jsvine
commented
1 year ago Alrighty! I think I have a basic version of this working:
- Viewer/renderer (public): https://data-liberation-project.github.io/epa-rmp-viewer/
- Converted JSON files (~just a sample for now, will upload the full set of them soon~ now all uploaded) that you can load into the viewer — filenames are
{FacilityID}.json, whereFacilityIDis that misnomer discussed elsewhere and is actually a submission ID: https://drive.google.com/drive/folders/12UQJVfdPdNLhVM-wbI2lESrVSAAnCrw7 - Code (public): https://github.com/data-liberation-project/epa-rmp-viewer
It's all quite bare-bones for now. For instance:
- It focuses on just the tables of greatest interest (submissions, processes, chemical processes, accidents, accident chemicals, and executive summaries), but shouldn't be too hard to add the others.
- The viewer is ugly, though it does the job and can be prettified in the future.
Here's a preview:
[...]
Feedback and contributions very much welcome. (Especially if you're an experienced Svelte developer, which I am not.)
As I've been scoping out the spreadsheets for #4, I'm realizing that (a) they're useful for getting a broad overview, but (b) aren't a full substitute for the more detailed/granular information in each RMP.
For instance: All the details about each facility "process" seems like too much to add to the core spreadsheets ... but as soon as you know there was an accident at a facility, you'll want to be able to drill down on that process.
Until we've built a full-blown interactive website, I think a decent middle-ground might be:
Generate a JSON file for each RMP submission, containing all database information corresponding to that submission.
Build a bare-bones static site that lets you select a specific submission ID and see a verrrry-lightly-styled rendering of that submission's full data.
I'll start working on pt. 1, will update with progress here.