Potential reasons for blogging include:
- to raise your profile and get a better job
- to share your opinions
- to “learn in the open”
- to share useful materials on the Web
- to document stuff you’ve learned which just might be useful for someone else one day
For me the main reasons are numbers 3 to 5, and rarely 2. Reason number 5 is probably the most important for me, because I’ve done a lot of very obscure stuff in global health informatics that I need to document somewhere. I’ve also benefitted from many other people who have put useful information out there just in case it was useful, and I’m very grateful to them. I don’t blog for reason 1, as I’ve painted myself into a niche techy corner in this late stage of my career and can’t imagine ever applying for another job.
At the moment this blog is mostly random stuff that I’ve managed to put together in limited time, but one day I’d like it to be a useful resource for epidemiologists, particularly epidemiologists involved in global health informatics. So I’ve put together a sort of roadmap for what this blog will ultimately include. The content will mostly be about technical subject areas that are adjacent to public health work in some way, and in some cases will be related IT hobbyist stuff which gave me the initial foothold in informatics. But overall you could call it technical epidemiology.
Here are the broad areas I have some interest in and experience with and hope to cover eventually, learning more about each area in the process:
- Knowledge management and learning, e.g. note taking, Anki, blogging, participation in e.g. Advent of Code
- Book/article summaries, teaching materials, best practices
- R stuff, particularly intermediate/advanced uses, packages, testing, API interaction, JSON, XML/XPath, reproducibility, Shiny, WebR, useful snippets/scripts/packages I have developed
- Python stuff, particularly for data engineering, app development, AI/machine learning, useful scripts/snippets I have developed, Jupyter
- Data management engineering/integration stuff, particularly technologies suitable for low-resource settings, including relevant R (e.g. data.table), Python (e.g. polars, dbt, dlt), SQL, Rust, databases esp PostgreSQL, DuckDB/DuckLake stuff, data modelling, data architecture, Apache Airflow, dashboards
- Git stuff including workflows, advanced features, GitLab, GitHub, alternatives like forgejo
- Statistics stuff including commonly used epidemiological methods, Bayesian methods, causal inference, data visualisation, mapping, diagramming
- Software stuff, including IDEs (Vim, Positron, VSCode, RStudio), Quarto, Typst, OpenRefine, D3
- Home server and cloud stuff: maintenance, Proxmox, VMs, Docker, Kubernetes, Ansible, command line, tmux, SSH, nginx, Nextcloud, Android, server hardening, Cloudflare, self-hosted software, digital sovereignty from Google or Apple
- Global health informatics: context and constraints, needs assessment, technology appraisal, implementation, troubleshooting, DHIS 2, SENAITE LIMS, SORMAS, WHONet, information security, interoperability and integration
- Software development, including hosting on the Internet, web development, JavaScript, Plone, Flask, Django
- Cybersecurity, including networking
- AI, including experience of AI-assisted coding or engineering, GitHub CoPilot, Aider
