Pages are written as markdown files, then converted to HTML
using md2html.sh
, which converts them to pandoc and
fixes some small bits (with sed
and
regex-chicanery.py
). Then a new RSS feed and
sitemap are generated using generate-rss.sh
and
generate-sitemap.xml
.
/blog/
/til/
/data/
/assets/
,
lining up with the organization of the page they go to (e.g.
/blog/marlin-boot-animations.{md,html}
->
/assets/blog/marlin-boot-animations/
)Error pages are provided by http-status-codes.