UNM-CARC/dust-2026 · docs
Three 50-minute lessons on open science, research data management, and the ethics of artificial intelligence for NIEHS Superfund Research Pr
Open source Repository Open in the app JSON README (API)
About
# DUST 2026: Open Science Training { .carc-visually-hidden }
<div class="carc-hero" markdown>
<p class="carc-hero__title">DUST 2026: Open Science Training</p>
<p class="carc-hero__tagline">Open science, research data management, and the ethics of artificial intelligence for Superfund Research Program trainees studying hazardous dust, mine waste, and metal exposure in the Southwest.</p>
<div class="carc-actions" markdown>
[:material-rocket-launch: Start Lesson 1](lessons/01-open-science.md){ .md-button .md-button--primary }
[:material-format-list-numbered: All lessons](lessons/index.md){ .md-button }
[:material-bookshelf: Resources](about/resources.md){ .md-button }
[:material-github: Source on GitHub](https://github.com/UNM-CARC/dust-2026){ .md-button target=_blank }
</div>
</div>
!!! info "For Superfund Research Program trainees in Arizona, New Mexico, and Texas"
This training is written for graduate students and early-career researchers in three NIEHS Superfund Research Program
Details
- Kind
- OKF bundles
- Topic
- Files & documents
- Publisher
- unm-carc
- Origin
- okf_github
- Category
- dados
- Version
- 0.2
- Last push
- 2026-09-12T19:32:50Z
- Repository state
- ativo
- Language
- Python
- License
- GPL-3.0
- Added
- 2026-09-13 21:04:14
- Updated
- 2026-09-13 21:04:14
- Origin id
UNM-CARC/dust-2026:docs/index.md
README
# DUST 2026: Open Science Training
[](https://creativecommons.org/licenses/by/4.0/)
[](https://zensical.org)
Three 50-minute lessons, each with a self-paced homework page, for NIEHS
Superfund Research Program trainees at the
[University of Arizona DUST Center](https://superfund.arizona.edu/), the
[UNM METALS Center](https://hsc.unm.edu/pharmacy/research/areas/metals/), and the
[Texas A&M Superfund Research Center](https://superfund.tamu.edu/):
1. **Foundations of Open Science** — principles, the six pillars, Gold
Standard Science, and the 2026 US public-access and publication-cost
landscape
2. **Modern Data Management** — the data life cycle, FAIR and CARE, the 2026
NIH and NSF plan formats, data rescue
3. **Ethics and Artificial Intelligence** — bias, responsible and agentic AI
use, NIH and journal AI rules, energy and water costs, regulation
Each lesson is a 50-minute lecture page plus a twelve-module self-paced
homework page with checkpoints, written so an AI tutor can deliver it.
**Website:** <https://unm-carc.github.io/dust-2026/>
This is the 2026 edition of [DUST 2025](https://tyson-swetnam.github.io/dust-2025/).
It is built with [Zensical](https://zensical.org), styled after the
[UNM CARC documentation](https://carc.unm.edu/docs/), and structured as an
[Open Knowledge Format (OKF) v0.2](https://github.com/GoogleCloudPlatform/knowledge-catalog/blob/main/okf/SPEC.md)
knowledge bundle so the content is first-class for both humans and AI agents.
