Skip to content
TopicTracker
From HackerNewsView original
TranslationTranslation

Context on Context

Providing LLMs with better context—such as relevant documents or examples—can improve performance more effectively than fine-tuning, making context engineering a powerful tool for steering model behavior.

Background

- David Manheim is a decision scientist and risk analyst, and this essay argues that providing "context" in journalism and intellectual writing often backfires when done badly. - The piece distinguishes between *explanatory context* (which helps readers understand unfamiliar technical or cultural terrain) and *performative context* (name-dropping, status cues, or group-identity signaling meant to establish credibility rather than educate). - Manheim warns that bad context burdens the reader with irrelevant detail, panders to insiders, or substitutes for clear writing. Good context, he argues, is concise, neutral, and designed specifically for the reader who needs it—not for the writer's own reputation. - The essay is relevant to anyone who writes for a cross-disciplinary or international audience, especially in tech, policy, or academic commentary where jargon and unspoken assumptions are common.

Related stories

  • The article discusses Opus 3: Henry VI, Part 2, continuing the exploration of early digital adaptations of Shakespeare's works on The Analog Antiquarian.