Skip to content

Stories for the MVPoC

Jump to bottom
kitblake edited this page Jul 15, 2020 · 2 revisions

This is a minimal version of the Stakeholder Stories, abridged for the PoC. Most aspects described here are meant to be implemented in the PoC. Items in [brackets] could be triaged.

There are two users, Sam the Civil Servant and Reilly the Researcher, plus the actions of the platform.

Country Civil Servant Reporting

Sam's task is to submit the latest NDCs for his country. He has one file to upload, a PDF.

  1. Using his browser, Sam accesses his country's submission page on the ClimateDataPool server.
  2. The submission page contains a form. Many of the fields are pre-filled. Sam fills in a few required fields with corresponding metadata.
  3. Sam adds the file to the form and clicks the upload button.
  4. The server confrims the succesful submission. At this point Sam is done with reporting.

Server Script

When the file arrives at the server it gets processed by a script. We assume that there is an IPFS instance running on the machine. In short, the script:

  • [Checks if it's a zip file and if so unpacks the archive.]
  • Copies each file to IPFS.
  • For each file it creates a record in the OrbitDB which contains the IPFS hash and/or path, plus basic metadata (details TBD, but e.g. it contains the country name).
  • For each file copied to IPFS, the script uses the Pinata API to pin the file. Perhaps this can be unified in one operation.

ClimateDataPool User Interface

The intention is for each country to have an environment on the ClimateDataPool server, but for the PoC we'll have one example country with a minimal environment.

The primary function of the UI is to provide listings of files. For example, Sam can view a listing of all the files submitted by his country.

A listing contains various metadata such as submission date, file size, etc (TBD).

The listings are generated using queries to OrbitDB. Each listing is the result of a search with parameters. For the PoC all the view actions use predetermined (hard coded) queries. There is no search form interface.

When rendered for the public each record in a listing contains a "Download" button.

Scientific Researcher Searching

A researcher can access the ClimateDataPool UI using a browser. For this story we're presuming that Reilly the Researcher is collecting files for a project.

  1. Using her browser, Reilly accesses the ClimateDataPool main page.
  2. The main page presents links to various listings. These might include latest submissions, submissions by country, submissions by type, etc.
  3. Reilly clicks to a listing for a country.
  4. The server returns a listing. The latest files are shown, ordered reverse chronologically. Much like the internal view, each file is a record with corresponding metadata, including a "Download" button.
  5. Reilly reviews the listing, locates the file she wants, and clicks the "Download" button.
  6. The server sends the file to Reilly. At this point her task is complete.
Clone this wiki locally