Keep in mind, RHEL backports and or cherrypicks commits from upstream into their kernel; so the 4.18 in RHEL8 is plus_plus and minus_minus 4.18 upstream

Just because we can MITM sms doesn't mean we should.... I'd like to at least marginally keep believing sms-based MFA is still useful and borderline secure, and I'm pretty certain we keep showing various companies and organizations keep loosing access to systems, exposing default credentials, etc.

Until we realize that building secure systems is actually really hard, and we can't just glob on security. There is more to security then making sure systems are updated regularly, audits are performed, and absurd password requirements are met. The GAO report on the SEC's systems (https://www.gao.gov/assets/690/686192.pdf) had these 2 recommendations: (1) Maintain up-to-date network diagrams and asset inventories in the system security plans for GSS and a key financial system to accurately and completely reflect the current operating environment. (2) Perform continuous monitoring using automated configuration and vulnerability scanning on the operating systems, databases, and network devices Stop rushing software and services to market, understand your software threat / overall architecture. Companies need to understand their applications, dependencies, attack space, how to actually implement security, and what having a secure system means. TLS and SSL doesn't mean anything if I can compromise an endpoint and now I (as an attacker) have access to your keys and can feed whatever I want into the pipe. Can you MITM your data access / caching layer? Where all can attacks come from and what is the impact at each level? What components are you using in the architecture? There is not a silver bullet to security. Sure, firewalls, IDS, password requirements, logging, encryption, etc. all help. They don't do anything if they aren't configured correctly, or they are bypassed, or some aspect of the system is wide open.

Was he (the passenger, and vehicle owner) keeping his hands to himself or were there also other distractions contributing to an already dangerous situation? Doesn't "insane" mode on the Tesla have to be specifically enabled? Was it enabled? Can they pull the black box and even find out if insane mode had engaged during the fatal trip?

The Internet itself is fully redundant. The Internet backbone can survive several major failures. E.g. take 9/11 for example, much of the regional fiber for NYC and the surrounding areas, ran in near vicinity to the WTC, of course this is going to cause small regional outages, however down in DC or over in Chicago, the backbone was fine and handling it's daily activities. The problem with redundancy and availability is diversity. Most content providers do not have regional data centers that advertise a single BGP route over multiple providers to their content, in most cases a provider of content will have a single data center and a single or possibly two distinct feeds to the Internet. This is obviously going to cause some problems. The Internet is only as redundant as the content providers make it. As far as the Bandwidth Crisis? it doesn't exist. This is hype. Sure, some of the dark fiber may not be getting used, but for the most part there is still plenty of fiber to go around, and the conduit is laid for the expansion of more fiber, glass just has to be pulled through the conduit and lit. It's expensive though to pull new fiber, terminate it, light it and connect it to a backbone router. The backbone routers such as say the Juniper T640 are not cheap! The problem is who pays for the routers and cross connects? What do we do do when every home has gigE fiber to the front door? Right now the Internet backbone runs at roughly 10gbps, although providers will generally trunk multiple OC-192s together to create large virtual pipes. It is currently cheaper to run trunked OC-192s instead of increasing the pipe "size". There is no money in the innovation (physics/electronics/etc) of networking in order to drive speeds faster. The only shortage we have is money and who is going to fund massive fiber rollouts, router upgrades, and innovation. As money flows into the universities, we will get faster pipes as folks devise ways to pack more light and encode more bits into a single strand of fiber. Want to eliminate the shortage? Fund the universities and innovation, or stop greed (but that just seems impossible)