My servers are one NUC clone and a 4*16tb NAS. I have a lot of docker containers running constantly and yet cooling has never really been an issue for me. A larger concern is I would rather not see it, so It’s hidden it under furniture. The fans on the NAS have attracted a layer of dust, and one day I might clean it. Kidding. I wont.

My security team involves a bull dog named Sophie, who has never done more than lick any other being, but I’m banking on burglars not knowing this.

I’ve played Ker Nethalas and a little bit of Thousand Year Old Vampire. I enjoyed both of them, but in both cases I found myself using Obsidian and later foundry VTT to manage the game. I imagine for some people the goal might be to move away from a computer, and in which case, my experience may not apply.

My experience was positive, but I would like to explore some other games. Ideally something where the randomisation utilises playing cards rather than charts and dice (which was the case with KN). I know KN has cards you can buy, but for some reason they weren’t appealing to me at the time. Perhaps incomplete or too expensive… I can’t remember.

I’m a dev and I think my experience is mostly similar to yours. Where AI seems to work well, is when the boundaries of the problem are very well defined. For example : “Take this C++ implementation of the LCS algorithm and convert it to JS”. That would have taken me a few hours at least, but AI appears to have nailed it. However, anything where a large amount of context is needed and it starts to fail fast, and suggest absolutely insane things. I have turned of copilot on my IDE because it slows me down, but I will still ask questions to chatgpt when I have a specific problem I think it can help me with. I also will ask pointed questions when I review other dev’s code, and my expectation is the author can explain why they wrote it.

I use authentik and like it. The learning curve isn’t that steep so not a lot of wasted investment if you decide to ditch it for something else. No password flow with webauthn is pretty cool.

Hahaha. She’s not wrong.

IMO, you want ram more than you want processing power. 16 gig ought to be enough. Most of the time your containers will sit dormant and just consume memory. However since you want to run Jellyfin, get a recent CPU which can do hardware decoding of popular codecs. There’s charts online that show what generation can handle what codecs. Ideally you don’t want that done by software. You should still be able to find something cheap.

In terms of placement. It depends a lot on noise IMO. If you’re running something small without magnetic storage, you’re probably fine to stick it anywhere. If you have several data-centre grade hard drives, you will probably want to keep it somewhere where you wont hear it all day.

In terms of upgrading, I’m not sure if its as much of a concern as you might think. I run probably about 30 docker containers off a NUC clone and a seperate NAS, and that has worked pretty well for the last few years. I can always add more drives to the NAS, but otherwise its fine. Also, many of my services scale to zero with sablier+traefik, and I schedule filesharing for low bandwidth times. This makes things pretty manageable.