docs(wasi): 04-04 notes

Author: ricochet

wasi/2024/WASI-04-04.md

@@ -30,4 +30,73 @@ The meeting will be on a zoom.us video conference.
 1. Proposals and discussions
     1. Vote: Advance `wasi-i2c` proposal to Phase 2
     3. _Submit a PR to add your announcement here_
-    
+
+### Attendees
+
+- Bailey Hayes
+- Luke Wagner
+- Colin Murphy
+- Friedrich Vandenberghe
+- Dave Bakker
+- Kevin Moore
+- Marcin Kolny
+- Joel Dice
+- Hung-Ying Tai
+- Stephen Berard
+- Yong He
+- Merlijn Sebrechts
+- Dan Phillips
+- Zalim Bashorov
+- Slava Kuzmich
+- Nuno Pereira
+- Ralph Squillace
+- Warre Dujardin
+- Johnnie Birch
+- Till Schneidereit
+- Jorge Prendes
+- Ashok Kumar Krishnamoorthy
+- Taylor Thomas
+- Wouter Hennen
+- Fedor Smirnov
+- Robin Brown
+
+### I2C implementation 
+
+**Friedrich Vandenberghe**
+https://github.com/Zelzahn/i2c-wasm-components
+Running implementations
+https://crates.io/crates/wasi-embedded-hal
+https://github.com/Zelzahn/wasi-i2c
+
+**Bailey Hayes**: Microcontroller might be a bit vague in the proposal
+
+**Chris Woods**: Let’s have a definition of this
+
+**Friedrich Vandenberghe**: Raspberry Pi Pico
+
+**Robin Brown**: ESP32
+
+**Till Schneidereit**: is a Cortex M0+ too big?
+
+**Emily Ruppel**: M0+ is one of the smallest ARM configuration. Which microcontroller might not be the right way to frame, instead we could focus on which OS. Perhaps WAMR or Zephyr
+
+**Thomas Trenner**: Zephyr is reasonable.
+
+**Stephen Berard**: I think we should likely select a few reference platforms and agree on OS/board/HW config. We should agree on a Dev board, OS, and HW configuration. Should be something small, much smaller than Cortex A53.
+
+**Friedrich Vandenberghe**: Will file Github issue. Define portability criteria. See https://github.com/WebAssembly/wasi-i2c/pull/5
+
+**Bailey Hayes**: Is there a use case proposal?
+
+**Thomas Trenner**: UART and similar spaces
+
+**Bailey Hayes**: Let’s keep the discussion going in the Github issue.
+
+**Till Schneidereit**: Since at least some of these microcontrollers have the option to run without one: would people see it as a requirement to run without an MPU as well?
+
+**Ralph Squillace**: Matrix of platforms, OSes, and configurations would be useful.
+
+**Robin Brown**: Github discussions might be better suited than an issue.
+
+**Bailey Hayes**: The proposal champions will decide if they will use discussions or issues. Decision to delay the vote until portability criteria is defined. 
+