From Sumatra to Panama, from Babylon to Valhalla
summary
Argues Rice’s theorem forces a mix of static and runtime checks, shows Loom’s virtual threads ship with a calculated technical debt (synchronization pinning) and that prototypes get thrown away: Project Sumatra demos GPU offload (started ~2011) but isn’t scalable so it spawns Panama, vector API and code reflection to handle native interop and lambda-to-GPU transpile; Valhalla runs 5–10 prototypes, yields value classes and no wholesale recompilation.
AI-written summary · about
chapters (14)
- 0:00Intro
- 1:54Obvious in Hindsight
- 9:01Technical Debt
- 13:26Embracing the Past and the Future
- 21:08From Sumatra to Panama and Babylon
- 34:11Memorabilia I
- 35:15Java Versions
- 40:15Turing Machines
- 41:49Memorabilia II
- 46:50Dynamic vs Static vs Rice
- 52:25Memorabilia III
- 54:44Peak of Complexity
- 58:58Memorabilia IV
- 1:02:05Outro
related talks
51mYour next JVM: Panama, Valhalla, Metropolis
45mBeyond the Hype: Unpacking the Latest Features of JDK
46mJava Next - From Amber to Loom, from Panama to Valhalla
57mJava Next - From Amber to Valhalla, from Loom to Leyden, from Babylon to Panama
49mJVM in the Age of AI: Babylon, Valhalla, TornadoVM and friends
47mJVM in the Age of AI: Babylon, Valhalla, TornadoVM and friends
45mJava Next - from Amber to Loom, from Panama to Valhalla
24mPostcards from the Peak of Complexity