Reference tooling and notes for measuring gesture counts in multilingual input method experiments.
This is a curated public portfolio copy of the active
GESTURAL_EFFICIENCY_ANALYSES project. It includes the analysis scripts and
dissertation-facing explanatory notes, without local project-management files or
large/generated data.
The project computes gestural-efficiency statements such as the cumulative share of frequency-weighted text available in one gesture, two gestures, three gestures, and so on.
The current research direction measures concrete GridPack-style input packages instead of relying only on older simulated input-method models. For each sendable text candidate, the analyzer can compute a least-gesture path and aggregate the result against a lexical frequency source.
tools/gestural_efficiency_statement.py: analyzer for tooltip rows, frequency rows, and actual manifest-plus-shard GridPack packages.tools/build_dissertation_section.py: helper for generating dissertation-facing analysis text from accepted report values.dissertation_gestural_efficiency_section.md: current explanatory source notes for the dissertation.sources/README.md: source inventory and provenance notes.
The public copy is intended to show the analysis model and implementation shape. Some input data and generated package artifacts live in separate private or local research repositories and are not included here.
The Python files in this public copy were syntax-checked with python -m py_compile.