Four things that change the answer when nothing changed

Slide 1 / 27

What this lesson covers

• One: the clock
• Two: the seed
• Three: the order files come back in
• Four: the locale
• The test that catches all four
• The deterministic-output habit
• Report it whole
• What comes next

Slide 2 / 27

One: the clock — In Python

from datetime import date
report_date = date.today()          # different every day, by design

Slide 3 / 27

One: the clock

• Anything that reads the current time makes every run differ — A generated-on footer, a filename with today's date, a…
• The fix is to take the date from the data or from a parameter

Slide 4 / 27

One: the clock — In Python

import pandas as pd

survey = pd.read_csv("data/raw/household-survey-2025.v1.csv")
as_of = pd.to_datetime(survey["interview_date"]).max()   # from the data

Slide 5 / 27

One: the clock — In R

as_of <- max(as.Date(survey$interview_date))

Slide 6 / 27

One: the clock

• This platform got this wrong and the failure is instructive — The course handouts and lesson decks originally took…
• The date is now scoped to the artefact — A handout's date comes from its own course and lessons, a deck's from its own…

Slide 7 / 27

Two: the seed — In Python

import numpy as np

rng = np.random.default_rng(20260729)      # stated, committed, reproducible
sample = rng.choice(households, size=200, replace=False)

Slide 8 / 27

Two: the seed — In R

set.seed(20260729)
sample(households, 200)

Slide 9 / 27

Two: the seed

• Anything random needs a seed and the seed belongs in the code — Bootstrap intervals, random sampling for verification,…
• Set it once, at the top, visibly — A seed buried three functions deep is a seed someone will move
• And say so in the output — The regression course's bootstrap interval is reported as "400 resamples, seed 20260729" for…
• Every dataset on this platform is generated by a seeded script — which is what makes pnpm datasets:generate a…

Slide 10 / 27

Three: the order files come back in — In Python

import glob
for path in glob.glob("data/raw/*.csv"):     # order is filesystem-dependent
    ...

Slide 11 / 27

Three: the order files come back in

• glob and os.listdir return files in an order that differs between operating systems and sometimes between runs — If…

Slide 12 / 27

Three: the order files come back in — In Python

for path in sorted(glob.glob("data/raw/*.csv")):
    ...

Slide 13 / 27

Three: the order files come back in — In R

for (path in sort(list.files("data/raw", full.names = TRUE))) { }

Slide 14 / 27

Three: the order files come back in

• Sort it. Always. It costs six characters
• The same applies to dictionary and group order — groupby in pandas sorts by default and dplyr::group_by does not; a…

Slide 15 / 27

Four: the locale — In Python

float("1,234")           # ValueError, or 1.234, depending on where you are

Slide 16 / 27

Four: the locale

• The decimal separator, the thousands separator, the date format and the sort order of accented characters are all…

Slide 17 / 27

Four: the locale — In Python

survey = pd.read_csv(path, decimal=".", thousands=None)     # stated, not inferred
dates = pd.to_datetime(survey["date"], format="%Y-%m-%d")   # explicit

Slide 18 / 27

Four: the locale — In R

readr::read_csv(path, locale = locale(decimal_mark = ".", date_format = "%Y-%m-%d"))

Slide 19 / 27

Four: the locale

• State the format rather than letting the reader infer it — pd.to_datetime without a format is a function that…
• Sorting is the subtler one — sorted() on French commune names orders Étroit after Zone under one locale and…

Slide 20 / 27

The test that catches all four — Shell

python run.py && cp -r outputs outputs-first
python run.py && diff -r outputs outputs-first

Slide 21 / 27

The test that catches all four

• Run it twice and diff — Anything that differs is one of the four, and the diff names the file
• Run it twice on different machines — and you additionally catch the locale and the file ordering, which a single…
• Put it in CI — which is a clean machine every time, and the check runs whether or not anyone remembers

Slide 22 / 27

The deterministic-output habit

Slide 23 / 27

The deterministic-output habit

• This platform pins SOURCE_DATE_EPOCH for both pdfTeX and pandoc — because a .pptx is a zip whose entry times and…
• The payoff is that a diff means something — When rebuilding produces no change, a change in the diff is a real change —…

Slide 24 / 27

Report it whole — Example

Determinism

  All randomness is seeded: seed 20260729, set in src/config.py and reported
  with every bootstrap interval in this report.

  Dates are taken from the data (the latest interview date) rather than from
  the clock. No output contains a generation timestamp.

  File iteration is sorted. Group order is stated explicitly rather than
  inherited from the grouping library's default.

  CSV reading states the decimal mark and the date format rather than
  inferring them.

  Verified: running the pipeline twice produces byte-identical outputs, and
  CI runs it on a clean machine on every push.

Slide 25 / 27

Report it whole

• The last line is the claim and the four above it are how it was achieved — A report that asserts determinism without…

Slide 26 / 27

What comes next

• A reproducible pipeline that produces one report is worth having.

Slide 27 / 27

Where this goes next

Read the full lesson, with runnable code Back to the lesson