[jack_tripper]

>And I'm guessing more people are likely to have heard about NextCloud before while probably not heard about Atos before, unless you're Austrian

You don't need to be Austrian for that. Atos is a pretty infamous IT services provider that operates in all of Europe and has the same issues as all such service providers like Accenture, Cognizant, Capgemini, Wipro, IBM, NTT Data, etc, and so far I haven't see ONE SINGLE CASE where these clowns were involved in a government project and it didn't turn out to be an expensive, over-budget, delayed, shitshow leaving the taxpayers holding the bag.

Like for example Austria has a national highway company(ASFINAG) and national railway company(OEBB) where the government is majority shareholder and they work pretty damn good to serve the taxpayers and the users of those services whether they're Austrians or not.

So then why not have the same for IT infrastructure instead of outsourcing it to all these parasites? It's 2025, when do we start treating IT infrastructure like road, rail, water, energy, healthcare etc already? How many decades more need to pass till the government realizes that the internet and associated services are also worthy of national importance and therefore ownership?

I'm not saying to nationalize the internet, on the contrary privatization and decentralization is better for consumers, but the digital interaction between taxpayer and government is something that should not be outsourced to the private sector, especially not to foreign publicly treaded companies like Atos, who have no skin in the local game and don't give a fuck if they leave an expensive mess behind as long as they can ride the gravy train while it lasts.

So excuse me if I have a high degree of skepticism when I hear about the involvement of the likes of Atos in taxpayer funded projects.

[tsimionescu]

So much this. It's absurd that in a world where every government interaction isoving online, we still ask every individual state institution to contract out the development and maintenance work to outside companies, instead of having a government IT provider.

The savings on bureaucracy and time spent analyzing puvlic offers alone would be immense over a decade.

[atonse]

At least in the US we started solving this by having high salaries for tech workers in government (see 18f and USDS, etc, both shut down by Trump), or UK's GDS which was a pioneer in this space.

If you want to attract good talent, there are successful models out there now, but you have to start by paying them way more than the average government salary. But the contractors throw lobbying money at these things and try to stop them every step of the way.

[tsimionescu]

The problem isn't that governments can't hire programmers. It's that they refuse to hire programmers, and prefer to pay the same consultancies for the same programs over and over again.

[jack_tripper]

I disagree. Unlike the US, tech salaries in Austrian private sector are not terribly amazing to begin with, so the Austrian government would have no issues to find labor within the budget that they gave Atos, not to mention that government workers in Austria have other perks that workers in the private sector don't have, like harder to fire, having their own private kindergartens for the workers' kids, much better pension funds and health insurance funds with more coverage and less waiting times, public housing, etc that to a lot of people will have more value than a higher paycheque in the private sector.

So IT IS technically possible to gather the labor force to build the project in house, it just isn't much political motivation to do so when you have lobbyists swaying leaders in the other direction, and the investigative journalists and voters are too tech illiterate to understand this type of grift because when the government pays a billion Euros for a bridge or a tunnel and after 10 years the bridge or tunnel is not there, everyone notices and someone needs to go to jail or at least loose their job in politics for that obvious theft.

But if you spend a billion to consultancies on a government IT project, and it's an offshored clusterfuck that barely works and could have been done better by a local shop for 1/100 of the cost, then the journalists and taxpayers have no clue they've been robbed blind because nobody understands the nitty gritty and costs of SW development, and unlike bridges and tunnels, the public can't see the source code in the open as they walk to school to see that there's nothing there, which is why government IT projects has now become the best and easiest way to funnel taxpayer money into private pockets.

[Ajedi32]

I think from an efficiency standpoint it makes sense to contract out to bigger players. Economies of scale are huge in software and IT since once it's written copying and running code is basically free.

The problem of course is that using someone else's proprietary, closed-source code makes you beholden to them. That's a problem for consumers but it's an even bigger problem for sovereign nations. Would be a great outcome if greater awareness of this problem lead to more state resources being invested in open source alternatives to proprietary software.

