2026 usrse CON talk to propose#3
Draft
yarikoptic wants to merge 11 commits into
Draft
Conversation
SOUL.md (mission):
- Promote Reuse/Compose/Extend/Standardize to five verbs by adding
*Automate*; capture the meta-automation (AI-maintained harness) angle.
- Add the SciOps maturity ladder (Johnson et al., 2024) alongside
STAMPED (Macdonald et al., 2026): STAMPED = per-artifact, SciOps =
team-level operations. Spell out the 5-verb -> L1-L5 mapping; cite
the L3 exemplars (BIDS, NWB, DataLad, DANDI, brainlife.io) the
SciOps paper itself names.
- Fix STAMPED expansion to the canonical Self-containment / Tracking /
Actionability / Modularity / Portability / Ephemerality /
Distributability.
- Add "Many lenses, one stack" recurring framing (Architectural /
Procedural / Sharing / Compositional / Purpose / Maturity ->
MVC / STAMPED / FAIR / YODA / project-purpose / SciOps+AI-ladder).
- Add "Static-first vs. service-tied UI" framing and Hexagonal /
Ports-and-Adapters citation (Cockburn 2005), Smart UI anti-pattern
(Evans, DDD 2003), Service Layer / Anemic Domain Model (Fowler).
- Add "HI <-> AI policy spectrum" recurring framing (Reject /
Accept-with-disclosure / Spec-driven AI-generated / Autonomous)
with CON-portfolio mapping; cite ICMJE Jan 2026 and the
melissawm/open-source-ai-contribution-policies catalog.
- Section 5 canonical-citations gains entries for STAMPED, SciOps.
- Section 7 story-arc list adds MVC-at-stack-scale; the five-verb
spine entry now includes the SciOps level mapping.
INDEX.md (catalog):
- Register two new talks: 2026-usrse-con-talk.html (US-RSE'26 draft)
and 202x-mvc-stack.html (stub for a future MVC-focused talk).
- Topic-lookup section gains rows for MVC framing, Schema-driven UIs
("the Model materializes the View"), Validation harmonization
(con/validation), Multi-dimensional framings, Backend-coupled
service-tied-UI anti-pattern / Hexagonal contrast, Automate /
harness / meta-automation, Five-verbs <-> SciOps maturity levels,
HI <-> AI policy spectrum.
Co-Authored-By: Claude Code 2.1.138 / Claude Opus 4.7 <noreply@anthropic.com>
Outline-only reveal.js stub for a future CON talk that argues the four
CON verbs (Reuse / Compose / Extend / Standardize) converge on
Model-View-Controller at the *stack* scale rather than the app scale.
Promoted from a 4-slide mini-section that lives inside the US-RSE deck.
What's in place:
- Title slide with [WiP] marker and pre-generated QR code (URL is
determined by filename per the talks-repo convention).
- "One thesis, two parts" + "Why at the stack scale" openers.
- Models section: dataset-layout / per-file / metadata / storage
stubs. Metadata stub spells out LinkML / pydantic / JSON Schema /
SHACL alongside HED / DUO / PROV.
- Views section: browse-the-archive / tabular / programmatic /
external-services / **schema-driven UIs** (vjsf -> DANDI meditor,
shacl-vue -> forms and research-group websites; cites Hanke et
al.'s LinkML metadata-driven workflow on ReproTube) / long-form
narrative.
- Controllers section: acquisition->layout / reproducible execution
/ logistics / data->derivative / **validation** ("the recursive
Standardize moment": con/validation harmonizes bids-validator +
HED + pynwb + zarr + nwbinspector + OME-Zarr, deployed in
dandi-cli, VisiData for triage).
- "Contrast: the 'service-tied UI' pattern" section (4 slides) names
Smart UI (Evans, DDD 2003), Anemic Domain Model (Fowler), Hexagonal
Architecture / Ports & Adapters (Cockburn 2005), GoF Adapter /
Strategy / Facade, Service Layer (PoEAA). Counter-pattern: static-
first archives, schema-driven UIs, JAMstack-style decks.
