---
tags: ome2022, plan
---

# NG-Metadata

## Introduction during viewers

- (send to mailing list?)
- 10 minutes
- Perhaps something that was missed
- NGFF / Metadata "detour" from OMERO
- Critical for IDR, more fundable, etc.
- Ecosystem, Java, portability, growing, more viewers

Note to the audience: NGFF specification discussions happening regularly elsewhere.

## High-level outline:

- [NFDI](#NFDI), i.e. Why am _I_ talking about this?
- [Web of metadata (_Any others?_)](#Web-of-metadata), i.e. _What_ are we talking about?
- [What are we trying to solve?](#What-are-we-trying-to-solve)
- [Overview of the solution-space (_intros or not_?)](#Overview-of-the-solution-space), i.e. _What could we possbily do?_
- [Use cases](#Use-cases), i.e. _Where else could we use this?_
- [Plan of action](#Plan-of-action), i.e. _Who is doing what when?_

For the audience: But there are going to be `...` sections for you to add your own, so be ready.

## NFDI (Personal)

(Internal: _Assuming a positive announcement today_ :crossed_fingers:)

- Building up resources as part of NFDI4BIOIMAGE
- Goal of linking all of research data in Germany
- Key components of that are of course common formats (NGFF)
- But also common descriptions for the datasets.
- OME's specific role


- Short history:
- See RIKEN presentation (key-value pairs, etc.)
- But update with Quarep.


## Web of metadata

- Metadata spaces that need linking:
- OME, IDR, REBMI, OLS Ontologies, NBOQ, NGFF, BBMRI, DICOM
- Working throughout the presentation to a "linked" version of these


- Examples of interactions: (Perhaps "ideal scenario")
- "I want to add my metadata to..."
- "I want to take responsibility for a part of the model"
- "I need to work together with bioschema/some library/etc."
- "I want to migrate my data from OMERO to a library"
- Stretch goals
- Google search
- DOIs & library sciencies
- Federated queries
- ...

## Current on-going efforts

Within our community, there have been multiple related efforts:

- 2017: REMBI workshop
- 2018: RIKEN, of course, with Norio and SSDB, ([link](https://gitlab.com/openmicroscopy/incubator/ome-owl/-/commits/master/ontology/owl/ome_core/ome_core.owl.ttl))
- 2020: 4DN, then BINA, the Quarep have been working hard on developing QC models
- 2021/2022: At the same time, the IDR has been working to develop search, https://github.com/ome/omero_search_engine
- 2022: Delta Tissue: initial sponsor, interested in building cross-domain (incl. OMERO) models of particular diseases
- Led to the development of (early doors) omero-rdf
- ...anyone else?...

A lot going on!

## What are we trying to solve?

- Of course, no one of us can do everything. Join forces, expedite
- Own choices versus long term maintenance
- But to do that we need a common framework, language agnostic
- Extensible (early use case), Ontology support
- Standards (re-use) (mention comparison to XML?)
- ...what else?...
- (include federation?)

## Overview of the solution-space

- "XML, JSON, RDF, openminds, salad, linkml, _oh my_"
- "assembly on the mainframe" (KG)
- "naive of one another"

### A deeper dive into linkml.io and recent work

* imports
* spreadsheets
* conversions
* a model
* Avoiding "not-invented-here"
* met with some in San Diego
* relationship to OBO community
* Other pros & cons


## Practical use cases and examples

Time permitting, with getting our hands dirty.

- Building your first model
- stick & balls
- Channels
- Rendering
- (Internal: other ideas?)
- Will: Image collections, e.g. linking to images in other collections

## Discussion / Open questions
- Use in NGFF or not?
- Naming (can't be "next" forever)
- Versioning
- ....

## Plan of action

- Goal of an import DAG
- Who would like to be involved?
- Concrete tasks