Skip to content

SOUL.md

Latest commit

 

History

History
554 lines (505 loc) · 28.8 KB

SOUL.md

File metadata and controls

554 lines (505 loc) · 28.8 KB
Raw

SOUL — what these talks are for and how they look

This file is the long-lived "soul" of the Center for Open Neuroscience (CON) talks repository. It captures the recurring mission, voice, visual style, resources, and citation conventions that any CON talk should inherit. New talks should read this first; reusable slide content is indexed in INDEX.md.

The technical "how to author / build" guide lives in CLAUDE.md.

1. Mission

CON talks exist to advocate, by way of working examples, for a single recurring thesis:

Build science as reusable, composable, extensible, standardized infrastructure — and, as much as pragmatic, bridge upstream rather than re-implementing.

Concretely, almost every talk in the corpus revisits some subset of:

  • Reuse — "the cheapest reproducible thing is the one you didn't have to build". pkg-exppsy → NeuroDebian, PyMVPA ←→ scikit-learn → nilearn, duecredit → citeproc-py, DataLad → git + ←→ git-annex, con/duct ← brainlife smon. We join upstreams instead of forking.
  • Compose — small units (HeuDiConv, ReproStim, ReproNim/containers, con/duct, DataLad subdatasets, DataLad extensions, BIDS dataset modules, ...) over silo'd monoliths.
  • Extend — when the commons we depend on need care, we stay on as (co)maintainers (fail2ban, citeproc-py) as long as needed, or even take over the development (heudiconv); we generalize and standardize ad-hoc work upstream (DataLad RUNCMD → BEP028 BIDS provenance); we push packages into Debian / Debian Med / Debian Science / conda-forge so others reuse them in turn but also extend our "workforce".
  • Standardize — common tech and underlying data models (git, git-annex), common data standards (BIDS, NWB, HED, DICOM), common metadata (LinkML, concepts.datalad.org), common organizational layouts (BIDS, YODA, STAMPED). Standards are the language across labs and across HI ↔ AI.
  • Collaborate — ad-hoc solo projects produce silos and fragile tools; only collaboration widens the range of use cases enough to force genuine robustness and generality. Every long-lived project in the CON portfolio is a sustained multi-stakeholder effort: PyMVPA with Michael Hanke (two decades and counting); NeuroDebian — itself embedded inside the even larger Debian community, contributing through and alongside Debian Med and Debian Science blends rather than standing apart; the principle is that specialized projects (NeuroDebian, Debian Med) should live within broader communities, not breed outside them as competing ecosystems; DataLad with Michael Hanke and the Psychoinformatics Lab at FZJ; DANDI with Satra Ghosh, Kitware, Ben Dichter, and the NWB / DANDI community; BIDS — a genuinely large-scale community standard now spanning hundreds of contributors. Contributions need not be code: testing, troubleshooting, and articulating new use cases are what turn a solo hack into infrastructure others can depend on.
  • Automate — none of the above scales without it. Unit/integration CI on every PR; daily-tested git-annex against DataLad; con/tinuous archiving CI logs and artifacts before they expire; auto-rebuilt ReproNim/containers; ReproIn + HeuDiConv driving DICOM→BIDS at the scanner; ReproStim capturing all stimuli; auto-mirrored dandisets; con/validation + dandi-cli on every release; auto-deployed handbook + per-archive docs. In SciOps (Johnson et al., 2024) terms this is Level 4 (Scalable) — what the paper calls "SciOps pipelines". Cost: the harness is itself code we maintain — and that is exactly where AI assistance becomes the most viable way forward as meta-automation: using Claude Code + con/skills
    • con/yolo to maintain the automations themselves, taking a small RSE center from SciOps L4 toward L5 (Optimizing).
  • Federate, don't recentralize — registry.datalad.org, DANDI/EMBER/ OpenNeuro, datasets.datalad.org, neurobagel, ... all distribute discovery and/or storage rather than becoming a single platform. Joining forces where we can contribute (neurobagel, babs) to instill our principles and make interchangeable to be "federatable", instead of taking over
  • Make things Convenient -- buggy, unreliable, hard to use, or requiring manual action when could be automated -- is INCONVENIENT.

