In the process, add docs: write up how to export data from BLS page to JSON via Google Sheets.

Under "Have the Talk," address no pay gaps talking point with actual data.

As well as occupations without enough data to derive one.

Update: instead of adding one-page gaps for these occupations, just add copy about the total occupations without enough data to measure, and total where women make more than men (3 or 4 in 2017)

Right now AllOccupations includes a ul for every li. The current code is hard to reason about, because the GapAPI data structure doesn't lend itself to hierarchical rendering.

To do: add a function to GapAPI which creates a nested data structure that will make ul/li rendering easier and more obvious.

Possibilities:

https://github.com/bvaughn/js-search

http://fusejs.io

https://github.com/moroshko/react-autosuggest

To reproduce: go to /all page, scroll down, click on an occupation low on the page. When you arrive at that occupation's page, you are scrolled down there, too.

• Set up new domain on Netlify
• Set up redirect to new domain in Hover
• Update all code references from "gapp" to "gap" and ".com" to ".co"

Using react-helmet

http://www.bls.gov/oes/#data

Break out footer as a reusable component on Gap page, All page, 404 page.

Manual data entry, plus re-export required.

Currently yarn run build completes with the error 🔥 RangeError: Maximum call stack size exceeded.

via Google Sheets

Something like:
https://github.com/matthewmueller/x-ray

Could parse the BLS page and generate the JSON structure we want without having to go through all the manual Google Sheet madness. This may or may not be worth doing for an annual data cleaning task.

Get rid of all inline CSS

This will minimize the file size significantly.