Comment by simpaticoder
2 days ago
I've worked with older governmental systems, and chances are they are running a wide variety of systems, some of which, the oldest and most critical, are probably written in COBOL running on IBM mainframe hardware. In those environments, there is no real distinction between "database" and "application". COBOL systems are very file- and batch-oriented, and are "monolithic" in the extremist sense. The technology itself makes it impossible to give read only access to such systems.
> The technology itself makes it impossible to give read only access to such systems.
This isn't true. Mainframe COBOL systems commonly store data in VSAM files, or DB2, or IMS, or sometimes some more obscure non-IBM database (e.g. CA/Broadcom's Datacom/DB or IDMS, or Software AG's ADABAS). But whichever one they use, there are multiple ways of granting read-only access.
For example, if it is VSAM, you can configure RACF (or TopSecret or ACF2) to allow an account read (but not write) permission to those VSAM datasets. Or, you can stick DB2 in front of VSAM (on DB2 for z/OS, CREATE TABLE can refer to a pre-existing VSAM file, and make it look like a database table), and then you can have a readonly account in DB2 to give you access to that database schema. Or, there's a lot of other ways to "skin this cat", depending on exactly how the legacy app is designed, and exactly how it stores data. But, probably this is already implemented – most of these apps have read-only access for export into BI systems or whatever – and if it happens for whatever reason not to be, setting it up should only be a modest amount of work, not some multiyear megaproject.
>Or, there's a lot of other ways to "skin this cat", depending on exactly how the legacy app is designed, and exactly how it stores data. But, probably this is already implemented
Given that neither of us knows the actual systems in question, what is more likely, that it's a well-designed system or one that has organically accreted over time? It seems like you tend to believe the former, and I the latter. I suppose my view is based on the fact that, like in statmech, you enumerate all possible systems that can do a particular job, the vast majority of those solutions will not have any organizing principle and will not be amenable to surgical analysis or change.
I think the difference is that I know that getting data out of mainframe COBOL systems is a long-known and long-solved problem, and I can list lots of different ways to do it (I mentioned a few, there's several more I didn't mention). Without knowing the details of the exact system, I'm not sure which one would be the best one to use, but the odds that you'd have a system for which none of these existing solutions is suitable is rather low – and indeed, likely most of these systems are already using one or another – there are whole teams of sales people who have spent the last 20-30 years convincing government agencies (inter alia) to buy these solutions.
Whereas, you don't seem to know anything about that topic, and are speculating based on parallels with completely different disciplines (such as statistical mechanics).
We both are speculating due to lack of details about the specific systems under discussion, but wouldn't you expect the person whose speculations are based on greater relevant knowledge to be more likely to be correct?
4 replies →
This implies good faith actor, which is not the case.
You can absolutely give read only access in COBOL systems. That's just lazy administration and IT security on a shoestring budget.
These old systems need to be upgraded but Congress never approves the financing or execution of it because they're too divided and won't increase taxes on anyone especially the rich and giant corporations to do it.
[flagged]
You know that annoying thing where someone joins a new team, looks around, declares all their friction points to be easily solvable, dives in & starts making changes, and turns out to make a big giant mess?
And the reason is they don't understand the specific domain & context well enough to know what the actual hard problems are. Instead they're just pattern matching to things they do know and extrapolating. And it usually doesn't go well.
Dealing with a system that's replicating 50 years of regulatory rules is going to be that times infinity.
I don't think that's annoying. If they make a mess, then by the time they're done cleaning it up, they'll be an expert, and you won't even have to train them. That is exactly what you need when the system is broken. The existing people should be encouraging, let them try, and lend their wisdom when they can. Disruption has always helped the tech economy thrive and government should welcome the opportunity to learn this aspect of our culture.
>They don't even know how to build a website that works.
What percentage of people who know how to make a "website" do you think could make an automated tax system?
>the tech industry has been the beating heart of this country
Agriculture? Construction? The heart means something without which you can't function. How did people in the 1950s survive?
The agriculture industry is a skeleton crew for something that's largely been automated by tech: https://justine.lol/tmp/agriculture.jpg There's not much of a construction industry either, since the government doesn't let us build anything except sprawl.
The USG does in fact know how to build a website and it is intellectually lazy (so very lazy) to suggest otherwise. A high profile illustration of this is login.gov, which is SSO used across USG agencies. It's not possible to take a comment like this seriously, at all.
Elon Musk is also not an auditor. DOGE is not an auditing entity. You bring in accountants to audit. These are 20 y/o something programmers. How DOGE has been operating has been completely opaque and this lack of transparency just plays to the point that what someone says their goals are and what their actual goals are are not mutually exclusive, so no, Elon Musk shouldn't be allowed anywhere near these systems.
Are you familiar with healthcare.gov? It was a disaster. So the government let some people from the tech industry come in and help. Techies saved Obamacare and then founded an agency called USDS, who did other sites like login.gov. DOGE is basically doing what USDS pioneered, except now tech people have earned enough trust to fix the government itself, rather than just being the wiz kid who fixes their website.
3 replies →
"fixing the government" in this case seems to mean "destroy the government" for somewhat hidden purposes.
hidden? I think tearing down government is a pretty damned good fix, and so does many others
11 replies →
DOGE literally took over the agency that competently modernized and integrated US gov technology (United States Digital Service), gutted it, and is now using that agency's pretense of needing access to data to now pilfer citizens' private information and grossly violate the constitutional separation of powers.
This is the mechanism by which this administrative coup (declared here in https://www.whitehouse.gov/presidential-actions/2025/02/ensu...) is being enacted. None of this is legal or constitutional in any way.
