shitty for something that's been in development since the 80s.

...

This, my workplace uses Attachamate Extra 8, can't develope banking and creddit apps so reliable and so good performing than that.
But we're trying to find a torrent/crack for MicroFocus Extra 9.5 to no avail.

Decades of technical debt make migrating almost impossible, and very very expensive.

we have clients still using SCO OpenServer 5

Getting off the mainframe is incredibly time consuming, expensive, and failure prone.
Especially when you have decades of COBOL/mainframe ASM in production but everyone who understood it has retired.

>Getting off the mainframe is incredibly time consuming, expensive, and failure prone.
Probably, but then again, it's going to get much worse if they wait another 10 20 years
Sooner or later they will have to because no one know this shit
Imagine the migration from the python-JS zoomer generation
Might as well rip the bandage now, this is the lesser suffering.

Imagine the migration *done from the python-JS zoomer generation

You're not paid to ask questions, back to work wagie

Attached: img-2022-04-02-23-08-41.png (1280x720, 1.04M)

Yeah, they're fucked and they know they're fucked. To migrate they need developers that are extremely competent in not just archaic languages like COBOL but also extremely competent in modern languages. You're not just talking about architecture changes but shit is weird from the ground up, those mainframes often don't use ASCII and are instead encoded in EBCDIC. It won't matter to management because they will retire or move on to another job, they just need to kick the can down the road for long enough.

Yeah that's fucked up, There are many others "Not on my watch, but it will happen eventually and fuck the next generation" things like this in other areas. Like in finance with national debt, or ecological catastrophe, etc..
Those guys are irresponsible. Makes me think of my eu country's head of state. fucking cunt.
Migrating from mainframes is sure as hell a big task but it can be done. I'm the kind of guy that could want to do something like this but with a massive salary. There is not really any problem. It's just the entire infrastructure must be replicated from scratch. EBCDIC is probably the most minor problem along the road it seems.
I've been thinking about related things, and I wonder if the best way to approach this would be to make a transpiler, for the language part. Transpiled code, is ugly, but it's better than doing everything by hand. And I'm sure we can add some complications on top of it to tranforms idiomatic patterns into an other, and to also conserve some of the identifiers of functions and variables.

Of course it's possible. The problem is that they have one system they know works and has been working forever, and they need to migrate it to a new system in a way that is massively expensive and could have bugs. They won't take that path until the cost of maintaining them plus the risks associated with switching are significantly more than sticking with them. Until that time there will just be thousands of jobs on job sites for mainframe developers that don't exist.

Only correct answer.

So is Linux.