I had enough troubling discussions with tech people that believe that these language models think and we are within moments of sentient computers.
Many variations on “we are just like the machines; we read and speak statistically valid sentences”
Folks are quickly dismissing that turning your complex inside thoughts into words is usually an art that requires work and practice.
These models are great tools, but I feel the cringe creep up when folks bless these models with supernatural powers.
@tcurdt the closest thing I have to “bare metal” is an AWS EC2 instance.. So I have an AMI pre-rolled for all my servers…
For all other situations, I just DD the image on a device (real or virtual) and let it resize itself on first boot
Just “inherited” a bunch of Cisco 2701 access points (in a frozen warehouse at -25C, I’m not going to swap them out until they die) and suddenly remembered why for the past decade I only used Ubiquiti…
The software stack (controller and device os) is seriously a pile of 💩, it’s like being stuck in the 90s: c’mon, who thought that TFTP would have survived till 2023???
The virtual controller runs only on Ubuntu 14, and even with that it needs serious hacks to run the network in “legacy” mode… 🤦
@jwz i second the suggestion for JAMF Now… Cheap and rather easy to use if you only have to manage some iPads (we then “upgraded” to JAMF PRO as we also had a ton of Macs and needed bonding to Google Workspaces).
ABM should do more or less the same, but it wasn’t available in EU when we started a year ago, so I can’t really tell…