Everything you ever wanted to know about Scala implicits, but weren't sure how to ask
summary
Empirical claim: implicits are common and mostly used safely — about 8.2 million lines of Scala yield ~12 million call sites and a pipeline that produces ~800 GB of analysis, showing roughly 16% implicit call sites overall and ~20% of implicit calls matching type‑class patterns. The study actually compiles repos, deduplicates forks, and resolves symbols with SemanticDB + Scalameta; many implicits come from the standard library and test libs, some cause long compile times, and the tooling mostly succeeds but skips projects that fail to build.
AI-written summary · about
chapters (14)
- 0:00Start
- 4:32Scala Implicits
- 8:45Implicit Conversion - Adaptation
- 9:51Implicit Conversion - Extension
- 14:32Implicit Parameters
- 15:12Use Cases
- 15:32Type Classes
- 22:00Evidence for properties (theorem proving)
- 23:33Establish Context
- 24:39Confusion?
- 33:11Semanticdb - Tree
- 37:51Pipeline - filter duplicates using déjàvu
- 43:25Use of Implicit Conversion
- 44:13Number of implicit conversion declarations
related talks
36mData pipelines engineering made simple with Scala
42mScala: The Unpredicted Lingua Franca for Data Science
01h 03mBig Data Pipelines With Scala - Part 1 of 2
32mWhy Use Scala in a Bigdata Context?
24mBig Data Science in Scala
23mAxle A Spire based DSL
15mApache Beam and Google DataFlow IRL with Scala Scio
01h 05mData Visualization with Scala tools