Universe
Create project in app

This repo contains a minimalistic template for writing a thesis. The main objective of the template is to produce a nice document with as little code as possible. No nonsense, no extras.

While this is not an official template for the “Universität zu Lübeck”, it is loosely inspired by its official LaTeX template by Till Tantau. The official LaTeX template is a lot prettier. However, it is also more code.

To be clear: Using this template does not guarantee any compliance with university standards. Consult your advisor for tips and rules on writing a thesis.

Installation

Typst web application

Just select this template by its name and version and start writing!

Typst CLI (local)

In your CLI, navigate to your empty project directory and use typst init @preview/minimal-thesis-luebeck:0.1.0. I recommend using VS Code and the tinymist extension for working with a local installation.

Usage

When installed, this template provides you with a thesis.typ file. This is where you write your thesis.

Note that by default this template is configured to write thesis using english language in a german environment. However, you can exchange all strings in the template to match your preferred languages.

Hint: You can take a look at tutorial.typ if you are new to Typst.

Template Contents

  • thesis.typ: The central document that finally produces your PDF document. Initially, this includes tutorial.typ to give some examples. Just remove the tutorial when you start working.
  • thesis.bib: The bibliography. Put your references here.
  • texts: Contains texts that are separate from your main thesis such as the abstract. If you like, you can also put chapters of your thesis here. Personally however, I like having everything in one place.
  • images: Contains all images. You should replace the example images with your own. Also, you might need to adjust some sizes afterwards (config/titlepage.typ).

Credits

This template is inspired by this one: thesis-template-typst

The example image of a cat was generated using OpenAI’s DALL-E.