Recurring framings the speaker leans on:

  • "Where data go to die / how data are reincarnated" → archives + reuse (2022-tx-big-neuroscience.html, 2025-ca-origami-retreat.html, 2023-brain-dandi.html).

  • "Sandwich" / "burger" / "minder" — DataLad as layered tech over git-annex over git, with extensions / remote helpers as additional layers.

  • "All standards are bad, but some are used" (D. Clunie, MICCAI 2017) — used to justify pragmatic standardization (2023-bids-dicom.html).

  • "Make re-use convenient" — every time data integration is mentioned, the conclusion arrives back at this.

  • YODA as a vocabulary of reproducibility (Version control everything; Look up you must not; Modular composition).

  • HI + AI — the recent (2026) framing: the same self-contained, well- described, version-controlled artifacts serve both human investigators and AI agents.

  • HI ⇄ AI policy spectrum — different projects need different AI-acceptance policies; CON's portfolio spans the full spectrum and we show that in talks rather than picking one. Four canonical stances:

    • Reject any AI-generated content (e.g. git-annex upstream contributions in practice — Joey Hess's "policy" page is a famous Feb-30 satire; analogues in OSS: Zig, Krita, Clojure, QEMU).
    • Accept with disclosure (DataLad / DANDI; Co-Authored-By: Claude… trailers; @pytest.mark.ai_generated; analogues: NumPy, Kubernetes, Linux kernel, Django).
    • Spec-driven AI-generated (HI specifies, AI writes, HI reviews & commits) — AnnexTube, mykrok, con/citations-collector, parts of dandi-cli (LAD specs + AI-generated tests).
    • Autonomous agents in the loop — con/skills + con/yolo for triage, PR review, dependency updates.

    Common ground (all four): AI cannot be an author (ICMJE Jan 2026 update); humans retain full responsibility; AI use must be disclosed inside the artifact (commit trailer, methods section, acknowledgments). STAMPED Tracking is what makes that mechanical. SciOps positions these as a climb from L3-floor up to L5 (Optimizing). External catalog of declared OSS stances: melissawm/open-source-ai-contribution-policies.

  • Static-first vs. service-tied UI — a recurring contrast: most academic data infrastructure is built as a "service-tied UI" (LIMS / ELN / bespoke Flask app / Drupal portal) where the View is welded to a running backend, so portability and reuse collapse the moment the server stops. CON's pattern is the inverse: the Model (BIDS / NWB / LinkML / DataLad-tracked file tree) is the artifact; Views are derived; services are one Adapter among many. Cite Cockburn's Hexagonal Architecture / Ports & Adapters (2005), Evans's Smart UI anti-pattern (DDD 2003), Fowler's Anemic Domain Model + Service Layer, and GoF Adapter / Strategy / Façade when the audience is enterprise-software-literate.

  • Many lenses, one stack — the same project / artifact can be legitimately described along several overlapping dimensions, each picking out a different property:

    • Architectural — Model / View / Controller (mini-section in 2026-usrse-con-talk.html; the standalone spine _backdrawer_/202x-mvc-stack.html is currently shelved).
    • Procedural — STAMPED (Self-containment, Tracking, Actionability, Modularity, Portability, Ephemerality, Distributability — see Macdonald, Baker, To & Halchenko, 2026, stamped-principles/stamped-paper).
    • Sharing — FAIR (Findable, Accessible, Interoperable, Reusable).
    • Compositional — YODA (modular, "look up you must not", version-control everything).
    • Purpose — acquisition / curation / archive / analysis / governance / sharing / training.
    • Maturity — operational maturity per SciOps (Johnson et al., 2024; arXiv:2401.00077; five-level Capability Maturity Model for rigorous scientific operations — cited from the STAMPED paper); agentic-readiness axis via the AI-coding ladder L1–L5 (2026-ca-origami-retreat-aicoding.html).

