I dont have any specific llms to recommend but if you do want to go that route you could always run it remotely through something like Google collab.

But I don’t know if I would trust the results of an llm doing this as any mistake would make your entire resume untrustworthy

And for context, it does this because cheaters are willing to run cheats that run at that kernel level, and the only way to detect and prevent them is if the anticheat is in your kernel first.

Correct, though to be pedantic anyone can be a CA- you just generate a cert with the right bits to say it’s a ca certificate and then use it to sign any other certificate you want.

But the only devices that will consider your signature worth anything are ones you also install your ca certificate on. So it’s useful and common in internal networks but isn’t really what is being asked here.

The hard part is getting in the root CA store of operating systems and browsers. As far as I know they are all maintained independently with their own requirements.

then I would install one

just say I help improve ad blockers on YouTube and refuse to elaborate.

They don’t provide stuff for free, they provide stuff in exchange for your data and to sell ads.

rather than allowing edits for invisible edits for X minutes, couldn’t your client just delay actually sending it for X minutes allowing to cancel or edit freely until that point?

Gmail allows a similar feature and it seems safer in a distributed system than relying on everyone else to respect what happens after you send a raw message and an edit right after

The average user uses sleep mode and wakes from sleep. Sleep mode should be under 10w, or around $1/mo.

though laptops are notorious for proprietary charging.

I’ve seen dells that can charge via USBc at full 140w but only on a Dell dock. On any USB PD charger it will only do 60w, and complains about it as it throttles everything.

well yes, but it’s profitable because customers continue to buy their products and services.

the problem with blaming companies is none of them do this out of desire to hurt the environment. they do it to meet customer demand.

as an example imagine if we all stopped buying gas from Shell. their environmental impact would plummet…and their competitors impact would go up as we continue to buy the same amount of gas from other companies

growing it like a garden is a perfect phrase imo

because on windows or Mac it may have just worked. …until it doesn’t, or leaves your windows scaled wrong or placed on monitors that don’t exist or some other failure condition. at which point you reboot and hope for the best.

when it doesn’t work on Linux I’d check logs, actual configuration, and even the source if I need to.and then I’d hopefully improve things and make it work the way I want it to.

even just knowing enough to not consider clothes ruined when a button pops out or a tear forms would be nice

If adopt systems then the question is easy to answer: no, journald does everything you need.

without adopting systemd… well. Are you evaluating going without any log handling at all and maybe just dumping logs ephemerally to tty0? DIYing all log stuff like your init scripts DIY things?

Personally if I had to go without journald I’d probably go back to using syslog-ng. But I guess there’s an argument for shipping straight into something like opentelemetry-collector if you’re willing to put in a lot of work.

If anything the gap is bigger than ever as the top end shoes are basically performance enhancers like the nike airflys used to set most records…and their new vaporflys being banned in the Olympics.

I guess it’s better than hyper expensive shoes just being a paying for a brand thing?

Before launching products*

walled gardens are only a little less awful when still supported

it’s typically up to the distribution to configure things like that, and many Linux distributions do come in both server and desktop or workstation variants like Ubuntu desktop vs Ubuntu server, or RHEL server vs RHEL Workstation

I can’t say how well they tune these things as I haven’t ran them personally, but they do exist.

You should look into IPMI console access, that’s usually the real ‘only way out of this’

SSH has a lot of complexity but it’s still the happy path with a lot of dependencies that can get in your way- is it waiting to do a reverse dns lookup on your IP? Trying to read files like your auth key from a saturated or failing disk? syncing logs?

With that said i am surprised people are having responsiveness issues under full load, are you sure you weren’t running out of memory and relying heavily on swapping?

I think starlink is more than that as even more things rely on a (good) Internet connection ingeneral than rely on satellites, and traditional connectivity methods leave many people underserved even in countries like America let alone the world.

It definitely has its problems, if nothing else that it’s privately owned and anyone who wanted to compete would then massively amplify those problems.

it sounds like the unlikely outcome of two reasonable policies.

1. you might not get back the device you send in - say it’s a simple broken screen and they’re willing to cover it. its easier to just send you an already refurbished identical model and then toss your phone into the queue to be fixed later.

2. unauthorized parts may violate your warranty and whatever you send in isn’t going to get repaired.

They should still just return it. but if you know it’s not covered you shouldn’t really send it in and it makes sense to cover their ass policy wise even if they do make an effort to just return them.