GeoJupyter core community meeting

GeoJupyter core community meeting (2025-12-09)¶

Please add new agenda items under the New agenda items heading!

• Join us on Zoom

  • What time is the meeting in my time zone?

• Previous meetings

• GeoJupyter handy links:

  • GitHub org

  • Community calendar

  • Zulip chat

Attendees¶

Your name / GitHub ID / affiliation

• Matt Fisher / @mfisher87 / Schmidt DSE

• Nicolas Brichet / @brichet / QuantStack

Standing items¶

Follow-up from previous meeting(s)¶

New agenda items¶

• JupyterGIS updates

  • Story maps: Greg making lots of cool progress!

    • Ex: https://projects.sfchronicle.com/2019/kincade-fire-origin/

    • geojupyter/jupytergis#994

    • Can set landmarks, attach text, advance through the landmarks, configure transitions (maybe more!)

    • Currently, the story text is in the side panel

    • Do we need a kernel for story maps? Maybe not for the first iteration on this feature. Enables embedding in static sites! But later, with a kernel, we could have cool features like widgets in the story!

      • Interactive story maps could enable new playful/sandbox experiences that are powerful persuasion or decision-making tools

  • Data model binding to the renderer

    • It would be nice to support more renderers for situations when OL doesn’t do what we want (e.g. large geoparquet dataset!)

    • This could be confusing to users. But maybe we have a default renderer and offer the ability to swap it out with e.g. pip install jupytergis-deckgl?

    • Our data model needs some work to not be coupled with openlayers concepts, and instead model for simplicity and use adapters between our data model and renderers.

  • STAC:

    • Greg working on adding Copernicus catalog

    • Interns working on generic STAC interface. Should have progress by Feb 15, but it may not be fully releasable at that time (?); more of a base for some finishing touches like additional filters to help users navigate large catalogs.

  • Strategy:

    • How do we support making the architecture more robust?

    • Path to more funding and capacity: More people and orgs using JGIS as a daily driver, a critical dependency

    • What’s stopping people from using JupyterGIS as a daily driver?

      • Bugs or features?

    • Path to getting new daily driver users?

      • Develop a single use case “happy path” and make it very robust, including necessary architecture improvements. E.g. tight integration with a Notebook workflow using Xarray and GeoDataFrame datasets.

      • Estimate number of developer hours required with team

      • Submit for funding

      • As more daily driver users onboard, more willingness to fund additional features and robustness will come.

    • Matt: Discuss with Sylvain!

Pushed to next meeting¶