- "One stack, many lenses" section: 6-row table for Architectural /
Procedural / Sharing / Compositional / Purpose / Maturity, mapping
to MVC / STAMPED / FAIR / YODA / project-purpose / SciOps + AI-
coding ladder. Maturity row spells out SciOps L1-L5 and names the
paper's L3 exemplars.
- "Three things to do this week" closer.
QR code: pics/202x-mvc-stack-qrcode.png (generated with
qrcode/`qr` from URL the file will resolve to).
TODO markers in-file flag what each stub slide needs filled in when
promoted to a real talk.
Co-Authored-By: Claude Code 2.1.138 / Claude Opus 4.7 <noreply@anthropic.com>
A draft reveal.js deck for the US-RSE'26 15-min talk and its
proposal-stage companion materials. The deck assembles a single
coherent argument out of slides reused from the existing CON corpus
(2022-nih-compcore, 2023-brain-dandi, 2023-bids-dicom, 2024-distribits-
datalad, 2025-distribits-YODA, 2026-ca-origami-retreat-aicoding) and
wires them onto a new five-verb spine.
2026-usrse-con-talk.html (the deck)
-----------------------------------
Title carries a [WiP] marker; pre-generated QR code references the
predictable datasets.datalad.org/.../<TALK-ID>.html URL.
The five-verb spine:
- Reuse: NeuroDebian + PyMVPA history; duecredit / citeproc-py;
con/duct (built on brainlife's smon); con/citations-collector as
the modern continuation of duecredit.
- Compose: DataLad sandwich mermaid + extensions diagram; small-
acquisition-and-compute units (ReproIn / HeuDiConv / NeuroConv /
con/nwb2bids / ReproStim / ReproNim-containers / con/duct);
registry.datalad.org federation stats.
- Extend: pkg-exppsy -> NeuroDebian -> Debian Med / Debian Science /
conda-forge; citeproc-py co-maintainership; BIDS Steering Group;
DataLad RUNCMD -> BEP028 BIDS provenance lift.
- Standardize: BIDS as a meta-standard (local BIDS-minder.svg);
schemas-as-first-class (LinkML / pydantic / JSON Schema / SHACL);
OBC consent layer; federated archives (DANDI + EMBER + OpenNeuro)
with a layered table for **common standards / infra (AWS S3 +
MinIO) / data logistics (git-annex) / processing (containers,
BIDS-Apps)**.
- Automate: where we automate (CI on every PR, daily-tested git-annex,
con/tinuous CI-log archival, auto-rebuilt ReproNim/containers,
ReproIn/HeuDiConv at the scanner, auto-mirrored dandisets,
con/validation+dandi-cli); cost = harnesses to maintain; meta-
automation move via con/skills + con/yolo; closing slide maps the
five verbs to SciOps L1-L5 with the paper's own L3 exemplars.
Plus:
- "Reuse, in reverse" 3-up (AnnexTube / mykrok / con/serve) with a
WiP Wayback-archive callout (con.github.io/serve/tools/web/
wayback-archive/) -- "local archive of the IA's archive of your
project's archive".
- HI+AI section: STAMPED (per-artifact) + SciOps (team-operations)
framed as complementary; AI-coding ladder presupposes SciOps L3;
L4 -> L5 gap is what AI closes for a small RSE center. New "HI <->
AI -- every project picks its own policy" slide with a 4-stance
spectrum (Reject / Accept-with-disclosure / Spec-driven AI-generated
/ Autonomous), CON-portfolio mapping (git-annex / DataLad+DANDI /
AnnexTube+mykrok+citations-collector / con/skills+con/yolo), and
ICMJE Jan 2026 + melissawm/open-source-ai-contribution-policies
citations.
- "Why it composes -- MVC at the stack scale" mini-section (opener
+ Models + Views + Controllers). Punchline: pick any cell and
swap it; the rest still works. The same block is the seed of the
standalone 202x-mvc-stack.html stub.
- Monday checklist (6 entries; one per verb plus Reuse-in-reverse,
with the Automate entry pointing at con/tinuous).