    Use lenses together, not as competitors. When two lenses disagree about whether something is "good", that's usually a real design tension worth surfacing on the slide, not noise to smooth over.

2. Audience and tone

  • Audiences range from neuro-domain (NWB/DANDI, BIDS WGs, ReproNim webinars) to general RSE / HPC / data-management (US-RSE, distribits, CA Origami retreat, NIH compcore). Talks usually pick a domain hook and then quickly broaden to the domain-agnostic stack.
  • Voice: first-person plural ("we"), occasionally first-person ("Yarik's first move was..."), liberal Yoda phrasing on YODA-flavored decks ("Track you must!", "Look up you must not"), and dry humor in Reality Check / disclaimer slides ("idiocracy-fixed.gif").
  • Concrete "Take home" / "Monday checklist" lists at section ends are a staple — the audience should leave with one thing they can do this week.
  • Respect for collaborators is loud: every deck ends with an Acknowledgements slide naming funders (NIH, NSF, BMBF, ERDF, BInC), collaborators (HBP, CONP, VBC, ReproNim, OpenNeuro, CBRAIN, brainlife, DANDI), and upstream people (Joey Hess, Michael Hanke, the DataLad team).
  • Solidarity tag: a small Ukrainian flag ribbon (pics/Ukrainian_Blue-Yellow_ribbon.svg) appears in the title slide logo strip — keep it.

3. Visual style and reveal.js conventions

All decks are reveal.js HTML, theme beige, with the highlight plugin's monokai syntax theme. The repository's vendored reveal.js/, reveal.js-mermaid-plugin/, and css/custom.css are the canonical assets. Do not introduce a new theme without reason.

Standard header

<!doctype html>
<html>
<head>
  <meta charset="utf-8">
  <meta name="viewport" content="width=device-width, initial-scale=1.0,
                                 maximum-scale=1.0, user-scalable=no">

  <title>Talk Title</title>
  <meta name="description" content="Slides for VENUE / DATE">
  <meta name="author" content=" Yaroslav O. Halchenko ">

  <link rel="stylesheet" href="css/custom.css">
  <link rel="stylesheet" href="reveal.js/dist/reset.css">
  <link rel="stylesheet" href="reveal.js/dist/reveal.css">
  <link rel="stylesheet" href="reveal.js/dist/theme/beige.css">
  <link rel="stylesheet" href="reveal.js/plugin/highlight/monokai.css">
</head>
<body><div class="reveal"><div class="slides">
  <!-- slides go here -->
</div></div>
<script src="reveal.js/dist/reveal.js"></script>
<script src="reveal.js/plugin/notes/notes.js"></script>
<script src="reveal.js/plugin/markdown/markdown.js"></script>
<script src="reveal.js/plugin/highlight/highlight.js"></script>
<script src="reveal.js-mermaid-plugin/plugin/mermaid/mermaid.js"></script>
<script>
  Reveal.initialize({
    hash: true,
    width: 1400, height: 1050,         /* recent default; older talks used 1920x1080 */
    margin: 0.1,
    minScale: 0.2, maxScale: 1.0,
    progress: true, history: true, center: true,
    controls: true, slideNumber: 'c',
    pdfSeparateFragments: false,
    pdfMaxPagesPerSlide: 1,
    pdfPageHeightOffset: -1,
    transition: ['slide', 'fade'],
    plugins: [ RevealMarkdown, RevealHighlight, RevealNotes, RevealMermaid ]
  });
</script></body></html>
  • Canvas size: 1400 x 1050 is the current default (2025-distribits-YODA, 2026-repronim-YODA-BIDS-webinar, 2026-ca-origami-retreat-aicoding); older decks used 1920 x 1080. Stick to 1400×1050 for new talks unless the venue insists on 16:9.
  • Plugins: always include RevealMarkdown, RevealHighlight, RevealNotes, RevealMermaid. Recent decks add RevealSearch (Ctrl+Shift+F) — fine to include.
  • data-src for images so reveal.js lazy-loads them.