[tsimionescu]

The economies of scale would work exactly the other way than you think. Right now, the same company can sell the same solution for the same money to 20 government agencies, ones that have broadly similar needs, because it costs too much for anyone else to compete with them. The company then extracts massive profit from every subsequent project, with none of the savings going to the government. And even if a new player wins some of the contracts, they have to start from scratch and thus need to charge similar prices.

If there was a government IT office, it could build this in house, and after the initial investment in building the base infra, re-use it almost for free in every government agency in the same country. In the context of the EU, they could even make moves to share this code with other governments, passing on the savings there as well.

[jack_tripper]

>Economies of scale are huge in software and IT since once it's written copying and running code is basically free.

If that were true, then all these government IT projects from these infamous consultancies would all come in-time and under budget, but that's never the case, because every government wants things completely different than the other government, so it's never a just a copy-paste, fire-and-forget type of job.

[1oooqooq]

oh it very much is. they just act and bill like it's not.

corruption requires costs you cannot verify after delivery. for construction it's the exagerated foundation which they only actually deliver what's needed and pocket the difference. for software it is the hundreds of rewrites that may or may not have happened and are now in the past.

[tsimionescu]

> corruption requires costs you cannot verify after delivery.

No, that is plain fraud. Corruption is paying so that no one notices or cares about the the costs that can't be justified after delivery.

[1oooqooq]

i guess your pedantry is right. it would be much more expensive to pay for corruption without the "safety" of some well executed fraud... but now it's open season and nobody even have to care about looking innocent anymore.

[Ajedi32]

On time and under budget is relative to what you set the budget and deadlines to in the first place. If these companies had to rewrite Excel from scratch for every client I guarantee you the budgeted cost would be a lot higher (and they'd probably still go over that figure).

[harvey9]

Nobody suggested rewriting Excel or even customising libre office. These projects are often ERPs which get customised to the client's requirements. Chaos and ballooning costs often follow for all the usual reasons.

[Ajedi32]

My point is "completely different than the other government" is only true to an extent. Even with significant customization, there's still a lot shared which benefits immensely from economies of scale. As you said, nobody's rewriting Excel.

[tacticus]

> I think from an efficiency standpoint it makes sense to contract out to bigger players. Economies of scale are huge in software and IT since once it's written copying and running code is basically free.

I mean sure if it wasn't for the fact that those bigger players are going to be looking at this as a way to print money.

[StopDisinfo910]

You think governments should have a full integration team full time for potential software deployments of bought on the shelf solution?

That doesn’t make much sense. Governments don’t have that much software integration to do. Especially when you consider that these projects generally require specific knowledge of what’s integrated. What will these people do the rest of the time? Are you suggesting hiring specialists of every piece of software the government is likely to use full time?

I mean when faced with something you completely fail to understand there are two solutions: it’s all a scam or you are missing something fundamental. Here I think a lot of people are completely misunderstanding what integration is about.

Consider that these are not IT issues which push all these departments to ask for different feature sets and customisations leading to every integration being different and these are not problems programmers will solve.

Consider also what happens if projects fail. With an integrator, you sue them or breach with penalties and move on to a competitor. No harm, no foul. If it’s internal, you have a full on restructuration on your hand for something that is not even your core responsibility.

Anyway, I would like to see the face of some you when you learn that it’s highly likely that’s the people managing the integrators from the customer side were probably mostly consultants for a big consulting firm because that’s another thing government agencies don’t know how to do.

[lou1306]

> Governments don’t have that much software integration to do

...What? Most European governments rely on herds and herds of pachidermic, segregated software systems and databases. There's surely enough to keep a whole team busy for years, if not decades. And I'd be surprised if the final costs would be higher than hiring consultants again and again.

[StopDisinfo910]

You are mistaking technical issues and organisational issues.

Projects are driven by business requirements and values not a desire to share more or rationalise. Segregation is more often a matter of governance and processes.

That's not things you will solve using a bunch of developers. This discussion makes me realise that most of the people commenting on HN probably work for software companies and have very little experience of how big projects, be them IT or organisational, are conducted in traditional companies and what are the challenges they face.

