A large percentage of the work required already exists as a part of existing extensions and CLI tools. Deciding how to best tackle these improvement is a good start.

Option 1

See if we can add a PlatformIO config to support Rust.

Option 2

Create an all in one extension like PlatformIO to create an integrated IDE experience

Option 3

Only create an extension with the bits that are missing and are unlikely to be added by others

Improvements required:

• Semi Automatic Board selection
• Board Manager (IE Download Board Configs)
• Memory Management GUI (Optional)
• Clear Check, Test, Build, Program Buttons
• Serial Monitor (IE Comms through the same cable that it is programmed with)
• Quick Flash (Like cargo install, but for embedded) (Designed for Bootloaders)
• Crate Management GUI

There are many crates that could be used on embedded that aren't in Awesome Embedded Rust.

Doing an audit would be a good start.

Motivation

Making RustEmbedded accessible as part of Patina would require a single, go-to hardware board that everyone is using (much like Arduino).

ToDo

• Make sure all requirements are collected and specified
• Identify and collect existing boards
• Create a feature matrix for boards + requirements to filter down candidates

Requirements

Must-Haves

• Hardware must be Tier 1 supported (Rust Platform Support List).
• Price needs to be about $25-45 USD.
• Hardware must have a comprehensive Board Support Crate
  • The BSP must come with examples
  • The BSP should have clear references to up-to-date training resources
• Minimal RAM: TbD
• Minimal ROM: TbD
• Minimal CPU Frequency: TbD

Nice-To-Haves

• Open sourced and cheap clone-able
• Wireless communications
  • If included, a complete and easy wireless stack is going to be needed
• QEMU support for the respective architecture to allow testing as a first step

To Be Discussed

• Debugger or not?

Collaborations & Partners

• Working with an existing supplier is fine as long as branding is clear
• Ferrous Systems offered to help in maintaining a "Ferrous-Supported Platform"

Concrete Board (Families)

• Sparkfun Artemis
• Adafruit Feather
• PJRC Teensy
• Potenta H7
• ESP
• nRF:
  • 51: micro:bit, calliope, ...?
• STM32 (e.g. Discovery, BluePill)

Feature Matrix (Sketch)

Copy matrix styling from Zola

Motivation

In order to get people intro RustEmbedded, there must be resources (guides / tutorials / ...) for various levels of knowledge.

Tasks

• Identify "Personas" that want to learn Rust & Embedded (e.g. high school students, University students, young professionals)
• Structure material to fit these different audiences
• Bring collaborators onboard (e.g. teachers, lecturers, trainers, ...)

Inspiration / Related Works

The EmbeddedRust WG already has two books "in print" that aim to fulfill a similar need:

• Discovery Book
• Embedded Book
• micro:bit Rust Book

While these books are great resources, they lack the following due to missing (human) resources:

• Regular updates to keep the material in sync with the state-of-the-arts crates. Examples:

  • The Discovery book has a custom HAL and examples cannot easily ported to the "production" HAL people shall be using
  • The books use tooling that is partially replaced by probe-rs.

• Multiple languages (especially for beginner material it could be essential to have a couple of things in the native language)

One of the main problems is that the initial work is unlikely to result in any big influx of people. As such, we need to formulate a plan to ensure that we both make progress towards a common goal and do things that provide value to as many people as possible.
I believe that we should focus our efforts on things not already in progress (other than ensure our requirements are met by those projects).

Patina was suggested in the comments as a good name that isn't already in use. As I like it too, the initial project name will be Patina.

Notable omissions of high level traits (as not reported in Awesome Embedded Rust):

• Storage
• Sensors:
  -- Gyro
  -- Compass
  -- Magnetic
  -- Current
  -- Voltage (?)
  -- Temperature
  -- Humidity
  -- Pressure
  -- Encoders (EG Speed, Angular Location)
  -- Capacitance
  -- Frequency/Time
  -- IR
  -- Imaging
  -- Other sensors as appropriate
• Motors
  -- Direct Drive
  -- Servos
  -- Steppers
  -- Brush-less
• LEDs (?)
  -- Bi, Tri, Quad Colour
  -- LED Arrays
  -- IR
• Analog Outputs
• Communications
  -- Generic Layer 3 and higher protocols like Modbus
  -- USB
  -- Ethernet
• Bootloaders

AVR now has Tier 3 Support.

It is unclear if it is going to be suitable for beginners or if we need to steer them towards something in the tier 2 list.

The first step is to have a play with the current AVR support on a number of Arduino (and related) and see how easy it is and if they work as expected.