STANDARD OF TASTE

HUME'S CRITERIA · PRACTICE

Practice

Practice is the criterion that makes this product a gym rather than a mirror. Hume is unambiguous: nothing improves the faculty of judging more than practice in a particular art — the repeated, attentive survey of works of one kind. Taste, in his account, is not an endowment you check once and frame. It's a capacity that sharpens with reps and dulls with neglect.

He even describes the beginner's condition: confront a work for the first time and the sentiment it produces is obscure and confused — you can tell you feel something, but not which parts of the work are doing it, or how well. Only repeated encounters let a judge resolve that blur into discrimination: this voicing, that transition, this specific flaw. Anyone who has learned to hear the difference between a good and a great recording of the same piece has lived this.

The gym takes the claim literally, with the same honesty rule as everything else: an improvement you can't measure is an improvement you can't claim. Sit a threshold ladder twice in the same browser and the result screen compares the two — against a noise floor we measured first, so that a difference smaller than the instrument's own run-to-run wobble is reported as no change rather than as progress.

That floor is high, and saying so is the point. Two sittings on the pitch ladder have to differ by roughly 3.5 times before the arc will call it movement; on the prestige test the label's pull has to shift by eight points of the scale. Most retests are therefore told that nothing changed the instrument could hear — which is the honest answer, and the reason the sentence names what it would have taken instead of leaving you to guess. The delicacy trials get no arc at all: fifteen pairs cannot resolve a change smaller than six of them, so that screen says so and points here.

What a second sitting genuinely buys is precision. The wobble of an average falls as the square root of the number of sittings, so the more often you come back, the smaller a real change has to be before this can see it. That is the whole return: not a badge or a streak, but a number that gets harder to argue with.

Practice alone isn't sufficient, though. Hume pairs it with breadth — you can rehearse one narrow corner of music forever and stay a provincial judge. That failure mode belongs to comparison, and knowing whether to trust your own sharpening judgment belongs to good sense. The gym starts where prejudice is caught in the act: the Prestige Test.

Questions, answered straight

Did Hume think taste could be trained?
Yes — explicitly. He wrote that nothing improves the talent of judging more than practice in a particular art, and that a first attempt at judging is always 'obscure and confused.' Delicacy sharpens with use; that claim is why a gym for taste is coherent at all.
How does the Taste Gym use practice?
Sit a threshold ladder twice in the same browser and the result screen compares the two sittings. It is free, because charging for the training loop would put the one honest question — did your ear actually move — behind a wall. The comparison is judged against a noise floor measured first, so a difference smaller than the instrument's own run-to-run wobble is reported as no change rather than as progress: on the pitch ladder two sittings must differ by roughly 3.5 times before it will call it movement. Most retests are therefore told that nothing changed the instrument could hear, which is the honest answer.

The library · Take the Prestige Test · Terms · Privacy