3 ms·
I’ve spent time with John and with the FrogNet work, including the simulator and the end-to-end use cases behind it. This isn’t a landing-page concept or a mock
by danieltone 3mo ago
I’ve spent time with John and with the FrogNet work, including the simulator and the end-to-end use cases behind it. This isn’t a landing-page concept or a mock-up: there is a working multi-node system, real application paths, shared/transient state, and a serious attempt to make constrained links useful for application traffic.
I think the right way to approach it is exactly the way John is inviting people to approach it: challenge the architecture, assumptions, measurements, security model, compression behavior, and deployment claims. There are ambitious claims here, and they deserve rigorous review—but there is also substantial engineering behind them, not just a pitch.
The aspects I find most interesting are the effort to make a node independently useful when disconnected, the transport-agnostic approach, and the semantic/structured-data path for low-bandwidth links. I’m looking forward to seeing the code and documentation receive the kind of public technical scrutiny that can either validate the design or materially improve it.