THALis: experiment design before the working pipeline

Need to decide what i'm asking the hand to do first... finger classification / continuous position / tracking a state are pretty different targets. hardware, signal, model, evaluation — split these up before changing everything at once

timeline evidence: the project journal labels September–October 2025 as an experimentation/failed-attempts period. It does not enumerate a dated sequence of failures in this section, so no specific experiment is assigned to this date.

reference questions for reconstructing this period: what was the intended decoding target—discrete movement class, continuous position, or a dynamical state? What recording arrangement and labels were available? Which failure concerned hardware, input quality, model fit or evaluation?

later implementation map: EEG streaming and preprocessing; fixed-window features; CSP/LDA classification; cue-driven collection; offline topological analysis; simulated hand output. Use those as separate subsystems when locating earlier sketches, rather than describing every unsuccessful attempt as a failure of the final theory.

evidence boundary: this is an index into a broad development period. Additional brainstorm and study documents are catalogued separately; their specific ideas should retain their own source and date evidence.

PyAudio missing portaudio.h error