New book available: This Is Server Country
on this page

Parallel Iliad: Brainrot Edition

software

A parallel Greek-English reader of Homer's Iliad (Palles 1904/1917) with AI-generated 'brainrot' translations, margin notes, and a searchable character bible

period: 2026-present
tech:
AI TranslationClassical StudiesGPT-5.4Alpine.jsTypography
Website

A parallel-text reader presenting Alexandros Palles’ Modern Greek Iliad (1904/1917) alongside a playful “brainrot” English translation — internet-meme-flavored language aimed at making Homer accessible to high schoolers. Think “what if Homer was a TikTok storyteller.”

Read It

Launch the Parallel Iliad Reader

How It Works

The translation pipeline is fully AI-driven, using GPT-5.4 with structured output enforcement:

Multi-Phase Pipeline

  1. Source Parsing — Palles’ text (Project Gutenberg #36248) parsed into stanza-structured JSON preserving the original paragraph breaks
  2. Character Bible — 184 characters with consistent “brainrot” epithets, 10 factions, 38 places, and 31 translation conventions generated from key books
  3. Book Summaries — Per-book narrative summaries with emotional arcs, key scenes, and brainrot titles
  4. Stanza Translation — Each stanza translated with full context: character bible, book summary, and previous stanza for continuity
  5. Margin Notes — AI-generated annotations for cultural references, character introductions, wordplay, and things a high schooler might not get

The Reader

The single-page HTML reader features:

  • Parallel text — Greek and English side by side, line by line
  • Typography controls — adjustable font size, line spacing, and justification
  • Reading modes — light, dark, and sepia themes
  • Margin notes — toggleable annotations with categorized symbols
  • Character Bible — searchable sidebar with all 184 characters and their brainrot epithets
  • Responsive design — works on desktop and mobile

Source Text

The Greek source is Alexandros Palles’ demotic Modern Greek verse translation of the Iliad, published in 1904/1917. This edition covers 23 of 24 books (Book 13/N was not translated by Palles). The source text is from Project Gutenberg eBook #36248.

Technical Details

  • Model: OpenAI GPT-5.4 with pydantic-ai structured output
  • Source: 23 books, 1,077 stanzas, 12,702 lines
  • Output: Single self-contained HTML file with embedded JSON data, Alpine.js for reactivity, EB Garamond + Noto Serif typography
  • LaTeX: XeLaTeX PDF also available with two-column longtable layout (verse pairs + margin notes)
on this page