Voice notes as a second-brain ingest layer

Average typing speed for adults is 35-50 wpm. Average speaking rate is 120-150 wpm. That's a 3x compression ratio between thought and capture - exactly the kind of difference that turns 'I'll write that idea down' into 'I'll forget that idea'. Almost every second-brain methodology underestimates this. Tiago Forte's 'PARA' assumes the bottleneck is organisation. The actual bottleneck is the front door.

Ideas that wouldn't have made it past the typing tax now make it into the system. You capture on the train, in the queue, on a walk, in the shower (well - in the bathroom afterwards). The capture step is small enough that you don't curate. The curation step happens later, when you skim the transcripts and decide what to keep, what to expand, what to delete.

Pass one: capture. Tap the green circle, talk, tap again. No editing, no thinking about taxonomy. The transcript appears under the note. Pass two: weekly review. Open the notes list, sort by last-recorded, skim every transcript. Promote the keepers (rename, tag, expand the prose body) and delete the rest. Twenty minutes a week is enough for a heavy capture habit; ten for a normal one.

An audio archive without transcripts is a write-only store. You can capture into it forever; you can't query it. The transcript is what makes the archive readable, searchable, and skim-able in pass two. A voice-notes app without transcription is a tape recorder with extra steps; a transcribing voice-notes app is the first real time-machine for ideas.