Title slide template

<section><section>
  <a href="http://centerforopenneuroscience.org/">
    <img data-src="pics/con-ccn-dartmouth-letterhead.svg" style="height:150px;"/>
  </a>
  <h2 style="margin-top:0.3em;margin-bottom:0.3em;">Title goes here</h2>
  <p style="font-size:0.75em;font-style:italic;">Optional subtitle / yoda one-liner</p>

  <div style="margin-top:0.3em;text-align:center">
    <table style="border:none;">
      <tr><td>
        Yaroslav O. Halchenko<br>
        <small><a href="https://twitter.com/yarikoptic" target="_blank">
          <img data-src="pics/twitter.png" style="height:30px;"/>@yarikoptic</a></small>
        <small><a href="https://fosstodon.org/@yarikoptic" target="_blank">
          <img data-src="pics/mastodon-icon.svg" style="height:30px;"/>@yarikoptic@fosstodon.org</a></small>
      </td></tr>
      <tr><td>
        <small><br>
          <a href="http://centerforopenneuroscience.org/">Center for Open Neuroscience</a><br>
          <a href="https://pbs.dartmouth.edu/">Department of Psychological and Brain Sciences</a><br>
          <a href="https://www.dartmouth.edu/ccn/">Center for Cognitive Neuroscience</a><br>
          <a href="http://www.dartmouth.edu">Dartmouth College, New Hampshire, USA</a>
        </small>
        <img style="width:160px;" data-src="pics/<TALK-ID>-qrcode.png"/>
      </td></tr>
    </table>
  </div>

  <small>
    <strong>VENUE — DATE</strong><br/>
    Live slides/<a href="https://datasets.datalad.org/centerforopenneuroscience/talks/.git">Sources</a>:
    <a href="https://datasets.datalad.org/centerforopenneuroscience/talks/<TALK-ID>.html">…/<TALK-ID>.html</a>
  </small>
  <small>
    <a href="http://datalad.org"><img style="height:100px;margin:15px" data-src="pics/datalad_D.svg"/></a>
    <a href="http://neuro.debian.net"><img style="height:100px;margin:15px" data-src="pics/neurodebian.png"/></a>
    <a href="http://repronim.org"><img style="height:100px;margin:15px" data-src="pics/repronim-logo-vertical.svg"/></a>
    <a href="https://open-brain-consent.readthedocs.io"><img style="height:100px;margin:15px" data-src="pics/OBC_LogoCheck.svg"/></a>
    <a href="https://dandiarchive.org"><img style="height:100px;margin:15px" data-src="pics/dandi-logo-square.svg"/></a>
    <a href="https://bids.neuroimaging.io/"><img style="height:100px;margin:15px" data-src="pics/BIDS_Logo.png"/></a>
    <a href="https://standforukraine.com/"><img style="height:100px;margin:15px" data-src="pics/Ukrainian_Blue-Yellow_ribbon.svg"/></a>
  </small>
</section></section>
  • QR code points to the live-slides URL on datasets.datalad.org/centerforopenneuroscience/talks/<TALK-ID>.html. Generate one per talk and save as pics/<TALK-ID>-qrcode.png.
  • Logo strip at the bottom: pick the relevant subset from the canonical set — DataLad, NeuroDebian, ReproNim, OBC, DANDI, BIDS, YODA, Ukraine ribbon. Keep order roughly stable.
  • Author block: keep the four-line affiliation (CON / PBS / CCN / Dartmouth) verbatim across talks for consistency.
  • The first <section> is wrapped in another <section> (reveal.js vertical stack convention) so a vertical sub-deck can sit under the title without restructuring.

Slide-construction patterns to reuse

These conventions show up across multiple talks; reuse them rather than inventing new ones.

