Anthropology of Ruby: Excavating its Gems
summary
A GitHub project called gemology mirrors Rubygems with ruby gems mirror, subscribes to gemwebhook feeds, and uses Sinatra + Resque to fetch from CloudFront into Rackspace Cloud Files, then extracts metadata (metadata.gz YAML and data.tar.gz) from .gem files. Analysis finds 50,000+ gem releases since gemcutter, a current rate ~150–200 releases/day, active_support as the top runtime dependency, many gems missing license fields, encoding and packaging errors, 4,460 gems declare extensions while only 2,333 target native platforms, and a clear need for a gem lint tool.
AI-written summary · about
chapters (19)
- 0:00Intro
- 0:20The Anthropology of Ruby
- 2:12Jim Webb Book
- 7:30Release Date Graph
- 11:40Ease of Publishing
- 16:50Ruby June Versions
- 18:25Prereleases
- 18:37Metadata
- 19:53Extensions
- 20:10How to determine extensions
- 21:50Platform Gems
- 24:34Ruby Language
- 24:50Required Ruby Version
- 27:29Twitter Lock
- 29:54Runtime dependency graph
- 31:29Development dependency graph
- 31:58How dependencies work
- 35:10What we learned
- 36:53The link command
related talks
24mExploring Big Data with rubygems.org Download Data
31mConfessions of a Data Junkie: Fetching, Parsing, and Visualization
35mRegulations, Ethics And Messy Data: Tales From An Open Source Analyst
42mMarket Analytics in EVE with Ruby
36mData-Driven Government and the Ruby Developer
27mCommunity Data is Not Community Metrics
01h 04mWhat can we learn from 750 billion GitHub events and 42 TB of code
21mTracking COVID 19 with Ruby