Quarkdown: A modern Markdown-based typesetting system

Original Article ↗ Hacker News Discussion ↗

Positioning vs existing tools

  • Many compare Quarkdown directly to Typst, LaTeX, Quarto, Pandoc, MyST, reStructuredText, etc.
  • Some see it as “Typst but more approachable” or “LaTeX with Markdown syntax,” others as redundant given Pandoc + LaTeX/HTML pipelines.
  • Several note major omissions or inaccuracies in the project’s comparison table (e.g., Typst support, LaTeX scripting, LaTeX→HTML via existing tools).
  • Quarto and R Markdown are highlighted as mature “Markdown in, many formats out” systems with strong editor integration.

Output formats and pipeline

  • Quarkdown is praised for targeting both HTML and PDF, but several point out PDF is just Chrome’s print-to-PDF over HTML, similar to existing headless-Chrome or WeasyPrint setups.
  • Some users ask for EPUB and LaTeX output; others want a compiled demo PDF and side‑by‑side LaTeX comparison.
  • For many, “Markdown → HTML/CSS → PDF” is already solved with existing tools.

Syntax, power, and Markdown compatibility

  • The function syntax (.function {arg} with indented bodies) is seen as powerful but contentious:
    • Some like the Smalltalk/DSL feel; others say it stops looking like Markdown and resembles reStructuredText or MyST.
    • Concerns about keyword/function naming collisions and the difficulty of evolving the language.
  • Debate over whether “slightly more concise than LaTeX” is enough value; some prefer full LaTeX/Typst power or plain Markdown minimalism.

Use cases and layout control

  • Supporters hope for a modern Markdown-based replacement for LaTeX in academic and scientific publishing.
  • Skeptics question table sophistication (merged cells, grids), page-numbering schemes, and fine-grained typography (drop caps, kerning, wrap-around images).
  • Several note that tools like Typst and LaTeX are still better for complex layouts, posters, and non-paper designs, though those are hard without WYSIWYG.

Tooling, runtime, and adoption

  • The Java 17/Kotlin/JVM dependency is a major turnoff for some; others argue Kotlin is fine and could go native later.
  • Multiple comments doubt academic adoption without publisher templates and without co‑authors switching away from LaTeX.
  • A few speculate that if LLMs start emitting Quarkdown by default, that alone could drive adoption.

Broader reflections

  • Thread repeatedly revisits whether extending Markdown is wise versus using HTML+CSS, LaTeX, Typst, Org-mode, or XML schemas like DocBook/DITA.
  • Some want a “universal Markdown-ish front end” compiled into various robust backends; others feel the proliferation of Markdown derivatives is itself a problem.