This is mostly a question out of curiosity, I’ve been using stable for my own configuration with a few select packages from unstable whenever they are broken or missing something vital to me. I’m curious as to what other users of NixOS are using for their default channel.
0 voters
Just want to say that your results will likely be very skewed, as this discourse is probably frequented most by contributors, who probably live on unstable.
I know when I first started with nix, I was on stable for a months. It wasn’t until I began contributing that i moved to unstable. If I chose to go with another community, I probably would have remained there.
No newbies here? I’m the only one? 
I can’t say who is, or isn’t here. But from post frequency, it’s around ~20 very active contributors/people, then drops of quickly. But that’s common in most forums.
Is there a nice chart of this available on the discourse UI?
I was speaking from personal observation. A lot of different people may ask questions or give input, but there’s a small active community which answers the questions.
I have a Radeon 5700 xt, which necessitates unstable. Otherwise I’d run stable with some unstable packages mixed in.
I run stable on all my systems. On my primary desktop I switch to the next release when it hits beta. For my personal systems I generally import a few packages and modules from unstable, but I try to avoid pulling anything from unstable on my servers at work.
I don’t run my machines off a channel is this bad (always wondered)? I use nixops for all my machines that way all my machines are pinned to the same nixpkgs commit. I update about once every month and usually pick the latest on unstable.
If you want exactly the same environment, then I think this is the only option.
I wonder if it would be possible to get this information from the binary cache. Tracking the number and date of download for a core package like stdenv maybe ?
Once every month could mean you’re exposed to security issues for up to a month. It really depends on your attack surface.