Objective JH-6: Improve map visualization in notebooks

NASA-IMPACT / veda-jupyterhub

VEDA JupyterHub technical planning and documentation

1 stars 1 forks source link

Objective JH-6: Improve map visualization in notebooks #21

Closed batpad closed 4 months ago

batpad commented 7 months ago

Motivation

Current map visualization tools in Jupyter notebooks are well behind the state-of-the-art of modern browser based map tools
Visualizing on the map inside Jupyter is still an extremely fragmented experience with no clear way of doing things across use-cases
Still hard to combine raster and vector data seamlessly alongwith analysis
Hard limits on size of datasets for interactive map visualizations

Owner(s)

@batpad @wildintellect @kylebarron

Success criteria

[ ] At least one demo of using lonboard / next-generation WASM + GPU-based mapping clients with NASA data
[x] A roadmap to collaborate with the larger community on supporting modern cloud-native data formats and improving GIS tooling within Jupyter notebooks

wildintellect commented 7 months ago

Need a better goal than that, or more specifics, we already have Lonboard examples in MAAP and VEDA https://github.com/MAAP-Project/maap-documentation-examples/blob/main/gedi-subset/notebooks/lonboard_exploration.ipynb

batpad commented 4 months ago

Lonboard continued to get more usage, and we were able to further support of sharing of notebooks that used lonboard on https://notebooksharing.space

@wildintellect would it be possible to compile some examples or even just make some brief notes on how lonboard has been especially useful - it seems like there were a few really good examples in the past few months of folks using lonboard for things that would not have been possible without it.

We've had some requests for features to add to lonboard, and we plan on spending some time in the coming quarter to add frontend features to improve usability.

batpad commented 4 months ago

On the Jupyter GIS front, we have continued to stay involved in discussions and being part of community meetings. It's great news that QuantStack is taking the lead on developing a framework for collaborative GIS tooling within Jupyter. We will continue being involved, contribute where we can, and be well setup to deploy these tools to our users as they become available.

With the focus of the Jupyter GIS project being on collaborative editing and providing more "desktop-app" like experiences, we feel our work on lonboard dealing with rendering large-scale vector data and optimizing for cloud-native formats continues to be well aligned and complimentary work.

batpad commented 4 months ago

Some good examples of using lonboard in this past quarter for real-world usage + notebooksharing.space to share can be found in the examples in this repo: https://github.com/MAAP-Project/geotrees-demo/