2026-usrse/talk-proposal-draft.md (the abstract)
------------------------------------------------
Tracks the deck:
- Five-verb framing in title and opening paragraph.
- Per-verb bullets including the new Automate bullet with the
SciOps L4 ("SciOps pipelines") attribution and meta-automation
pitch toward L5 (Optimizing).
- MVC-at-stack-scale bullet covers the architectural rereading.
- HI+AI bullet folds in the 4-stance policy spectrum.
- References add STAMPED (16), SciOps (16a), ICMJE Jan 2026 (16b),
and the melissawm OSS-policies catalog (16c).
- Connection-to-mission Automate bullet ties to SciOps L4/L5.
2026-usrse/projects-audit.md (companion audit)
----------------------------------------------
Gap-list and recommendation tracker comparing this deck against the
broader CON ecosystem (50+ active repos per `gh api orgs/con/repos`).
Records that recommendations 1-5 (con/tinuous, con/yolo+con/skills,
con/citations-collector, con/nwb2bids, OBC) have been applied; flags
the remaining marginal-namedrop candidates (con/external-services,
con/shub, con/noisseur, con/upptime, con/flux, datalad-installer,
fail2ban, psychtoolbox-3-debian).
QR code: pics/2026-usrse-con-talk-qrcode.png.
Co-Authored-By: Claude Code 2.1.138 / Claude Opus 4.7 <noreply@anthropic.com>
Member
Author
|
pushed more changes and also envisioned supplemental talk on "stack level MVC pattern" also filed since unfortunately previews ATM are hard/impossible here. And I should concentrate on abstract alone -- the rest is nice to see how it would shape up but I wonder now if not to propose two talks, where first would be along the dimensions of the presenting historical perspective and actions like automation etc, and then next on MVC, diving more into tech aspects of structuring, integration etc. WDYT? |
yarikoptic
commented
May 11, 2026
|
|
||
| ## Keywords | ||
|
|
||
| reuse, modularity, community standards, federation, reproducibility, distributed data management, neuroinformatics, FAIR, open source, research software engineering |
Collaborator
There was a problem hiding this comment.
Suggested change
| reuse, modularity, community standards, federation, reproducibility, distributed data management, neuroinformatics, FAIR, open source, research software engineering | |
| reuse, modularity, community standards, federation, reproducibility, distributed data management, neuroinformatics, FAIR, open source |
a bit much for a RSE conference 😉
|
|
||
| ## Abstract | ||
|
|
||
| For two decades the team of the **Center for Open Neuroscience (CON)** has been building an open, largely domain-agnostic research-software stack — first for neuroimaging, then for neuroscience broadly, and now used well beyond. We did it by repeating five actions: **Reuse** what already exists; **Compose** small modules into environments and systems instead of shipping silo-ed monoliths; **Extend** the upstream projects we depend on; **Standardize** the "languages" of data we exchange; and **Automate** everything we can — because nothing scales without it, and the harness itself is now another thing AI helps us maintain. These five verbs are not just our private taxonomy: they map directly onto the operational-maturity climb described by the **SciOps Capability Maturity Model** (Johnson *et al.*, [arXiv:2401.00077](https://arxiv.org/abs/2401.00077), 2024) — Reuse/Compose/Extend/Standardize are the practices of *Level 3 (Defined)*, where Johnson *et al.* themselves cite BIDS, NWB, DataLad/git-annex, DANDI, and brainlife.io as canonical exemplars; Automate is *Level 4 (Scalable)* with its "SciOps pipelines"; and AI-in-the-loop is *Level 5 (Optimizing)*, the pinnacle Johnson *et al.* reserve for "closing the discovery loop with the assistance of artificial intelligence". Architecturally the result is a familiar pattern lifted to the *stack* scale — **Model–View–Controller**: layered, standardized **models** (BIDS / NWB layouts; [LinkML](https://linkml.io/) / [pydantic](https://docs.pydantic.dev/) / [JSON Schema](https://json-schema.org/) / [SHACL](https://www.w3.org/TR/shacl/) for metadata; `git`+`git-annex`+DataLad as a content-addressed storage model); a plurality of interchangeable **views** — including ones the *Model itself materializes*: archive UIs (datasets.datalad.org, DANDI, OpenNeuro), schema-driven editors and websites ([vjsf](https://vjsf.koumoul.com/) for the DANDI meditor; [shacl-vue](https://github.com/psychoinformatics-de/shacl-vue) from M. Hanke's group rendering forms *and entire research-group websites* from SHACL), tabular surfaces ([Datasette](https://datasette.io/), [VisiData](https://www.visidata.org/)), programmatic APIs, and the DataLad Handbook; and small, single-purpose **controllers** (HeuDiConv, NeuroConv, ReproStim, `datalad run`, `con/duct`, BIDS-Apps, `con/validation`). The "Age of AI" doesn't change any of that — it expedites the actions, and in turn benefits from a modular, transparent stack that lets humans and agents "divide and conquer" without duplicating effort. |
Collaborator
There was a problem hiding this comment.
