alunduil / alunduil/blog.alunduil.com

A backlog with a high enough turnover rate isn't a backlog

Abierto
#533 0 comentarios 0 reacciones 0 asignados Ver en GitHub
idea
Lenguaje dominante
Astro
Estrellas
1
Forks
1
Merge medio
6 h 34 min
PR fusionados (30 d)
118

Descripción

## Spark

Digest, week of 2026-08-08: 120 issues opened, 99 closed, 82 pull requests, ten
repositories, seven days. The same week's reading ran squarely into backlog and
management skepticism — rands's "R.I.P. Your Backlog", "Middle management roles
are also a trap", "The Big Management Lie: Overpromising".

## Why it could be interesting

The activity is "I filed a lot of tickets." The claim to test is that the
standard critique of backlogs — they accumulate, they rot, they become a
graveyard that guilts you — assumes a *low* closure rate, and stops applying
when open and closed run at roughly the same rate. At 120 in and 99 out, the
queue isn't storage; it's a working surface, closer to a task list than an
inventory, and the items that rot are visible precisely because the rest move.

The honest counter-position, which the post has to take seriously: a 1.2:1
open-to-close ratio still grows without bound, and "my backlog is fine because
it churns" is exactly what someone with a rotting backlog would say. The
interesting version of this post measures rather than asserts — age distribution
of open issues, what fraction of a week's filings are closed within that week,
and how much of the churn is agent-generated rather than chosen.

## Open questions

- [ ] Get the actual numbers before drafting: age distribution, same-week close
rate, net growth over several months. Without them this is a vibe.
- [ ] Is the churn a property of how I work, or of the window? One week during
a fleet-wide sweep is not evidence of a steady state.
- [ ] How much of the 120 is decomposition of work already decided versus new
scope? Those are very different claims about the queue.
- [ ] Does rands's argument actually conflict with this, or is he critiquing a
different artifact (product backlog, prioritised and negotiated) than
mine (a personal issue graph)?
- [ ] Note or post? #461 argued a similar "if the number has no story it's a
note" test — this may be the same shape.

## Source material

- Window totals, digest 2026-08-08 → 2026-08-15: 120 issues opened, 99 closed, 82 pull requests, 10 repositories
- R.I.P. Your Backlog (rands) — https://read.readwise.io/read/01kzh2wqqpypydv7j4gy6d95bq
- Middle management roles are also a trap. (Irrational Exuberance) — https://read.readwise.io/read/01kzhkd6991y5v9srhxa2434my
- The Big Management Lie: Overpromising (Stay SaaSy) — https://read.readwise.io/read/01kz9c3zkpkzbg3jr3bmhp8rfv
- Could vs. Should: The First Year Managing an SRE Team (Reid Savage) — https://read.readwise.io/read/01ky3r9xjxag32yegwkbg851eq
- Fifteen Tics (rands) — https://read.readwise.io/read/01kzp8kny645ppdh9e51tq0dtd
- Adjacent existing ideas: #461, #229, #283

Guía de contribución

No hay ninguna guía de contribución indexada para este repositorio

Evaluación

Este issue todavía no se ha evaluado.

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.