How I built this website
This website started from a simple need: a site of my own, clear and almost self-maintaining. These are the decisions I made.
The stack
Quarto builds static pages from text. A Pandoc filter, written in Lua, adds the header, the menu, the footer and the date to every page, so pages hold only their content. There is no database, no cookies and no third-party scripts.
The brand is in the code
Colours, fonts and sizes come from a written specification and a single stylesheet. More than six hundred automated checks compute the contrasts (WCAG AA), the order of the menu, or that every page has a single title. If something breaks, it is not merged.
Two languages and two themes
Every page exists in Spanish and in English. The light or dark theme follows the system, and a button with a sun and a moon lets me change it.
Publications
I took the structure of Andrew Heiss’s research page as a model: a side index, categories that filter the list, and an action strip under each entry. References come from a BibTeX file exported from Zotero. Each publication has its own page, its BibTeX, a RIS file and COinS metadata so that Zotero can import it.
How it is published
I build on my computer and push the result to a public GitHub repository. A webhook makes the UPV server download it within seconds. I do not use remote actions: everything is checked first, locally.
What is missing
The real links of the publications, the teaching materials and the data of my studies are still missing. This is the first note; I will add more.
Last updated: