Blaming the employee for the failure of the company is wrong. The company failed because they didn't have good data management or access controls. If the password was compromised due to being "weak", then the company also didn't have good password controls.

Who did that? Many years ago, I had to get a math minor (which I ended up with a second major) to get my computer science degree. Looking at my school's current requirements, they don't have quite as much math required, but it still has all the calculus, linear algebra, and probability/statistics (with a note that you can add one more math class to fulfill the technical elective in CS and also get a math minor).

I haven't touched an HP switch in an age, but a large chunk of the switch market is owned by Broadcom chips, and most of the vendors take the Broadcom reference OS (based on Linux) and just put their branding and spin on it. There's a Cisco IOS-ish CLI, and then some will let you access a little more of the under-the-hood Linux (like the FiberStore datacenter switches I've used recently have some level of python installed although I haven't messed with it).

Realtek has been getting a foothold at the lower end, and I think their reference OS is also Linux based.

Juniper has been moving to Linux over time. They started with a lot of customizations in their FreeBSD that always made it hard for them to rebase to a new upsteam release and to support new hardware (like they had issues with PowerPC CPU SMP support, so the MX80 and MX104 had multiple cores but JUNOS only ever used one).

Instead they've been basing a lot on Linux (Wind River Linux distro IIRC). The move is two-fold: they've been running Linux on the hardware and their FreeBSD-derived JUNOS in a VM on a lot of platforms, plus in parallel they've been porting their whole system to Linux as JUNOS Evolved.

Who knows what HP will do with all of that though... probably have to pay license fees for every new SFP you plug in (and go back to vendor-locking optics).

Ugh, thinking about on-hold music brought back the old BellSouth Business Repair (800-247-2020) hold tune. I don't think I've called that number in 25 years (looks like AT&T still uses it for reporting line trouble)... the fact that I still know the number AND CAN HEAR THE TUNE shows just how much time I spent with it in the late 1990s.

Building a Linux distribution requires a fair amount of infrastructure, and that's something that's pretty different from distro to distro (and not all make all the necessary tooling public). Changing to a different base distro would most likely require a significant rework, and may be more than someone, especially someone who has mastered the intricacies of one distro's tooling, wants to do.

Incorrect response - some manager will believe that THEIR meeting is SO important that it's worth paying for (and you said it's fine if they pay). Even if they pay time-and-a-half for extra time, it's still almost certainly not worth the intrusion into your life... living a life outside of work is worth more than some extra $$.

Which, that's kind of a privileged point of view, because I've never needed to work more than one office-hours job to survive. I've been on-call, but that's kind of a necessity in my line of work (systems and networks for small ISPs and other Internet companies). I've never worked for a company large enough to have a 24x7 high-level staff (and I probably wouldn't enjoy being in that large of an organization).

It's not all of them. One thing that tends to happen with a major outage is everybody treats downdetector.com as the gospel truth. Then end users try to make their own reports, and don't even know what service they're trying to connect to (like reporting Netflix is down when they try to load Hulu). Then lazy news services just pick up downdetector results and publish stories, which are then picked up and copied by other news services.

What's known: GCP (Google Cloud Platform, not to be confused with Google Actual Cloud Platform) has experienced a major outage. It affected some Google services as well as a lot of customer services. CloudFlare had a significant outage at the same time, although it looks like mostly just that CF's systems got backed up and overloaded trying to talk to the GCP-hosted customers.

[citation needed]

For example, the 3-reactor plant near me hit about 89% capacity for the year last year, and part of the less-than-100% was due to refueling/maintenance outages. The peak annual capacity was 96% in 2017.

Artemis II is not a landing, it's a manned flight to Lunar orbit (similar to Apollo 10). Artemis III is the mission for landing, assuming it still happens.

The Artemis II rocket is being stacked now for testing.

Eh, I have no trouble with USB-C to RJ-45 cables. I had way more trouble with the device end than the computer end before everybody finally settled on the same pinout (at one job we had to have a basket of adapters with every combination of DB-9, DB-25, male, female, and null-modem - that was the real PITA). Use quality USB adapters with genuine FTDI chips and they "just work" everywhere (including my phone and tablet).

CentOS is upstream of RHEL, so it's the other way around.

It depends on the Discord server. At minimum, they'd have an email address that the user controlled at least at some point (to confirm it). Due to spam bots, a lot of servers require a phone number verification via SMS as well.

There are multiple companies that provide bulk SMS sending for corporate use (plus some large companies do it themselves). All my notifications are from such senders - I don't get any email-to-SMS gateway messages. Bulk senders would otherwise get blocked; the email-to-SMS gateways have low rate limits. Even things like restaurants that use SMS to notify waiting customers a table is ready use real SMS senders, not email gateways.

Email-to-SMS (at least on the carriers I've seen) is really obvious and stands out as weird (which no sender wants), and the only time I remember seeing it was when I was testing it for personal use.

Email-to-SMS has always been a lightly-documented, "best effort" (don't call support if it doesn't go through or is delayed), and probably lightly-used service. Due to email abuse, it's never going to be a service that doesn't cost some money to support, so I can see it getting the axe from all the providers. I also wonder if the shift to RCS meant they were going to have to spend development money to keep it working and decided it wasn't used enough.

I've been on T-Mobile for ages, and I've seen their email-to-SMS gateway be unreliable at times, with no real effort apparent to update it.

For my personal stuff, I shifted to Pushover (one-time $5 fee, per receiving-platform but I only use one platform to receive), which a lot of things already support and can easily be called with a simple wrapper around curl. For a couple of email-only things, I have a special address in my own domain and a procmail rule to forward to Pushover.