[lou1306]

Well of course going towards integration vs segregation is an organizational (in this case, political) decision. I was saying that the average EU state machinery has a lot of room for integration, should there be a political intention to go that way.

Case in point: in Italy, different towns used to have different systems for their resident registrations. I doubt there was an extreme need for customization in this context, it was just that bigger/wealthier towns had a chance to digitize earlier and so on, leading to extreme fragmentation. Moving to a nationwide register took literally a dozen years or so, for a single service of a single country.

[woodson]

> So then why not have the same for IT infrastructure instead of outsourcing it to all these parasites?

That exists already (and has for a long while): the Bundesrechenzentrum (BRZ, https://www.brz.gv.at/en/). They do a lot of public facing government websites and portals. If you lived in Austria, there’s a good chance that you’ve used at least one of them. The question is, why haven’t they been tasked with this migration?

[protocolture]

>So then why not have the same for IT infrastructure instead of outsourcing it to all these parasites?

Not a comment on Atos directly, but I am betting that its an opex/capex thing.

Government departments see ongoing maintenance as internal, opex driven activity and right size their infrastructure teams to handle the ongoing workload.

Government departments see a large implementation project as an opportunity to save money through capex. Rather than increasing their resourcing internally to complete the project, they send it to tender. The tender document will require that the vendor hand over to their internal team for ongoing maintenance and support at the completion of the project. No ongoing impact to department opex budgets or headcounts.

Even if it blows out by like 50%, they probably still save money by not managing the labor costs internally. And they can force the vendor to take a hit on some overruns.

[mk89]

I prefer Atos + a solid open source solution to "an own IT department that will ditch the battle tested open source solution because XYZ" and then 6 months later bugs rain from the sky with users' data searchable in google.

Opening up a whole department requires skills. If you don't have such skills, please hire the "parasite". I prefer that. At least they provide a service, overpaid, ok, but they have at least some knowledge in the business.

[jack_tripper]

>overpaid, ok, but they have at least some knowledge in the business.

With all due respect, setting up a Nextcloud instance for a government entity is not really rocket science requiring a 150 IQ, Stanford grad, PhD, galaxy-brain labor force, but it's a skill that's easily abundant in Austrian and can be easily transferred to more of the tech labor pool to achieve the same results of what Atos did.

We're talking about a Nextcloud instance here, not building an entire hyperscaler from scratch, like AWS or Equinix, which is indeed a skill next to inexistent in most of Europe, which does indeed require contracting FAANG corps to build because we lack that capability in Europe.

[mk89]

I am not talking about the Austrian people's skills ;) I bet the employees from Atos were locals, so...

I am talking about politicians that are supposed to create the conditions to set this up in a proper, honest and "good" way. As soon as this becomes a "department", nextcloud is not an option because it lacks xyz.

So let's reimplement it worse. All this to justify the need for having an IT department at all.

This is why sometimes I prefer that they just hire some company and that's it. One and done. (More or less).

Also, on a more disturbing note: how do you reduce the costs, when you have public employees....? You can't fire them, or it's nearly impossible to do so. Atos, on the other hand, you can switch.

[franga2000]

But this is the exact issue you have with IT outsourcing - instead of taking the obvious and sustainable solution, there's a clause somewhere in the 500 page requirements doc that doesn't even make any sense, but means you have to use something nonstandard and even add some of your own hacks on top. Because it's a tender, you can't really change the spec and you don't care to either, because a terrible bodge means they have to go back to you whenever it needs changes/fixes.

An in-house development department on the other hand doesn't have to stick to the strictly disconnected way of tenders and the development team can actually work with the stakeholders to develop and evolve the spec throughout the project. They also don't need to guarantee future business for themselves through vendor lock-in or boost their corporate partners through technology choices.

This is an unprecedented case where a private company decided to go the open source route for a government project, usually it's only the in-house teams that pick open source.

[bluGill]

> Like for example Austria has a national highway company(ASFINAG) and national railway company(OEBB)

Since only the government is doing these, there is no real gain from outsourcing - either way you pay the full costs (it need not be that way, but that is how it typically is). For IT lots of others also need that work and so you can share the overhead costs if you outsource.