nexus is a threefold documentation resource for open research: native, external via URL pointers, and hybrid. Hybrid documentation is commentary generated when working through an external resource such as a tutorial. This is intended to fill context gaps, note gotchas and so on: Make the path easier to follow.

You are reading meta-narrative for the nexus repo.
nexus content resides in the gh-pages branch
and this is published at https://robfatland.github.io/nexus.

nexus

This repo started life as 'how to disable garish fonts in a typical bash interface'. I just want green characters on a black background and one day I got sick of looking up the procedure yet again. To generalize the process, nexus is now here to remind me how open science procedure works and, related, how to build research computing cyberinfrastructure.

Example motivating tasks

Each of the following examples links to some corresponding work.

• Task Zero: How do I manage Python environments?
• Task I wrote some code that is very specific to some oceanography data. How do I publish it as a package?
• Task 2: I derived some results as useful data. How do I publish it as an API-accessible resource?
• Task 3: I am working with some sensitive data and I want to be really careful about restricting access.
• Task 4: I like having my code backed up in a GitHub repo. What are the ten git commands I need to master?

Navigating this GitHub repository

This repo has a main branch which primarily consists of this README file. Then there is the publication component in a gh-pages branch. Every time there is a commit to this branch: A (much prettier) documentation website is refreshed.

• nexus content gh-pages branch
• nexus published website.
• nexus 'main' branch.

nexus aspirations (the to-do list)

• TCOTU: Understanding, building, maintaining, using Python environments
  • corollary: customizing bash, vi to a simple appearance
  • ssh tunnels from one machine to another
  • headless jupyter running on a cloud VM with a localhost laptop interface
• Comparison: miniconda versus anaconda
• PC-based Linux / IDE: Ubuntu bash, WSL-2, VSCode
  • Containerization
  • conda environments
• cloud infrastructure building - Data systems, APIs - Building a Flask server - Cloud security - ML/AI: CNN, NLP - varieties of cloud CLI - cloud spend notification - stopping runaway cloud spend automatically
• Python's guide to publishing a software package
• My repository-based projects
  • epipelargosy, an oceanography project
  • steam: university-level STE(A)M notes
  • sodium: a study of distributed forms of information
  • analytic number theory
  • middle school level STE(A)M outreach