Some time ago, I ran my simple server on the Windows Subsystem for Linux (WSL) and it could communicate with a normal Windows browser (Chrome). But my Windows laptop is broken right now, so I don’t have a way of verifying that this still works (it should, of course).
What I don’t understand is why you get this error: Timeout in read from <stream>(0000013f7b324c20) instead of some kind of “can’t open file” error. Perhaps there’s a strange cascade of exception handling. When I try to reproduce this on Linux (by setting the --static option to a non-existent directory), I just get a failure message, nothing about read_util:read_line_to_codes/2. Did you enable some other debugging, to get this? (I tried putting a spypoint on http_files:http_reply_from_files/3, but that didn’t get anything)
I think that could be the choice of the folder where you downloaded and expanded Peter’s example repo that could make a difference. This morning on Windows 10 I did