fuzheado / jupyterkg Goto Github PK

Provide a way to save or load the state of each facet

Too many facets and they start to disappear. Instead they should scroll.

Provide a list of sample queries people could load in from the UI

Ideally, you could specify a LIMIT to the number of nodes, but you'd want them randomly sampled. Here's a possible way:

https://stackoverflow.com/questions/5677340/how-to-select-random-dbpedia-nodes-from-sparql

Put a way to specify the number of starter facet panels in the YAML config file

Right now it's a statically sized iframe but instead it should be responsive to the size of the browser window. Need to fiddle with styles of a parent DIV in Jupyter for this to work.

Could you add a license file?

This could be either in the config file or with a color picker in the interface

Right now for speed and simplicity, a local CSV file (the result of a SPARQL query) is provided to fill in the parameters list. It would be nice to provide some alternatives that allow for some more dynamic loading of choices:

• From a URL - pull in a properly formatted (JSON?) stream
• From a SPARQL query
• From a wiki page - use a table from a wiki page

Parameter list could support different orderings - original list order from the CSV, alphabetically, or otherwise.

The entire configuration could be loaded from a wiki page (like Cradle), allowing for more collaborative ways to customize the use of the tool

Support property-qualifier or more specific ways to filter

Provide a way to return more than just a knowledge graph - perhaps a image grid or some other type of display