@yarikoptic I recommend you take a pass through and purge the em-dashes and egregious colon/semi-colon, rephrasing into a more natural human prose
|
|
||
| ## Abstract | ||
|
|
||
| For two decades the team of the **Center for Open Neuroscience (CON)** has been building an open, largely domain-agnostic research-software stack — first for neuroimaging, then for neuroscience broadly, and now used well beyond. We did it by repeating five actions: **Reuse** what already exists; **Compose** small modules into environments and systems instead of shipping silo-ed monoliths; **Extend** the upstream projects we depend on; **Standardize** the "languages" of data we exchange; and **Automate** everything we can — because nothing scales without it, and the harness itself is now another thing AI helps us maintain. These five verbs are not just our private taxonomy: they map directly onto the operational-maturity climb described by the **SciOps Capability Maturity Model** (Johnson *et al.*, [arXiv:2401.00077](https://arxiv.org/abs/2401.00077), 2024) — Reuse/Compose/Extend/Standardize are the practices of *Level 3 (Defined)*, where Johnson *et al.* themselves cite BIDS, NWB, DataLad/git-annex, DANDI, and brainlife.io as canonical exemplars; Automate is *Level 4 (Scalable)* with its "SciOps pipelines"; and AI-in-the-loop is *Level 5 (Optimizing)*, the pinnacle Johnson *et al.* reserve for "closing the discovery loop with the assistance of artificial intelligence". Architecturally the result is a familiar pattern lifted to the *stack* scale — **Model–View–Controller**: layered, standardized **models** (BIDS / NWB layouts; [LinkML](https://linkml.io/) / [pydantic](https://docs.pydantic.dev/) / [JSON Schema](https://json-schema.org/) / [SHACL](https://www.w3.org/TR/shacl/) for metadata; `git`+`git-annex`+DataLad as a content-addressed storage model); a plurality of interchangeable **views** — including ones the *Model itself materializes*: archive UIs (datasets.datalad.org, DANDI, OpenNeuro), schema-driven editors and websites ([vjsf](https://vjsf.koumoul.com/) for the DANDI meditor; [shacl-vue](https://github.com/psychoinformatics-de/shacl-vue) from M. Hanke's group rendering forms *and entire research-group websites* from SHACL), tabular surfaces ([Datasette](https://datasette.io/), [VisiData](https://www.visidata.org/)), programmatic APIs, and the DataLad Handbook; and small, single-purpose **controllers** (HeuDiConv, NeuroConv, ReproStim, `datalad run`, `con/duct`, BIDS-Apps, `con/validation`). The "Age of AI" doesn't change any of that — it expedites the actions, and in turn benefits from a modular, transparent stack that lets humans and agents "divide and conquer" without duplicating effort. |
Collaborator
There was a problem hiding this comment.
Might want to do your markdown thing for the PR review (one line per) - makes it easier to leave comments
Claude suggests "SCALE (Standardize, Compose, Automate, Leverage, Extend)" as an easy way to remember/present these principles
Collaborator
There was a problem hiding this comment.
@yarikoptic Please render a simple PDF view as they describe from their Google sheet instructions - need to see how this (or your humanized version) fits in formatting/length
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.