Hacktoberfest 2020 + Nerves

Do you like Hacktoberfest? Also enjoy working with Nerves and want to contribute?

Fantastic! :tada: :beers:

Here are some potential starting points for you ¬

  • Nerves docs
    • Its boring, but things move fast and its a huuuggee help to get little updates when things look wrong. Or if guides are incorrect, etc
  • Search Nerves, NervesHub, and related projects by issue tag:
  • Add utilities to toolshed
    • Add df - see #41
    • Anything counts, but things for cpu utilization would be cool.
  • Do anything related to Erlang tracing and write up examples for Nerves people. (Frank thinks we rely way too much on logging, but he’s painfully aware that tracing is hard.)
  • BlueHeron examples
    • small Bluetooth projects that people can replicate
  • Find some sensor/i2c/spi device that doesn’t have an elixir lib and build one
    • Its totally random, but helps when you need it. These can range from really really simple to very complicated implementations
  • Circuits.GPIO pinout task
  • Help push circuits_cdev over the line to use char dev interface with GPIO
  • Speak another language? We want need help putting translation pieces in place for NervesHub

Just curious - doesn’t cpu_sup is what you need? It is out there ready for use.

I feel I should respond since toolshed is a project of mine and the reference was to adding a cpu utilization helper to it.

Many helpers in toolshed delegate to Erlang/OTP functions to aide in discovery, muscle memory, ease-of-use, or to make the output easier to read. The point hasn’t been to bring in new features, but more to expose existing functionality to lessen the need to drop to a bash prompt to get work done. That’s important on Nerves, since it’s frequently the case that Erlang/OTP provides more capability than what’s available in a shell due to the minimal Linux userland.

Anyway, back to a CPU utilization reporting utility, I think it would require a few iterations to figure out an actually useful helper. I agree that if it doesn’t provide some advantage over cpu_sup that there’s no reason to replicate it. I’m not sure this is a good Hacktoberfest PR opportunity, since there will probably be discussion and iteration that would drag the PR out past the October window.

1 Like