After having made a lot of personal projects, I realized that I had forgotten a ton of them, and there wasn’t really a place to show them off. This website was made to solve that issue.
- Tagging system
- Relatedness measures between tagged items
- List of projects
- Comments on blog and project pages
- Pie chart of open source licenses
- A resume
- Webmentions, sending and receiving
- Feed aggregation of activity from all my accounts, plus my self-hosted “microblogging” system
Tech stack summary
- Frontend - statically generated, optimized SPA
- Backend - API-only backend (except for the admin panel)
I use Gatsby to statically generate the website. However, I am considering moving off of it to Next.js.
There is a Github Actions workflow that automatically builds on every push to verify that my code compiles. Additionally, if there was a push to the
main branch, it will publish the build output to the astralbijection.github.io repo.
UX design methodology
I’ve tried to make the browsing experience as user-friendly and inclusive as possible by:
- Adopting a mobile-first methodology for designing views, and responsively sizing elements in CSS
- Statically generating the site to reduce bandwidth consumption for users
- Utilizing semantic tags and designing the site to be accessible for screen readers
Currently, the backend’s only functionality is to serve and receive possibly anonymous comments.
I have pointed Docker Hub at my repo to build, test, and release a new
:latest image on every push to main.
It is deployed as a Docker container on a Contabo VPS located in Missouri for about $8/mo. See the
docker-compose.yml for deployment details. I will soon set up Watchtower to automatically pull the image, so that the only thing I need to do with deployment is push to
The content is in its own Git submodule as a set of raw files that get aggregated and compiled on each frontend build. It is written in the following file formats:
- Jupyter notebooks for a few blog posts (I wrote a custom Gatsby plugin that auto-transforms the Jupyter notebooks into Markdown files)