This is the kind of attitude that drives people away from open source.
Yes, people should read the manual, but at some point they will have questions, and there are a lot of projects that aren't clear on certain things. Such as YAML changes.
Syncthing itself is fine. Syncthing-Fork, a completely separate project that wraps Syncthing into a neat app for Android, is what's going through the repo drama.
Besides - it looks like the new repo owner is pretty transparent about the whole thing and appears to be making good-faith efforts to keep the original Syncthing-Fork devs involved.
I mean, 35W maximum is still incredibly low. At that point, you're looking at a cost difference in the single-digits over the course of an entire year.
My little lab has 5 machines, 3 of which are tiny/mini/micro PCs. Total draw from my entire setup, including the t/m/m machines, is right around 100W. And since I started measuring it back in February, it's used a total of 635 kWh. And most of that is from the ~spinning rust~ hard drives. For reference, my whole household's monthly usage averages around 1200 kWh.
I haven't figured out the whole high availability thing yet; I just move VMs/containers to different nodes if I need to bring a node down, or shift resources around, or whatever.
A 7th-gen i7-powered tiny/mini/micro is perfect for HA. Plenty of grunt for lots of HA addons and integrations, lots of USB ports for dongles (zigbee, z-wave, etc), often with 2x M.2 slots (usually one B/M key and one A/E key) and SATA interface, very low power draw, and cheap due to businesses offloading them all the time.
I have a Proxmox cluster and still went with a separate machine for HA. I figure the thing controlling my house should be on it's own, since the cluster is more a playground for me than anything else.
My Wyze is based on an ODM unit, the 3irobotix CRL-200S. Companies like Wyze, Xiaomi, Viomi, iLife, Conga, and other brands customize and sell it as their own models since that's cheaper than manufacturing their own units. Parts are swappable between them as they are all the same robot underneath... Kinda like how car companies rebrand models based on region. As far as I'm aware though, iRobot builds their own robots.
This is the kind of attitude that drives people away from open source.
Yes, people should read the manual, but at some point they will have questions, and there are a lot of projects that aren't clear on certain things. Such as YAML changes.