Brake fluid is hygroscopic and will accumulate water over time despite being in a nominally sealed system. Water in solution with brake fluid noticeably lowers the boiling point which leads to issues under repeated braking (e.g. down long steep hills) as the fluid boiling means you lose braking capacity in that circuit.
You should ideally be changing the brake fluid every few years (2-3 being the typical recommendation) and that applies even if the brakes are used less often.
Electric cars do have oil that will ultimately need changing but it’s less exposed to contaminants than the engine oil in a internal combustion engine and therefore will last a lot longer. EVs typically have a reduction gearbox and differential and these will require oil changes in a similar fashion to a manual gearbox or differential in a ICE vehicle - i.e. barring exceptional circumstances it will last long enough to get out of warranty but don’t believe it will never need changing.
I think it was 2007, the family upgraded to a 3G modem when Telstra got around to putting up a tower that provided mobile reception where we were living. I was pretty happy as with the quality of rural phone lines we weren’t even getting the full potential of dial up (maxxed out at 30 ish kB/s).
Of course the next problem was trying to keep under the tiny download caps of the time, I remember having to wait until the end of the month (when usage was about to reset) to download large files or risk having my parents and siblings annoyed at me for using up all the quota…
Chirp ran fine on Linux when I needed it to program a UV-5R a year or two back - was provided in a flatpak then but looks like they use a Python wheel file now.
It’d be interesting to see how much this changes if you were to restrict the training dataset to books written in the last twenty years, I suspect the model would be a lot less negative. Older books tend to include stuff which does not fit with modern ideals and it’d be a real struggle to avoid this if such texts are used for training.
For example I was recently reading a couple of the sequels to The Thirty-Nine Steps (written during WW1) and they include multiple instances that really date them to an earlier era with the main character casually throwing out jarringly racist stuff about black South Africans, Germans, the Irish, and basically anyone else who wasn’t properly English. Train an AI on that and you’re introducing the chance for problematic output - and chances are most LLMs have been trained on this series since they’re now public domain and easily available.
Closest I’ve come in a truck is an annoyingly loud alert for everything the computer reckoned was an issue and that was painful enough. Every time I’d drive it it’d be blaring the alarm for some reason or another and if it had been a long term company truck instead of a rental I probably would have ended up removing the speaker.
For example the lane departure warning would fire off every time you moved over to not run into someone parked on the side of the road, the close distance warning would fire off regularly when people merged in front of you, and if it was windy it’d set off an alarm to let you know the truck was being blown around when driving. Could be useful if you’re mentally challenged or blind but that sort of thing is just going to annoy anyone who isn’t. You couldn’t even turn the alarms off properly - you could go through the deliberately prolonged procedure to turn them off temporarily but then they come back again every time you start the truck.
I’ve driven an SUV with lane keep assist and it would pull at the wheel trying to follow lane markings that were outdated or ones it just made up, I hope that particular bit of ‘safety’ tech doesn’t make it to any truck I have to drive.