The rule of law is not a partisan issue nor a matter of "government efficiency". Those who aid this coup should be considered traitors.
If it ain’t broke.. move fast and break things?
All I've seen about this DOGE stuff is negativity based on hypotheticals, this is the first optimistic hypothetical I've seen so far.
It's an interesting point. As a thought exercise, tech is absolutely the core of modern America, #1 export (I assume) and a key market. Private sector influence probably can give huge amounts of low hanging fruit.
I think peoples main concerns stem from not trusting Trump (which seems odd given he's a second term president, he is objectively wanted) and not trusting Musk (which is probably fair, he's publicly and openly an arsehole).
Speed probably concerns people too, however "move fast and break things" is a pretty fundamental American tech mantra, so entirely unsurprising and usually effective.
Trump winning the election wasn't necessarily because he was "objectively wanted". It could be because he was less disliked than Biden at the time. Plus I wouldn't be surprised if a lot of people voted Trump but then his first couple of weeks made them go "hang on a sec...".
3 replies →
> not trusting Trump (which seems odd given he's a second term president
You might recall that at the end of his first term, he tried to overturn the results of the election he lost, calling up the Georgia attorney general to demand the vote total be blatantly altered, and even siccing a mob against the Capitol to physically prevent the certification of the results.
That's why many people don't trust him.
> That's why all this stuff is backed up to an iron mountain.
When one of your threat vectors is a massive ball of nuclear fire right on top of the federal government in DC, your offsite backup policy is going to be absurd overkill by the standards of any other organization on this planet. That doesn't mean it's flawed.
> ...many of the people in charge don't even know how to use a website. Now for the first time, tech industry people have the opportunity to help run these computer systems, and you're afraid they're the ones who'll be incompetent and accidentally break everything?
Are you honestly suggesting that the people who built these systems, maintained them, and updated them to reflect often significant changes in rules and regulations over the course of decades somehow don't know how those systems work? If they were so damned clueless, those COBOL systems would have sputtered out and died decades ago. The fact that they've continued to run for all this time is practically prima facie evidence that the system works just fine by industry standards for that kind of legacy code.
No doubt there's plenty of stuff buried in the codebase that bugs the hell out of the developers working on it, but you get that with any complex legacy code. It's the nature of the beast. Do you think there's nothing in Google's monorepo that some of their engineers don't quite like but doesn't rise to a big enough issue to warrant refactoring right now? Any other FAANG company? Or large tech company in general?
You're writing as though a bunch of junior developers--and that describes pretty much all of the publicly known DOGE employees so far--are wizards who can just waltz right in and magic up a better solution just because they're from the "tech industry."
Setting aside the unlikely chances that those juniors--no matter how skilled or talented--have any experience with COBOL, mainframes, or even just decades-old legacy code, is anyone going to suggest that something like the federal government's payment system isn't defined by an immense amount of complex business logic so as to comply with legislative requirements? It's not something you just start playing around with.
I can't think of any tech company that would take a junior developer, toss them overboard in the middle of the freezing Atlantic, grant them sudo access, and tell them to do whatever the hell they want with critical systems before they drown and--somehow--take the ship with them. Worse yet, those juniors were chosen for ideology fervor and/or purity, so what happens when the normal review processes and experienced senior developers are pushed aside because they're in the way and part of the "deep state conspiracy" that doesn't want them to "[fix] the government" as you put it?
Not only is that a recipe for disaster for the company itself, it's a damned good way to take an otherwise talented junior developer and permanently ruin them. Instead of mentoring them so they can work well as part of a team, you're basically creating a toxic working environment that's going to turn them all feral. By the time they crawl out the other side and the public hears all about what they've been up to, what company is going to be stupid enough to a developer with "DOGE" on their resume? Beyond that, you're conflating a whole bunch of different issues here with federal software contracts and IT, while putting the tech industry on a really peculiar pedestal.
Besides, if the goal is to discover waste/fraud/abuse, the obvious answer is to hire a bunch of forensic accountants and let them dig into everything. Those are the people who actually find that kind of stuff, and they're incredibly skilled at their job. If it's there, given the time, they'll find it. But it's a slow-going process, so we instead see a bunch of engineers focusing on random transactions so they can ask themselves (1) "do I like that one?" and (2) "do I think it's legitimate?" because it's faster.
That's not exactly how you fix anything, least of all a country.
I'm not questioning the reliability of their systems but the content of their databases.
The DOGE workers are already legends in their own lifetime, having saved $55 billion, and they haven't even gotten started. That's like 20% of Google's yearly revenue, all in a few weeks, and without needing to write petabytes of code in a monorepo.
I don't think it's accurate to mentally model these payments as though they were counter intuitive algorithms in a deeply embedded software system. Waste fraud and abuse can be painfully obvious. So it's not the complexity of the problem that has prevented it from being solved. It's the political cost. Senior people have spent a lifetime accruing political capital. They're afraid to lose it. They're only going to spend political capital if they get something in return. They know and have cultivated relationships with the people who will be unhappy if particular instances of waste get solved.
So it makes sense that Elon is unleashing his crackerjack juniors.
They're perfect for the job.
> What's with you people
Right?
> For decades the tech industry has been the beating heart of this country that's kept the American dream alive
By "tech industry" do you mean the consumer surveillance industry? Maybe your vision of the American dream involves inescapable corporate control, but mine certainly doesn't!
I'm talking about the tech industry that invented a self-driving bulletproof truck that looks like a DeLorean and is faster than a Lamborghini which anyone in the middle class can afford. If Elon can make science fiction real for the masses, then he should be able to balance one itsy bitsy tiny little federal budget.
1 reply →