Well, Crowdstrike did the excact same thing to Linux servers earlier this year, sending them into kernel panic. https://www.tomshardware.com/s...
What caused the Crowdstrike outage wasn't a driver update, it was their version of a virus definition file (Crowdstrike call it channel files). This gets updated many times per day (can be hourly), is a normal data file pushed directly to the Crowdstrike data directory and read by Crowdstrike software, which requires a live connection to their "security cloud" to function. So everyone saying IT admins should have tested the update first don't really understand how this work, it isn't possible.
Unfortunately this particular definition file contained malformed data that provoked a malfunction in the already installed kernel level agent. And as anyone who has read Crowdstrike's PIR know it was entirely due to Crowdstrike''s test and release process being shockingly lacking in normal safeguards.
Btw. Crowdstrike did the exact same thing to Linux servers earlier this year (Debian and RH), so if this is a Microsoft issue it is a Linux issue as well.
Crowdstrike did the excact same thing to Linux servers earlier this year, sending them into kernel panic with an update.
Microsoft willingly let untrusted, unverified code run in kernel space. Why didn't Crowd Strike have to get their signing changed / updated after they made an update on their product? It's useless to have a signing system for trusted kernel access, then completely ignore when someone who interfaces with the kernel makes changes. Essentially the kernel should have detected the update was bad, because it wasn't signed, and therefore should have ignored it.
Because what caused the issue wasn't code to be installed in Windows at kernel level at all, it was "just" Crowdstrike's version of a virus definition file -- Crowdstrike call it a Channel File -- and it gets updated several times per day. Somehow corrupted data in this file triggered a fault in the already installed agent.
Your link mentions 100,000 year cycles. Don't you think it's a bit alarming that we're seeing 100,000 year fluctuations in human observable time?
This. What people pointing to historic fluctuations being normal don't take into account is both the unprecedented extreme speed at which it is happening now, making it much much much harder for natures ecosystems to adopt. But also, humanity. The difference now is that there are 8 billions humans on this earth, with a fragile and inter-dependent ecosystems. Few are anywhere near food self sufficiency, and fewer still are self-supplied on all the elements and components needed for the products and lifestyle we take for granted. And that is ignoring the effects of mass-migration.
Anyway, I wouldn't trust any output a LLM generates just because it is inherently a very very very advanced prediction engine.
Humans get things wrong all the time too, the bar isn't for the generative models to always be right, but to do at least as good a job (error/quality wise) as an average human doing the same work. There is a reason we have quality control already, also LLMs will need that.
, and service is where dealers make a lot of their money.
EVs also require less service in general, with fewer moving parts and components. The maintenance on electric cars can be significantly lower and cheaper than maintaining a gas car. According to Consumer Reports, in 2020 electric cars cost about $900 per year to maintain, while gas cars cost about $1,200 per year. This is quite a bit of lost revenue for the dealers which is part of the resistance.
So if I get some old documentation PDF and it's unreadable due to unsupported fonts - what do you think I shall do then?
Would understanding that documentation be dependent on pixel-perfect rendering of the design? Because that is what you will lose with substitution fonts.
I've had "5G" internet and it is absolute trash. I'm sure it's "fine" for cell phones, but for normal internet access on your PC it is incredibly slow, not a real competitor to cable, fiber, or even DSL.
It doesn't have to be if implemented well by mobile provider. I'm not in the US but I have seen 1.5 Gbps actual speed (measured download from remote site) over 5G internet, beating that my home broadband is only 750 Mbps. Sure, the cable broadband speed is likely more stable and a bit faster ping (though the 5G ping is surpisingly good), but I'm basically unable to notice any diference on my laptop with built in SIM card whether I'm running on 5G or 750mbps cable even for bandwith heavy loads.
Competence, like truth, beauty, and contact lenses, is in the eye of the beholder. -- Dr. Laurence J. Peter