  • Section divider with a "Challenge:" headline on a soft gradient: <section data-background-gradient="radial-gradient(white, #f7dfd3)">. Used heavily in 2022-nih-compcore.html, 2023-brain-dandi.html, 2023-lbl-building-dandi.html.
  • Buddha / Yoda backgrounds for "nirvana" sections: data-background="pics/digits-budda.svg" data-background-opacity="0.9".
  • Markdown sub-decks: long content uses `
    <textarea data-template>… markdown …</textarea>
    ` — this is the dominant style in the 2025/2026 talks.
  • Layered fragment images (showing a dataset directory tree progressively reveal): r-stack or absolutely-positioned <img class="fragment"> layered with descending top/left offsets. See 2024-distribits-datalad.html "2017: DataLad crawler pipeline".
  • Mermaid diagrams for pipeline / sandwich layering / "the Vault" inbound/outbound. The mermaid plugin is loaded in every deck.
  • Aside notes (<aside class="notes">) — used freely; keeps speaker notes alongside the slide for the RevealNotes s shortcut.
  • "Take home" closer at the end of each section: a markdown bullet list, often with a single bolded action verb up front.

What NOT to redesign

  • Do not switch theme. The beige theme is the visual identity.
  • Do not introduce additional CSS files; extend css/custom.css if needed.
  • Do not introduce per-talk node_modules / build steps. Slides ship as static HTML opened straight from disk or via the published datasets.datalad.org mirror. The repo-level npm install / npm start exists for live-reload only (see README.md).

4. Resources (where things live)

  • Logos & images: pics/ (443 tracked files; see ls pics/ and grep). All the CON-flavored marks plus borrowed third-party assets in pics/borrowed/ — though that convention was applied inconsistently over the years, so plenty of borrowed material sits in pics/ root too. The lists below are a findability index, not a provenance record: for who owns what, read REUSE.toml.
  • YODA artwork: pics/yoda*.png/svg, pics/yoda-principles-reordered.png, pics/principle-vcs.png, pics/principle-computeenv.png, pics/principle-structure.png, pics/yoda-do-not-look-up.png, pics/yoda-all-the-way-down.png, pics/yoda-hierarchy-with-containers.png. Source poster: https://github.com/myyoda/poster/blob/master/ohbm2018.pdf.
  • DataLad ecosystem: pics/DataLad-minder.svg (the canonical DataLad ecosystem diagram), pics/datasets.datalad.org-*.png snapshots, pics/datalad-extensions.png.
  • DANDI: pics/dandi-slide-modalities.svg, pics/dandi-slide-services.svg, pics/dandi-slide-standards.svg, pics/dandi-slide-schema.svg, pics/DANDI-ecosystem.svg, pics/DANDI-shot-and-interactions.svg, pics/DANDI-FAIR.svg, pics/DANDI-users-{submitter,researcher,developer,PI}.svg.
  • BIDS: pics/bids-logo-wide.png, pics/BIDS_Logo.png, pics/BIDS-minder.svg (and .png), pics/bids-yoda.png, pics/bids-nipoppy.png, pics/bids-princeton.png.
  • NeuroDebian / PyMVPA: pics/neurodebian.png, pics/neurodebian_logo.svg, pics/neurodebian-overview.png, pics/neurodebian-user.png, pics/nd_overview.svg, pics/neuropy_history.svg, pics/pymvpa_icon.png, pics/pymvpa_logo_fromfusionposter.svg, pics/pymvpa-features.png, pics/pymvpa-classification.png, pics/pymvpa-searchlight*.png, pics/pymvpa-transfusion.png, pics/pymvpa_on_phone.jpg.
  • CON / Dartmouth letterhead: pics/con-ccn-dartmouth-letterhead.svg (always at the top of the title slide).
  • CI / con/tinuous: pics/con-tinuous-github.png, pics/con-tinuous-term.png, pics/con-tinuous-term-dandi-cli.png.
  • con/duct + traces: pics/webshot-con-duct.png, pics/screenshot-duct-{1..4}.png, pics/duct-mriqc-cerebra.png.
  • ReproNim 5 steps: pics/repronim-5steps.png (http://5steps.repronim.org).
  • OBC: pics/OBC_LogoCheck.svg, pics/obc-{main,ultimate,tools}.png.
  • Yoda goal cartoon (the speaker's "north star"): pics/yarik-goal.svg.
  • Funder / collaborator logos: pics/borrowed/ — nih.png, nsf_2020.png, nsf1.jpg, bmbf_2020.png, binc.png, erdf.png, cbbs_logo.png, LSA-Logo.png, fzj_logo.svg, hbp_logo.png, conp_logo.png, vbc_logo.png, repronim_logo.svg, openneuro_logo.png, cbrain_logo.png, brainlife_logo.png, dandi_logo.svg, bannerthanks.svg.
  • Stock comics / cartoons: pics/borrowed/phdcomics-notfinal.png, pics/borrowed/scm-history-2016.svg, pics/borrowed/versioncontrol.svg, pics/borrowed/idiocracy-fixed.gif, pics/borrowed/twitter-unsolicited-advice.png, pics/borrowed/chatgpt_german_sandwich.png, pics/borrowed/2011-book-avogadro-corp.png, pics/borrowed/2026-ai-intensifies.png, pics/borrowed/ai-ladder-skills.png, pics/borrowed/HH12-webshot-20130331.png, pics/borrowed/reproin-logo.jpg, pics/borrowed/crcns-vim-1-takendown.png, pics/borrowed/con-webshot-20150812-front-down.png.
  • External deps under 3rd-party/: MICCAI_2017_Clunie_DICOM.pdf (used as a citation source).

