Buy the parts and assemble it yourself. Should be around the same price if i remeber correctly

Did i read correctly that the user pays for converting csv in their own browser?

the government ids should be enough. Germany has an quite interesting system. Federal issued ids that contain an nfc powered chip that provides only information if the user has the credentials that allow access. The minimum is a datum stored/generated (name, age, fullfills age requirement, service pseudonym, ...) and the maximum is full access without pin requirement. After over an decade i haven't heard of hacks that broke the system. Effectivly it is possible to use it in an offline system that allows access to an venue for someone at least 16 years and all that the system sees is that the owner of the id is at least 16 years old

The "Ukranian" nukes were sowjet. It was part of the deal that the sowjets get them back and Ukraine gets its independence. A lot of countries made that deal. And its a good one. Ukraine had never the full blueprints and capabilities to maintain them, so after a few years they would have gotten relative useless and a liability.

in the Price for an nvidia card is not only the price for the card. Its also the limitations of producing it. The Hyperscalers have much deeper pockets than the consumers, so they compete via the price to buy as much as possible from the limited batches nvidia/amd/... sell. As a result, manufacturers are changing the number of different chip types, so that instead of 4 graphics cards and 1 CUDA card, there is now 1 graphic card and 3 CUDA cards.

My current understanding is that regular consumer GPUs are very much capable if they had enough VRAM, and buying like 2 isn't out of this world.

A 64GB consumer GPU not existing has no technical nor financial (buyer side) reason to it, other than market segregation. If a 24GB one costs 1000 bucks, people are ready to pay quadruple that for a 32GB GPU. And VRAM is the main bottleneck for running larger LLMs, not performance. And some resort to offloading to RAM etc...

Segregation between pro and consumer hardware are in a big part artificial I think, for fatter margins.

Otherwise, yeah Hyperscalers have deep pockets, but they can only have it by getting back money from their users/customers, and we're gonna give them that by using their products and paying for it (indirectly or directly).

I don't think it's a good thing that only select companies gatekeep AI and it feels to me like it's going that way with chip prices.

The european auto industry is in jeopardy, because they still want to milk the ICE. Alternatives are treated as concepts or to circumvent fines.

- MS forcing AI

- the uptime is dropping

Thats what i implemented for myself. I made myself pre commit hooks for running clippy nightly and deploying my standard lints(clippy and non-clippy). Thanks to jq-lang and the jq-wrapper tomlq it can even work around the lints/workspace.lints difference with little problems.

For myself i have forbidden unwrap/expect, since this can lead to panics. For cases where an unwrap IS unavoidable, i use unsafe {unwrap_unchecked}. The unsafe is to make sure i can't overlook it.

I wonder what they have planned for the Chaos communication congress

Homework was only something i had to do, but i remembered nothing from it. It was used to shorten the time for going through the textbooks