Tracking COVID 19 with Ruby
summary
A tiny Ruby stack scrapes ArcGIS feature layers and parses JSON into CSVs, caches results in Redis, offloads heavy caseline tabulation to Sidekiq, and reports timeouts to BugSnag; a helper app called paperboy queries the News API and a simple Rack app named ovid serves aggregated tables from those endpoints on Heroku. Pagination limits (200 / 2000 items) force incremental fetches; the project finds 2,000+ Florida cases, shows 0 of 67 counties meeting reopening criteria, and helps unlock hospitalization data.
AI-written summary · about
chapters (27)
- 0:00Intro
- 0:27Frank
- 0:57January 2020
- 1:09How to Prevent an Outbreak
- 1:42Is it spreading?
- 3:22The COVID Tracking Project AT The Atlantic
- 5:58undistinguishable from magic
- 6:33hunting for good data
- 10:46scaling
- 11:12pagination
- 11:43threads
- 12:52intuition
- 13:28Rebekah Jones
- 15:03tipping point
- 15:13researchers
- 15:26excess mortality
- 15:42local journalism
- 16:33data issues
- 17:03reopening criteria
- 18:56Takeaways
- 19:18Pay for News
- 19:56Vote for Science
- 20:06Promote Healthcare for All
- 20:18Fight Systemic Racism
- 20:34End Private Prisons
- 20:56Contribute your Time & Money
- 21:11Volunteer
related talks
42mVisualizing COVID-19 Data
31mConfessions of a Data Junkie: Fetching, Parsing, and Visualization
31m'Good Luck With That' : Tag Teaming Civic Data
20mJournalism with RStudio, R, and the tidyverse
35mVisualising Public Service: Seeing Inside Birmingham City
45mRuby Prototyping for Impact
15mCivic Coding In Ruby
29mKEYNOTE: Big Opportunities in Small Data