5. Citation conventions

CON talks cite primary literature inline near the relevant slide, in a small, gray, three-line block. The author of the slide (Halchenko) is bolded, e.g. <b>Halchenko, Y. O.</b>. URL is a direct DOI / PubMed link. Pattern:

<small>
  <a href="https://doi.org/10.21105/joss.03262">
    Y. Halchenko, K. Meyer, B. Poldrack, …, M. Hanke. <em>DataLad: distributed
    system for joint management of code, data, and their relationship.</em>
    Journal of Open Source Software, 6(63):3262, 2021.
    doi: 10.21105/joss.03262
  </a>
</small>

A QR code linking to the paper sometimes accompanies the citation (e.g. pics/2021-datalad-joss-qrcode.png next to the JOSS DataLad citation in 2024-distribits-datalad.html).

Canonical citations to keep at hand:

When citing other CON talks (frequently done — talks build on each other), link the live-slides URL on datasets.datalad.org/centerforopen neuroscience/talks/<TALK-ID>.html, optionally with a #/<slide>/<sub> fragment so the deep link goes to a specific slide.

When citing YouTube recordings (distribits 2024 talks, ReproNim webinars, the AI-coding talk), embed an image-as-link to the thumbnail, with the YouTube URL in a <small> underneath.

6. License and sharing

  • Repository is CC-BY-SA (LICENSE). Reuse and remix freely; share alike — but that covers what we authored. Slides quote the world, and the borrowed screenshots, figures, cartoons and logos are not ours to relicense. Saying so plainly is the same instinct as §2's loud acknowledgements and §5's citation blocks: credit is part of the work, not paperwork bolted onto it.
  • REUSE.toml (+ LICENSES/) is the authoritative per-file record of copyright and provenance — including for material whose origin could not be reconstructed, which is marked unknown rather than quietly absorbed under our license. Mechanics live under "Not everything here is ours" in README.md; what to do when you add an asset lives in CLAUDE.md.
  • Live mirror: https://datasets.datalad.org/centerforopenneuroscience/talks/.
  • Sources: this repository (git clone https://datasets.datalad.org/ centerforopenneuroscience/talks/.git).
  • Slides do not depend on a build step for viewing; opening <TALK-ID>.html directly in Chromium is the lightest path. PDF export via ?print-pdf URL fragment, or the decktape Docker recipe in README.md.

7. The recurring "story arcs"

For composing a new talk, these arcs are the materialized templates the existing decks reach for. Pick one as the spine, then borrow specific slides from INDEX.md.

