Hacker News Reader

Comment by insane_dreamer | Hacker News Reader

I find it really hard to justify spending $600M on a system like that? Look, if you had 500 skilled FTEs working on the project for 2 years, at $250K per FTE, that would be a total cost of $250M. Say 50 FTEs for ongoing support at $12.5M/year or $125M for 10 years. So $375M.

But the above numbers are hugely generous. This is not building an ERP from scratch. Do you really think it would take 500 people (say 400 engineers and 100 non-engineers), to build and deploy such a system? I would imagine you could get it done (and done right) with half that many, or less.

Anyway, just mind-blowing.

View on HN

IF you go to a body shop to get a dent pulled, you will receive one quote. If you tell them insurance is paying you will receive a second, much higher, quote.

The dealer will say it's because the insurance company is such a hassle to deal with that they have to dedicate employee time to jumping through hoops to get paid. The insurance company will say the quote is higher because the dealer sees the insurance company as a big bag of money and if the insurance company doesn't ride them then prices of insurance will go up.

Dealing with government is this times a thousand.

Your analogy is somewhat backwards. Insurance companies squeeze body shops pretty well. If you go in there and ask for the same work you will pay more.

The real difference is that often times a person will just say make it look okay and I'll be fine. Whereas insurance work is basically making a car good as new because that is legally required. If you try to get the same level of service at retail you will pay more.

Interesting. I've been lucky in that I haven't had non-cosmetic damage repaired, but it always seemed like there were several iterations when insurance was involved.

There are tons of body shops that won't even take insurance work because of the low pay and somewhat low standards of work as well.

With Oracle? 400 Engineers is likely way to high, it's more like 100 Engineers and 900 non-engineers. Oracle likes to throw in a ton of people into contracts with lie that they will provide "white glove" teaching and various other things.

It might be reasonable, if it was quality software. But it is crap.

Governments should unite to create basic open source software and then individual organizations can tailor it to their needs.

The trick is the find some way that a vendor can monetize it. Sometimes you have management consulting groups building crappy unmaintained (except by them, if you extend the contract) OSS.

One thing that seems to work, sometimes, is having cloud vendor support. Could you support maintenance of an OSS platform that AWS (and/or their competitors) operates - that way AWS could "win" the contract, but the OSS system gets funded with a core team and various groups still contribute to make it better

I agree. Any tax payer money used for creating software should be required to be open source. Ideally, governments and similar organizations would utilize open source software, then use contractors to modify, support, and maintain it (again, releasing any changes as open source).

Contractors would then be responsible for providing excellent support, not some huge bloated product.

It is a partisan fight for the government to build and offer tax software. Huge uphill battle to build out infrastructure code projects.

Governments are controlled by rich people's interest. Of course they are lobbied to death to avoid any such kind of solution.

Bribery isn't control. It's just bribery.

Direct or indirect, it doesn't matter. It is a strong form of control.

Pretending it is impossible to get a semi-democratic government to work for the people instead of just very difficult is defeatism. In fact, getting people to believe "It's hopeless" is a big part of the way bad governments maintain control of people, like in Russia or in the USA

I'm not saying it is impossible, I'm only talking about the situation in the US. Currently it is dominated by the billionaire tech and financial oligarchs and we need to do something to change it. It doesn't make any good to sugarcoat this reality.

Governments aren't able to manage their way out of a wet paper bag unless it threatens their existence.

And yet 90% of all startups fail, which could also be seen as a indicating a certain level of incompetence in the private sector.

Consider perhaps that incompetence is distributed across the economy, and government and private industry share in that, with successful and unsuccessful projects as a consequence.

Governments can print themselves a bigger budget. Startups can't.

I would say you probably could get it done with 5 people (1% not 50%)

Maybe not 5, but fewer than 50, yes, even with QA/testing/deployment. So long as they're skilled.

5 is a stretch for sure, but Retool, for example, grew to a 9 figure valuation with 2 engineers, and AI is now a big boost to developer velocity.

Reply on news.ycombinator.com