GitLab

The Overviews are one of the most useful features, but it is difficult to keep track of all the information available (long, long scroll).

I would like to try another way of exploring accession data by using a drill-down method and pie charts with onion layers:

These onion layers work well when there are no overlaps in data (e.g. storage can have multiple values and affects the count!).

Querying path

For Genesys, there are many querying paths that make sense:

1. Genus
2. Species
3. Provenance
4. Country of holding institute
5. Holding institute

Or perhaps:

1. Country of holding institute
2. Genus
3. Holding institute
4. Species

Or:

1. Availability
2. Genus
3. Holding institute
4. MLS status

User-defined layers

We declare the default order of layers for all categories we support (in ES): genus, species, institute.country, institute, provenance, ...

But the user may have a different preference and would like the onion layers ordered differently. We need to allow the user to reorder the layers. A simple drag-drop list will do.

Drill-down

We render the onion pie chart based (on ES data) for the categories. By clicking any of the sections all relevant filters are applied and the selected section becomes the root of the onion chart.

Problems

The first problem we have is that we may get too many values for genus (or another field) to properly render all of them and the smaller chunks. We need to select the number of "Hits" to display and use the "Other" in a clever way. Then by clicking "Other", we need to exclude the other displayed values in that category and re-render the chart with the same root.

API

We don't want to make many calls to the server to retrieve the onion layer data. The server needs to return group by aggregation for the next 2 or 3 fields (for 2 or 3 layers) for the applied filters. This is currently not implemented.