Metadata Updating with Miriam

[jacobdadams]

Benefit

We're using a new metadata schema stored as .md files in github to rewrite and unify metadata based on all the current versions/sources.

This supports the MATT project by creating a One True Version of metadata for at least our most used layers, and a guideline/process for the layers Miriam isn't able to finish.

Acceptance Criteria

• An updated metadata schema.
• As many layers rewritten as possible.
• A new style guide based on Miriam's experience with all our various written work.

Notes

No response

Risks

• Getting bogged down with one or two big ones
• Poor management
• Lost or unknown metadata (where did this layer come from anyways?)
• Perfection vs good enough

Issue Reference

• Update UGRC style sheet gis.utah.gov#2763

[jacobdadams]

Priority list based on this sheet from 2022/23: https://docs.google.com/spreadsheets/d/1yzsY3udaIRH9HtQvXxgFHtr4nolk8xdpJiDwVrCnnPk

[steveoh]

• The initial list has been written and worked on
• Miriam has an export from agol of item views to help create the next list. Steve will provide google analytics and api usage analytics to help shape new list
• Review the new list

[jacobdadams]

Carrying into Q2

[jacobdadams]

Carrying into Q3. In Q2, work slowed down due to Miriam taking on other tasks (blog posts, some data stuff, etc) and being back in school.

[miriamseely]

Current Status of the MATT Repository:

Layers merged, currently in pull requests, or will be in pull requests by the end of the week: 140
Layers that have been skipped so far: 20
Total layers in the SGID meta table as of August 2024: 379
So, 140 layers out of 359 are complete. That is about 38% of the layers.

Some Caveats:

If we are only creating metadata for layers we host, which has historically been part of the methodology, then there are about 329 total layers, and 140 out of 329 is 42% of the layers.

If we continue to skip the Lidar layers, then there 140/300 are complete. (46%)

Thus we see that how "complete" the MATT project is depends upon which layers will receive new metadata. Of layers we host, all layers with greater than 9000 views have been completed. The view count is total views since the publication of the layer.

What's Next

Going forward, there are a few factors to consider with determining when the project will be "done". Once decisions have been made on which layers will receive fresh metadata (whether to add layers we do not host, or Lidar layers to MATT, the DEQ layers, etc) that will likely give us a better idea on how much more there is to do. Many of the remaining layers can have the same metadata (e.g. all the different years for the Tax Entities metadata can likely have almost identical metadata) so that will make things go faster.

Since all of our most used layers are already complete, I am now getting into the layers that folks don't know much about, or otherwise have limited or no documentation. This affects the time it takes for me to push the layer through my workflow.

I have a document that indicates which layers are coming down the pipeline next. I only have layers on it that have greater than 4000 views (on the Layers To Do Next tab) and have UGRC listed as the host. As I've mentioned, once we decide the exact conditions for a layer to receive metadata, I can update that document with the rest of the layers we are intending to add to MATT.

Hopefully this gives everyone a general idea of where we are at with the MATT project. Let me know what questions you have!