My bad I didn’t see that proportion thanks for clarifying. As more than half are already going over the speed limit then there is indeed a more systemic problem. I thought it was about say 10% fringe that go with heavily modified bikes. I’m not sure more tech would help though, rather than fines with explanation of the risk until people do start respecting the limit. If people are unable to respect that and it causes more accidents, then yes ISA on e-bikes, cars, everything causing accident on the road.

They can, cf https://en.wikipedia.org/wiki/Intelligent_speed_assistance with example in this thread

Apparently starting ISA in July 2024 https://www.brusselstimes.com/855280/self-braking-cars-all-new-cars-to-be-fitted-with-speed-limiting-technology

at least easier to find those who do it and fine them.

Missed that part, can you please clarify how?

I can’t see why this is not a good idea

I believe the argument here is that it’s security theater, i.e it looks positive but in practice has literally no effect. To clarify if people buy a “normal” e-bike today, they are already speed limited. Consequently people who have bike going faster that said limit are doing something already beyond the ordinary. The likelihood that such people would suddenly change their behavior to buy typical bikes when they have even more restrictions is probably not high, but the announcement still makes it look like something is done for the greater good.

Genuinely no idea how Linux gaming could be better. I’ve been playing on desktop and Steam Deck for years, both “flat” games and VR games and it just works. Sure I don’t try literally everything but with ProtonDB I’m confident it will work, or not, and decide accordingly. Obviously not all games work on Linux but definitely more quality games that I have time for. For me it just works, I spend at least 99% of my time gaming on Linux actually gaming, in fact I can’t even remember when is the last time I tinkered. I don’t even have problems with GPU drivers despite tinkering with containers with machine learning. I’m not trying to say nobody has problems or dismiss problems people do have, just sharing my experience.