  1. Origin → Stack → Today (2024-distribits-datalad.html, 2022-nih-compcore.html): History of one tool → what got layered on top → the ecosystem now.

  2. Challenges → Solutions → Take home (2022-nih-compcore.html, 2023-lbl-building-dandi.html): each section opens with a Challenge: slide on the radial-gradient background, then the answer, then a small "Overall:" / "Take home" wrap.

  3. YODA principle a day (2025-distribits-YODA.html, 2026-repronim-YODA-BIDS-webinar.html): one section per principle (Track / Portable Compute / Modular Composition / Look Up You Must Not), each ending with a Yoda quip.

  4. Nirvana / data archives (2022-tx-big-neuroscience.html, 2025-ca-origami-retreat.html, 2023-brain-dandi.html): Where data go to die → Reincarnation → Make re-use convenient.

  5. AI-coding ladder (2026-ca-origami-retreat-aicoding.html): layered "maturity" slides (Levels 1–5), then a 5-stage development loop, then real examples.

  6. Reuse / Compose / Extend / Standardize / Automate (2026-usrse/talk-proposal-draft.md, this repo's overarching framing): the five-verb spine that subsumes all of the above and is the recommended skeleton for cross-domain RSE audiences. Automate was the late-named fifth verb (added 2026-05-10) — it covers everything CI/harness-related (con/tinuous, daily-tested git-annex, auto-mirrored dandisets, ReproIn/HeuDiConv at the scanner, etc.) and carries the meta-automation (AI ↔ harness maintenance) handoff into the HI+AI section. The five verbs map directly onto the SciOps Capability Maturity Model (Johnson et al., 2024; arXiv:2401.00077):

    • L1 Initial — ad-hoc, manual; no verbs in play yet.
    • L2 Managed — lab-local Compose (YODA layouts; internal pipelines).
    • L3 Defined — Reuse / Compose / Extend / Standardize; FAIR data
      • FAIR workflows. The SciOps paper itself names BIDS, NWB, DataLad/ git-annex, DANDI, and brainlife.io as L3 exemplars.
    • L4 Scalable — Automate: "SciOps pipelines", semi-automated continuous workflows. Our CI + con/tinuous + auto-mirrored dandisets are exactly this.
    • L5 Optimizing — AI in the loop; closing the discovery loop. This is where the AI-coding ladder (2026-ca-origami-retreat- aicoding.html) plugs in, but it presupposes L3 below it.

    Practical implication for talk authoring: when you cite one verb, also state which SciOps level it corresponds to — RSE audiences recognize CMM language even when they don't know the specific paper.

  7. MVC at the stack scale (mini-section inside 2026-usrse-con-talk.html; the standalone stub _backdrawer_/202x-mvc-stack.html is currently shelved): the architectural re-reading of the four verbs — layered standardized Models (BIDS / NWB / LinkML / git+git-annex+DataLad), small single-purpose Controllers (HeuDiConv / NeuroConv / ReproStim / datalad run / con/duct / BIDS-Apps), interchangeable Views (datasets.datalad.org / DANDI / OpenNeuro / mykrok / AnnexTube / Datasette / VisiData / handbook). Useful as the integrating punchline near the end of any retrospective, or as the spine of a stand-alone talk for an RSE-heavy audience.

8. Author identity / fixed metadata

  • Speaker default: Yaroslav O. Halchenko (yaroslav.o.halchenko@dartmouth.edu, ORCID 0000-0003-3456-2493, @yarikoptic on Twitter, @yarikoptic@fosstodon.org on Mastodon).
  • Co-author centroids per the proposal draft (CON team): Cody Baker (0000-0002-0829-4790), Austin Macdonald (0000-0002-8124-807X), Isaac To (0000-0002-4740-0824), Vadim Melnik (ORCID TBD).
  • Affiliations block (verbatim): Department of Psychological and Brain Sciences, Dartmouth College / Center for Open Neuroscience.