Qbrt Bytecode - Interface Between Code and Execution

Explore a conference talk on Qbrt, a bytecode assembly language designed to bridge the gap between code and execution. Delve into the world of high-level programming language design as Matthew Graham introduces Qbrt's built-in primitives for concurrency and inline asynchronous I/O. Discover how Qbrt aims to simplify the implementation of complex runtime features, allowing language designers to focus on creating user-friendly programming interfaces. Learn about the narrowing performance gap between JIT-compiled bytecode and native machine code, and understand why bytecode VMs are becoming increasingly appealing for front-end languages. Gain insights into Qbrt's standalone bytecode language and VM, which provides accessible implementations of inline asynchronous I/O, concurrency primitives, multiple dispatch, and pattern matching primitives. Follow along as Graham demonstrates Qbrt's capabilities through examples, including Hello World programs, worker processes, message passing, and multiple dispatch scenarios. Examine the advantages of Qbrt's approach to error handling and stack traces, and explore its potential applications in the evolving landscape of programming language design.

Intro
The Preview
Architectural Fit
Interoperability
Applications
VM Priorities
Language Priorities
Precedents
Obligatory Hello World
Hello UTF-8 8
Workers
Processes
Code to Create a Process
Code to Send a Message
Forks
Code to Fork
Why Exceptions Suck
Why Error Values Suck
Qbrt Stack Trace
Multiple Dispatch: Protocol
Multiple Dispatch: Int, String
Multiple Dispatch: Test Code
Multiple Dispatch: Output