What Go Got Right (and Wrong): Lessons from Rob Pike’s 14‑Year Anniversary Talk

At GopherCon AU in Sydney, Go co‑creator Rob Pike delivered a retrospective talk titled “What We Got Right, What We Got Wrong” to mark the 14th anniversary of the Go programming language. He shared personal memories of Go’s early history, offered insights into key design decisions, and clarified that the views expressed were his own, not official statements from the Go team or Google.

“Go has always been the result of a dedicated team and a massive community’s effort. If you agree with anything I say, thank them. If you disagree, blame me, but don’t say it.”

Pike emphasized that the quality of a programming language is largely subjective. In a 2022 article co‑authored with Ken Thompson, Russ Cox, Robert Griesemer, and Ian Lance Taylor, he noted that Go was deliberately built for concurrency and parallelism, leveraging modern multicore chips, and that its sustained “developer‑centric philosophy” and vibrant community were crucial to its success.

He recalled that the original goal was not to create a new language but to devise a better way to write software. He credited the Go gopher mascot—released under a CC license—as an early and unexpected factor that helped unite the community and foster a culture of technical excellence mixed with fun.

Pike listed several concrete decisions that contributed to Go’s growth:

The gofmt automatic formatting tool.

A rich standard library of packages.

The formal Go language specification.

Early compatibility guarantees for the language.

He also expressed wishes for future features, such as arbitrary‑precision integers to eliminate a class of security risks, and more compiler checks for dynamic interfaces and resource‑sharing deadlocks, stating that anything that makes programs safer at compile time is valuable.

Pike highlighted Go’s portability, noting that its ability to compile for many platforms stems partly from Ken Thompson’s original C implementation of the compiler, even though some have argued the compiler should be written in Go itself or built with LLVM tools.

The talk also covered the pivotal decision to prioritize concurrency. Pike described Google’s early reluctance to allow multithreading, which motivated the design of Go’s lightweight goroutine model. He argued that Go helped convince the programming world that concurrency is a powerful tool, especially in the multicore era, and that this focus was a major driver of early adoption.

In closing, Pike credited the supportive, diverse Gopher community as the most important factor behind Go’s success, summarizing that the language’s mission—to make high‑quality software development simpler and more efficient—remains unchanged.