DOGE Plans to Rebuild SSA Codebase in Months, Risking Benefits and System Collapse
-
Do you have reason to think that? Organizations that use mainframes keep them up-to-date in my experience.
-
I've worked on teams converting legacy code for most of my life. The planning for something like this would take longer than six months.
If this proceeds in Trump's corrupt government, Elon will get the contact, will claim it is too broken to salvage, and will privatize it. The only way this goes anywhere is if Trump and musk stand to gain money, and they stand to gain a lot.
-
In theory, it wouldn't be a necessarily bad idea to port the COBOL code to something more modern, but I cannot trust Muskrat and a few vibe coder youngsters with this task.
in theory yeah it's not a bad idea but it's also 60 million lines of a cobol
-
"ROFL"
Signed, everyone who has been involved in migrating a codebase before.
That is the mother load of all code bases. Probably still some COBOL if not mostly cobol.
-
Okay but have you ever tried just throwing genAI at the problem and not caring about the consequences?
With Grok looking more and more like the only one working for Musk with enough (digital) balls to stand up to his boss, that might be better than the alternative of "Big Balls" and the rest of the Digital Oblivous Goons of Elon
-
I wish Java was declared deprecated back in 2017 when Kotlin was adopted for Android and supported by Spring. It was the only sensible way forward for JVM. Sure with containerization there's some debate for the necessity of JVM at all but its GC and runtime optimizations are nice.
Java has still an enormous adoption, though. It'd be very self harming making it deprecated. And I don't think containerization solves majority of cases, perhaps it does it better in backend. So both GC and Java are here with us like forever
️ -
I've worked on these "cost saving" government rewrites before. The problem is getting decades of domain logic and behavior down to where people can be productive. It takes a lot of care and nuance to do this well.
Since these nazi pea brains can't even secure a db properly I have my doubts they'll do this successfully.
Not just domain logic. The implementation logic is often weird too. Cobol systems have crash/restart behaviour and other obscure semantics that often end up being used in anger; it's like using exceptions for control flow, but exceedingly obscure and unfortunately (from what I've seen of production cobol) a "common trick" in lots of real-world deployments.
-
In theory, it wouldn't be a necessarily bad idea to port the COBOL code to something more modern, but I cannot trust Muskrat and a few vibe coder youngsters with this task.
Bro. Check it out bro, we're gonna like make it this dope Electron app, bro. It'll interface with X, bro and everyone will have to login there to get their money, bro. Don't worry tho, you'll get paid in recession-proof Trumpbux crypto currency as long as you claim it in time. But X gets a fee of 60% bro.
Seriously bro we like hired a bunch of grads that took a one week X created code boot camp that like you know revolved around a language big balls created called "cyber coin purse++". On second thought bro we're rewriting it in that. Should be like 2 weeks to rewrite it cuz old people wrote the current code and they're like old or whatever bro. Like I live in an old person's basement and they're just like old, bro.
-
Months? I don't k ow how to code, and even I know that's impossible.
I know how to code, how to manage programs, how to architect huge safety-critical systems, and quite a few other things, and I know that you are right. I'd give it 5 to 7 years if it were adequately resourced, there was political commitment, and the stakeholders could be made to agree a set of requirements, then not change them unless there's a really convincing reason (conflicts with other requirements, impossible to implement, breaks everything, etc).
And the validation and verification of such a system could itself take a year or more, if it's well-planned and correctly executed.
-
If Trump is smart, he'll let Musk do all the unpopular project 2025 stuff, then throw him in prison at the end and escape the blame personally. This way he gets to keep popularity with his base while telling his donors they got everything they asked for. It's what all dictators do, really.
I'd be fine with the Musk in prison part. Launching him into the sun in one of his own rockets would be cooler, though.
-
This post did not contain any content.
They're really playing with fire here.
So many MAGA supporters are seniors who are entirely dependent on OASDI. If Trump's minions break this, we're going to see torches and pitchforks strapped to electric scooters and golf carts coming out of Florida retirement communities in droves.
-
"Whoops, we pushed to prod and have no backups. Sorrryyyy!"
They'll have backups. And those backups and associated restore processes would never have been properly tested.
I've come to places where that happened and there were very few remaining IT ops people after the mass firing that followed.
-
If SS payments stop, there will be hundreds of thousands of people with nothing left to lose.
Suicide bombers on mobility scooters. Be very afraid.
-
My mom is over 80. Little old suburban white lady. She already volunteered - "Just get me close." She'd be super-thrilled to have her shot with a suicide vest.
I also know a 93-year-old lady who has loudly talked about taking a few of the bastards with her. The sentiment may be more widespread than we think.
-
Nah I think it will just be
const benefits = false;In Javascript, no less
-
This is how you know Musk is a fraud. This far into his career and he’s leading teams into rookie mistakes.
Or, he knows this will break it and that’s the goal. I’m just not sure how he avoids the blame.
When he suggested PayPal convert its data centers from Linux to Microsoft Server, I knew he was a wrong'un. That was one of the reasons he got booted out of PayPal, by the way. Even the other bullshitter tech bros knew how stupid that idea was.
-
This post did not contain any content.
This has the stench of junior engineer all over it. This rewrite will go way over budget and come limping across the finish line late, with more bugs and less features than the system it replaces. I guarantee it.
-
They're really playing with fire here.
So many MAGA supporters are seniors who are entirely dependent on OASDI. If Trump's minions break this, we're going to see torches and pitchforks strapped to electric scooters and golf carts coming out of Florida retirement communities in droves.
-
I don't think Rust is a bad language for doing same things people do with C++, but with a smaller standard and less legacy.
But yep, that's the kind of people.
About dinosaur things - I've started learning Tcl/Tk and it's just wonderful.
Oh Rust is great, and it's on my learning to do list...but its evangelists are annoying as shit.
-
It's not a case of "seeing the code isn't perfect" but rather, not understanding the myriad problems the code is solving or mitigating.
I'm reminded of this shitshow:
https://en.m.wikipedia.org/wiki/2010_Queensland_Health_payroll_system_implementation
Queensland is a state of about 3m people in Australia. Their health service employs about 100k people. They ended up spending about 900m USD to develop their payroll software and fix the fuck ups it caused.
I'm an accountant by trade, there's a classic "techbro does accounting" style of development we see a lot. Like if you hadn't spent a career learning how complex accounting can be, it would be easy to look at a payroll system and conclude "it's just a database with some rules".
Yea, that's a mich better way of putting it.