My server is 2 GB, and for self hosted stuff it's good enough. Granted, I use Gentoo, but I wouldn't expect debian would be much higher unless you're running something.
I run postgres and a file server on mine. I've previously ran NextCloud. It was a liiittle slow, but I think I was CPU bound (raspberry pi).
The big choice here is: do you want rolling release or stable? Most servers are stable, but endeavour is rolling.
Stable release means your distro's repository's packages rarely change behavior. This is because they lock the versions and only bring in security updates. Pros: things will almost never change when you update, which means you won't have to be fixing things when updating (unless updating major version of your distro). Cons: you're stuck with frozen versions. Those can be years old sometimes. As long as you're okay with not using new versions of what you're using, you should be fine.
If you like stable, go with debian stable for 5 year release / update cycles, debian testing for 2 (or Ubuntu server), or the red hat one for 6 months (I forgot its name).
Rolling release means you have to update frequently and you always get the latest version, remaining very close to upstream. This is unpopular for servers as it means an update might bring changes you don't expect, and you might have to change a configuration or maybe even more. If you like this, OpenSUSE tumble weed is a good choice, or good old Arch.
You also have Gentoo and NixOS, but I don't know if you'll be wanting those.
The size of the code still impacts your deployment. Moreover, if you're using something like AWS lambda, small changes can have significant influence on cold start time.
I agree that an extra step is not desirable, but this would only be done for production deployments (and consequently pre-prod if you do that).
Ideally you'd only do this for live deployments (production and possibly pre-production or staging / QA). For all other testing, you would keep it unbundled.
Book reader on android.
Scala has some of that effect too.
I personally don't fully agree. Libertarianism just doesn't work at all. It is not even a complete system from a logical sense. It falls apart when faced with basic scrutiny, or they just theorize a system that's basically the same as a central government but with a private entity name stamped on it.
It is an ideology stemming from a basic principle, but they sadly don't seem to think of the entire system as a whole.
Government is only part of the problem. When they fail to see the ruling class behind it, they don't get too far.
To be fair, nix is not super hard, it's just that its more than your typical distro. You'll run into rare compatibility issues. Yes, rare, but if you're not a tinkerer, you may not like it.
This is false, even by the CIA's own admission:
https://www.cia.gov/library/readingroom/docs/CIA-RDP84B00274R000300150009-5.pdf
You must be speaking about the USSR's early period, transitioning from a rural backwater into an industrial power house. They experienced a famine then (and unfortunately it was the routine even before communism), but once they completed collectivication, there no longer were any. In other words, communism ended the pattern of famines in Russia and Ukraine.
I see it as the fact that we already has more than double the capacity to feed everyone, yet we still choose not to.
This is how we will likely end up paying for services AND STILL having our data sold. It's just the nature of capitalism. Businesses have to grow, and in today's world selling data is always the natural progression towards it.
A parallel example is streaming services starting out as "tv but no ads and on demand!" or "just pay for the service and you won't see ads!" but now we are paying and there's still ads.