Hacker Newsnew | past | comments | ask | show | jobs | submit | franch's commentslogin

Italy's unconsolidated budget for 2025 is projected to be around 700 billion euros in revenue and 900 billion in expenditures:

https://www.rgs.mef.gov.it/VERSIONE-I/attivita_istituzionali...

So yeah, whoever talks about these fines as a strategy for fixing the budget knows nothing about the actual budget of a G7 state, these fines are completely immaterial to Italian fiscal policy.

For perspective, that's roughly equivalent to someone with a €50,000 annual income finding €7 on the street and someone claiming they're doing it "to survive."


From a post I saw on reddit:

> In 2024 EU fined US tech companies €3.8B meanwhile public internet tech companies paid only €3.2B in income tax

How is it not a major budget contribution to have fines on American companies bigger than revenue from your entire tech industry?

That is a de facto tax, particularly when they announce these new fines monthly like clockwork.


Income tax paid by public internet tech companies is not the same as “revenues from the entire tech industry”

This report is indicating around 800B in value for the sector (https://ec.europa.eu/eurostat/statistics-explained/index.php...)

While other reports have significant higher numbers https://en.ilsole24ore.com/art/tech-europe-is-worth-4000-bil...


> fines on American companies bigger than revenue from your entire tech industry?

1. As someone already mentioned, taxes != revenue

2. On top of that, "public internet companies" != "entire tech industry"

3. On top of that, tax evasion and creative accounting by "public internet companies" companies is well known, documented, and is subject to additional fines (not as often or as much as they deserve)

4. On top of that "announce these new fines monthly like clockwork" speaks volumes about the state of the "public internet companies" and there continuous disregard for the law.


The relevant comparison is fines vs. actual budget, not fines vs. some cherry-picked industry segment. EU general government spending (across 27 nations) in 2023 was around €8.4 trillion. €3.8B in fines is 0.045% of that, again, completely immaterial.

https://www.statista.com/statistics/1379290/government-expen...


By that standard any actual tax or other income source that’s not at least 1-2% is immaterial and should be removed?


It has been done. See how the leonardo supercomputer is cooled. Closed circuit water that is passively cooled completely outside the building: https://www.youtube.com/watch?v=mckX_R0GeRY


In our Alpine region there is a long tradition of male choruses singing folk songs about mountain life and tales through rich harmonizations of pieces. An example from coro SAT: https://www.youtube.com/watch?v=xQZyZggh3SQ


They keep all the marquees because they sell better in the various European national markets, but many models are now the same platform with some small aesthetic changes. For example the new electric citroen berlingo / fiat doblo / opel combo are the same car:

https://www.citroen.it/modelli/berlingo/elettrico.html

https://www.fiat.it/e-doblo/e-doblo

https://www.opel.it/veicoli/gamma-combo/combo-e-life/panoram...


Heh, I debated putting a second paragraph about this. And yes, while it's not like they're paying the entire cost of developing a car from scratch each time, there's still a lot of duplication of work.

They still have a fair number of unique parts, they still have a fair amount of individual development and certification, they still have their own parts supply chains… these all add costs compared with literally having the same vehicle.

It's not that I see no reason to have multiple marques—after all, there is meaningful differentiation between the different VW ones, for example. But at some point you've got to stop and ask when you have too many.


Yes and unfortunately that's what Chrysler are well known for. Hopefully their culture of churning rebadged (I am a bit harsh here) car has died with the merger of all of those corporation.

Chrysler has been taken over by different entities throughout the last 40 years but somehow managed to drag the new owner down every time. A bit like Rover before the Chinese takeover.


The reason for being a Dutch company is purely for a taxation advantage, and it goes back to the fact that FIAT/FCA moved its headquarters to the Netherlands under Marchionne's leadership. Other italian companies (Mediaset owned by Berlusconi Family) did this too recently: ( https://www.ilfattoquotidiano.it/2021/09/18/mediaset-la-sede... )


no it was not taxation but legal reasons. as it takes years in italy to get a decision from judges.


The original mother company of Fiat/FCA (Exor, controlled by the Agnelli family) now merged in Stellantis, also own stakes in Ferrari, CNH Industrial (New Holland Tractors, Case and other industrial and farming equipements), Iveco and a bunch of other brands. Quite a concentration of companies.


I'm using it for a small project and it works fine. Self hosting is documented here: https://develop.sentry.dev/self-hosted/

Upgrading is a little bit tricky: in my (limited) experience the upgrading procedure for the self-hosted version sometimes fails and requires to take the application offline to perform the upgrade (but it is very possible that I may be doing something wrong).

It is really an invaluable tool.


Not all stone houses are castles or luxury residences. My family's house (northern Italy in the Alps) is at least 600 years old (there is a painting on the outside that has been dated to around 1420) but it is a working-class home. The interior has been repurposed many times over the centuries, but it is still there and inhabited.

edit: spelling


Entire cities in the UK are famous for the kind of stone they are largely built from e.g. Aberdeen and its granite, Edinburgh and sandstone:

http://earthwise.bgs.ac.uk/index.php/Building_stones_of_Edin...


Also, physics-based models (Numerical weather predictions NWP) assimilate much more data than just the last few radars scan. The assimilation phase takes a couple of hours, but when the physics kicks in, the NWP models win hands down.


I've been out of the field for about a decade now. So how often are they running data assimilation into models now? It was basically just 0Z and 12Z back then, because it's computationally intensive. (And of course, you only have radiosondes every 12 hours, and satellites only get you so far.) As are model runs.

Like, of course NWP models win. It's not just the physics, it's all of the other assimilated obs that are advecting over your area of interest.

But it's the problem is always always computation time, to an extent that most people on HN won't get. Maybe the finance guys. But you have to process a mountain of new data, then run the model very quickly for it to be any use at all to the public.

We should use reanalysis for nowcasting, that would be super accurate. /s


I chime in on this because I'm working on deep learning for precipitation nowcasting using radar for my Ph.D. I was very excited when google released the press statement at NeurIPS about their work in this area. Unfortunately, after reading the paper, I have to say that their approach is fairly basic. Basically they threshold the precipitation in 4 thresholds (no rain, light, medium, heavy) and then use a U-Net like architecture, treating it as a classification problem. I think that the works of Shi et al are much more interesting in this regard:

https://papers.nips.cc/paper/5955-convolutional-lstm-network...

http://papers.nips.cc/paper/7145-deep-learning-for-precipita...

What I think is that Google wanted to use a lighter model that can be applied to the whole continental US. I expect them to integrate this in google assistant, like: "hey google, tell me when it's going to rain"


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: