Decide the structure of the storage of markdown files

[shapeseas]

Story Researchers will need to create markdown files -- whether typing them out or using the interface -- to design their deliberation experiments. These files needs to be stored.

Description Storage of markdown files. Researchers will need to be able to edit treatment file and create/edit/delete any files that the main treatment file references.

Potential approaches In discussion with @JamesPHoughton: 1) Store files on server 2) Store on browser as a serialized string, using localStorage

Additional approach: 3) Store on browser in db, using IndexedDB (or wrapper like idb, PouchDB)

IndexedDB seems like a strong candidate because it allows for secure, persistent storage that allows storage of documents, instead of just strings. It enforces the same-origin policy, uses transactions, handles asynchronous requests, and facilitates queries in a NoSQL style. It would facilitate offline usage of the application, which may be a bonus feature for researchers who are on flights or somewhere without wifi.

[shapeseas]

Additional notes:

• This video series is a solid intro to IndexedDB.
• Overview from Chrome Devs on storage on the web: https://www.youtube.com/watch?v=NNuTV-gjlZQ
• An additional wrapper is DexieJS.

[JamesPHoughton]

• what would the path be to port what we have in nextjs to a vs code extension - then we can depend just on local files using vscode, and use github more directly for managing files, etc.

[JamesPHoughton]

Possibly useful:

• https://code.visualstudio.com/api/extension-guides/webview
• https://github.com/microsoft/vscode-webview-ui-toolkit-samples/tree/main/frameworks/hello-world-react-cra