Even without toxicity, as a domain expert, I'm cynical about community experiments in this space in general.
Hobbyist "mesh networking" communities readily adopt suboptimal modulation, frame specification, packet format, encoding, ineffective mesh arrangements, flood-prone routing mechanisms, poor reference hardware, and bloated user interfaces, while often ignoring basic practices like forward error correction, collision detection/multiplexing, trust mechanisms. Community solutions tend to be opinionated in the worst ways ("we assumed you'd use an arduino" or "here's a mandatory packet header field to indicate whether you're currently in a hot air balloon") and under-constrained in areas where opinion would be helpful "our client UI has 75 configurable options -- get one wrong and you won't be able to communicate."
By the time there is a "community" it's too late to address these issues: network effects have already locked in the worst design decisions.
Hobbyist "mesh networking" communities readily adopt suboptimal modulation, frame specification, packet format, encoding, ineffective mesh arrangements, flood-prone routing mechanisms, poor reference hardware, and bloated user interfaces, while often ignoring basic practices like forward error correction, collision detection/multiplexing, trust mechanisms. Community solutions tend to be opinionated in the worst ways ("we assumed you'd use an arduino" or "here's a mandatory packet header field to indicate whether you're currently in a hot air balloon") and under-constrained in areas where opinion would be helpful "our client UI has 75 configurable options -- get one wrong and you won't be able to communicate."
By the time there is a "community" it's too late to address these issues: network effects have already locked in the worst design decisions.