In round one focus was on the nix command more then anything else.
While we already expected, the need to write flake.nix and the need to
know Nix language very early on almost destroys the nice user
experience.
In this round we tried to explore the idea:
Changing to some known format (toml, json, yaml, ...) how far could we
use Nix and still not need to know about Nix language.
The reason to explore this idea was the realization that learning Nix
language is quite a big barrier. Currently we require to know Nix
language at quite an early stage - right after using nix-shell -p. How
far could we allow new user of Nix to use Nix without knowing Nix and
using a simpler and more familiar format instead. Clearly this is not
going to replace Nix, but only to help on-board new Nix users, until they
reach a problem where learning Nix language is worthwhile of their
investment. And obligatory finish of a paragraph with a cliché quote:
Simple things should be simple and complex things should be possible.
In round one focus was on the
nixcommand more then anything else. While we already expected, the need to writeflake.nixand the need to know Nix language very early on almost destroys the nice user experience.In this round we tried to explore the idea:
The reason to explore this idea was the realization that learning Nix language is quite a big barrier. Currently we require to know Nix language at quite an early stage - right after using
nix-shell -p. How far could we allow new user of Nix to use Nix without knowing Nix and using a simpler and more familiar format instead. Clearly this is not going to replace Nix, but only to help on-board new Nix users, until they reach a problem where learning Nix language is worthwhile of their investment. And obligatory finish of a paragraph with a cliché quote: