---
name: Publishing Industry
type: system
status: legacy
version: 3.1.4
released: "1440-01-01"
maintainer: a rotating cast of conglomerates and their anxious subsidiaries
dependencies:
  - authors
  - paper
  - cultural anxiety
  - gatekeeping
  - [Amazon](/amazon)
  - [capitalism](/capitalism)
license: Proprietary (rights retained, options held, sequels negotiated)
tags:
  - media
  - culture
  - slow
  - hierarchical
  - print
  - legacy-infrastructure
---

Publishing Industry

What it actually is

A centuries-old system for deciding which ideas become books, then deciding who gets to read them, then being surprised when the numbers are bad.

How it works

An author generates a manuscript. This takes anywhere from eight months to fourteen years.
The manuscript is submitted to an agent, who reads it in three to eighteen months, then submits it to an editor.
The editor loves it. The editor's boss is uncertain. A committee meets. The committee is cautiously optimistic. An offer is made.
The book is published eighteen to twenty-four months later into a market that has moved on.
The book sells moderately. Everyone agrees it was ahead of its time, or behind it.

"We really believed in this one." - every acquiring editor, twice a year, for 580 years

Features

The Advance: money given upfront that statistically never earns out, functioning as a gift with complicated feelings attached
The Slush Pile: an infinite queue. No SLA. No acknowledgment. Assume lost.
The Midlist: a tier of reliable, quality authors who get progressively worse deals until they quietly leave
Backlist: the part of the catalog that actually makes money; quietly subsidizes every bad bet on the frontlist
Cover design by committee: produces results that satisfy no one and somehow still sell

Known bugs

Royalty statements are issued biannually and are legible only to people who enjoy confusion professionally
Discoverability is unresolved and has been marked WONTFIX since approximately 1995
Debut authors receive massive marketing budgets OR nothing. No middle state exists.
The system occasionally surfaces a book about the publishing industry that the publishing industry then fails to distribute
Diversity initiatives run as a plugin, not core infrastructure. Frequently crashes under load.

Error codes

ERR_RIGHTS_REVERSION     : Author requested. Publisher lost the paperwork.
ERR_OUT_OF_PRINT         : Book exists. Copies do not.
ERR_AGENT_UNRESPONSIVE   : Not a bug. Expected behavior.
ERR_ALGORITHM_UNKNOWN    : Bestseller list methodology classified.
ERR_FORMAT_WAR           : ebook vs. print vs. audio. All three are winning. None are.
WARN_CONSOLIDATION       : Five major houses. Four. Three. Watch this space.

Dependencies

This system cannot run without bookstores, even as it slowly defunds them. It also requires libraries, which it views with complicated affection, like a parent who never calls but cries at the graduation.

Heavy runtime dependency on nostalgia. Removing it causes immediate instability.

Changelog

v1.0 (1440): the printing press ships. Everything changes immediately, then slowly.
v2.0 (1800s): Mass literacy. Serialization. Charles Dickens emerges as a stress test.
v2.7 (1990s): Consolidation begins. Borders opens. Borders closes.
v3.0 (2007): The Kindle. A moment of genuine existential panic.
v3.1 (2020s): self-publishing normalized. AI enters slush pile from both directions.
v3.1.4 (current): Functional. Structurally similar to a building being renovated while occupied.

FAQ

Is the industry dying? It has been dying since 1455. Current vitals: stable, irritable, restructuring.

Do publishers read unsolicited manuscripts? Technically yes. Practically: query an agent. Query another agent. Consider therapy first.

Who actually makes money? literary agents (sometimes), Amazon (always), authors whose first book was a sensation (briefly).

drafting spec…