seed concept

Definition	The number that fixes a run's pseudo-randomness. Two runs with the same seed agree only when code, data, environment, and execution are also fixed, and different seeds are necessary but not sufficient for an independent replication. Chapter 8.
Example	Seed 20260827 fixed the run, and seeds 20260828 and 20260829 were the two replications.
Book	Data Mining as Observation, draft 0.2, commit `f3914f0`; entry id `seed`, kind concept.
Status	no ledger row names this entry. Corrections: none recorded.
Defining equation	Book equation 8.1.
Assumptions and scope	The number that fixes a run’s pseudo-randomness. Two runs with the same seed reproduce each other only when the code, data, environment, and execution are also fixed, since nondeterministic execution can differ at the same seed, and runs with different seeds are necessary but not sufficient for an independent replication. The standard error over n seeds falls as one over the square root of n, a paired comparison across seeds has smaller variance than an unpaired one exactly when the two arms covary, and a preregistration names its seeds before the run.
Prior art	none recorded
Evidence	`geometric-observation/claims/LEDGER.md`, `observation-theory-campaigns/experiments/DATABASE-FRESHNESS-TRACK.md:1-45`, `lean/DataMiningAsObservation/StandardError.lean`, `lean/DataMiningAsObservation/Bootstrap.lean`
Reviewed	semantic review 2026-09-06; generated 2026-09-10 from records at the commits on the provenance page.

The number that fixes a run's pseudo-randomness.

Equation

Book equation 8.1.

\[O_{\mathrm{harness}}=\big(C_{\mathrm{score}},\ G_{\mathrm{metric}},\ B=\text{folds}\times\text{samples}\times\text{seeds}\big).\]

Conditions

The number that fixes a run’s pseudo-randomness. Two runs with the same seed reproduce each other only when the code, data, environment, and execution are also fixed, since nondeterministic execution can differ at the same seed, and runs with different seeds are necessary but not sufficient for an independent replication.
The standard error over n seeds falls as one over the square root of n, a paired comparison across seeds has smaller variance than an unpaired one exactly when the two arms covary, and a preregistration names its seeds before the run.

Conditions are curated in entries.toml rather than read from a record.

Ledger

none

First stated

Chapter 8 section 8.1 of Data Mining as Observation, with the disjoint-seed rule of the database freshness track in observation-theory-campaigns/experiments/DATABASE-FRESHNESS-TRACK.md.

Measurements

Where the book states it	Numbers, as the book’s sources table records them	Source
chapter 4 section 4.4	GO-4 budget inversion, fixed m 10 rises, matched m 121, 126, 159 collapses, 3 seeds	`geometric-observation/claims/LEDGER.md` row GO-4
chapter 13 section 13.3	ZooKeeper hot 0.99 cold 0.01 witnessed 0.0, Postgres 0.50 to 0.06, MongoDB 0.47 to 0.03, production Postgres 0.47 to 0.02, real substrates, disjoint seeds	`observation-theory-campaigns/experiments/DATABASE-FRESHNESS-TRACK.md:1-45`

Failures and corrections

none

Invariance envelope

none declared

Machine checked

lean/DataMiningAsObservation/StandardError.lean, theorems se_pos, se_quarter, se_antitone, se_tendsto_zero, thousand_folds, at observation-data-mining f3914f0; what the check covers is stated in the book’s appendix C.

lean/DataMiningAsObservation/Bootstrap.lean, theorems mean_sub, var_sub, paired_lt_iff, cov_comm, at observation-data-mining f3914f0; what the check covers is stated in the book’s appendix C.

Used in

Data Mining as Observation primer S, chapters 4, 6, 7, 8, 9, 11, 12, 13.

harness; standard error; bootstrap; preregistration; sealed.

Status

Generated 2026-09-10 by encyclopedia/generate.py; book at observation-data-mining f3914f0; the commit of every record is listed in the encyclopedia’s provenance.

← sealedsensitivity →