## Quick start
```bash
uv venv --python 3.12 .venv && source .venv/bin/activate # or: python3.12 -m venv .venv
uv pip install -r requirements.txt # zensical + pyyaml (Python 3.11+)
zensical serve # live preview at http://localhost:8000
zensical build --clean # static site in ./site
```
## Repository layout
```
├── zensical.toml # Site configuration (theme, nav, extensions)
├── docs/ # The OKF knowledge bundle + site content
│ ├── index.md # Landing page (declares okf_version: "0.2")
│ ├── log.md # OKF update log (reserved filename)
│ ├── lessons/ # Three lecture pages + three self-paced pages + listing
│ ├── about/ # Overview, resources, credits, AI tutor, accessibility, AI-agent guide
│ ├── assets/ # Images and logos
│ ├── stylesheets/extra.css # UNM cherry + turquoise theme, DUST additions
│ └── llms.txt, llms-full.txt # Generated agent indexes (do not edit)
├── scripts/
│ ├── okf_validate.py # OKF v0.2 conformance checker (CI)
│ ├── check_links.py # Links, images, and nav coverage (CI)
│ ├── gen_llms_txt.py # Builds docs/llms.txt + docs/llms-full.txt
│ ├── postbuild_agent_surface.py # Markdown mirror, OKF meta tags, robots.txt
│ └── externalize_links.py # Adds {target=_blank} to external links
├── MIGRATION.md # 2025 URLs -> 2026 URLs
├── AGENTS.md / CLAUDE.md # Rules for coding agents working in this repo
└── .github/workflows/docs.yml # Validation + GitHub Pages deployment
```
## Machine readability (OKF)
Every content page carries YAML frontmatter with:
- `type` — `Lesson`, `Guide`, or `Reference` (required by OKF)
- `title`, `description`, `tags` — used by search, indexes, and agents
- `generated: { by, at }` — who or what produced the current content
- `sources` — provenance links to the exact upstream DUST 2025, FOSS, or
GPT 101 file, with `last_modified` from git history
- `status` / `stale_after` — lifecycle markers; the lessons carry a
`stale_after` date because the policy and AI facts they cite move quickly
- `lesson` (Lesson pages) — objectives, key terms, duration, delivery modes,
and an accessibility profile; the build turns it into a schema.org
`LearningResource` record so AI tutors and learning platforms can pick a
delivery mode (see `docs/about/ai-tutor.md` and `docs/about/accessibility.md`)
Pages rewritten by an agent are intentionally **unverified** (no `verified`
key). When the author reviews a page, they add:
```yaml
verified: { by: "human:tswetnam", at: "2026-XX-XXT00:00:00Z" }
```
The deployed site is directly consumable by AI agents:
- `/llms.txt` ([convention](https://llmstxt.org)) — linked outline;
`/llms-full.txt` — the full corpus with frontmatter in one file.
- **Markdown mirror**: any page URL + `index.md` returns that page's source
with OKF frontmatter (e.g. `/lessons/01-open-science/index.md`).
- Rendered pages carry `<link rel="alternate" type="text/markdown">` and
`okf:*` meta tags; `robots.txt` advertises all of the above.
`docs/about/ai-agents.md` is the human- and agent-readable guide.
After content changes, regenerate the indexes (CI fails on drift) and check
the bundle:
```bash
python scripts/okf_validate.py docs
python scripts/check_links.py docs
python scripts/gen_llms_txt.py
zensical build --clean && python scripts/postbuild_agent_surface.py site
```
## Using these materials
All content is licensed [CC BY 4.0](LICENSE): use it in workshops and
courses, adapt it to your discipline, remix it, with attribution. Suggested
attribution:
> Adapted from "DUST 2026: Open Science Training" by Tyson L. Swetnam
> (<https://unm-carc.github.io/dust-2026/>), CC BY 4.0, which builds on
> DUST 2025, CyVerse FOSS, and the GPT 101 workshop.
**Citation (BibTeX):**
```bibtex
@misc{swetnam2026dust,
title = {DUST 2026: Open Science Training},
author = {Swetnam, Tyson L.},
year = {2026},
howpublished = {\url{https://unm-carc.github.io/dust-2026/}},
note = {Licensed under CC BY 4.0}
}
```
See [docs/about/credits.md](docs/about/credits.md) for source materials,
contributors, and funding.
## Deployment
Pushing to `main` runs OKF validation and the link check, verifies the
llms.txt indexes are current, builds the site, adds the agent surface, and
deploys to GitHub Pages via `.github/workflows/docs.yml`. The workflow enables
Pages automatically (`configure-pages` with `enablement: true`); if the first
deploy fails on permissions, set Settings → Pages → Source to "GitHub Actions"
once.
## Contact
- **Author:** Tyson L. Swetnam, UNM Center for Advanced Research Computing
- **Email:** <tswetnam@unm.edu>
- **ORCID:** [0000-0002-6639-7181](https://orcid.org/0000-0002-6639-7181)
- **Issues and suggestions:** <https://github.com/UNM-CARC/dust-2026/issues>