Skip to content

SYSTEM Cited by 1 source

Netflix 10-18 Monkey

10-18 Monkey (short for Localization-Internationalization, i.e. l10n + i18n) is the Simian Army member that detects configuration and runtime problems in instances serving customers across multiple geographic regions, different languages, and different character sets. Introduced in Netflix's 2011 TechBlog post (Source: sources/2026-01-02-netflix-the-netflix-simian-army).

Purpose

"10-18 Monkey (short for Localization-Internationalization, or l10n-i18n) detects configuration and run time problems in instances serving customers in multiple geographic regions, using different languages and character sets."

Role in the Simian Army

10-18 Monkey is a drift detector for the globalization surface of Netflix's services. The failure mode is the class of bug that only appears under non-English character sets, non-ASCII input, non-default locale formatting, or non-US region configuration — the kind of bug invisible to an English-language CI run but severe for customers in other regions.

Scope (2011)

The 2011 post is notably thin on specifics for this monkey. Named problem classes: "configuration and run time problems" involving:

  • Multiple geographic regions (different Netflix service deployments / CDN edges).
  • Different languages (UI, captions, metadata).
  • Different character sets (non-Latin scripts, RTL, emoji, etc.).

No detail on the detection rules or the enforcement action is given in the 2011 post. Unlike Conformity / Security / Janitor Monkey, which have clear enforcement actions (terminate the offender), the post does not describe how 10-18 Monkey acts on its findings.

Lineage

As Netflix grew internationally — especially with the 2010 Canada launch and the 2011 Latin America / Caribbean expansion that was contemporary with this post — globalization correctness became a first-class concern. The naming of 10-18 Monkey as a Simian Army member reflects that this concern was peer to infrastructure reliability in Netflix's early cloud architecture.

Implementation gaps in the 2011 post

The most under-specified monkey in the post:

  • Detection rules undocumented.
  • Signal sources undocumented.
  • Enforcement action undocumented (kill? alert? queue for review?).
  • Regional partitioning of the monkey itself not described.
  • Integration with A/B testing across regions not mentioned.

Operational numbers

None disclosed.

Seen in

Last updated · 319 distilled / 1,201 read