James Gosling

@dougmerritt@mathstodon.xyz @kentpitman@climatejustice.social @screwlisp@gamerplus.org @cdegroot@mstdn.ca @ramin_hal9001@fe.disroot.org The GC I wrote back in the early i90s is (thankfully) long gone. Many Actual Experts in garbage collection joined the project, with many backgrounds. Lisp and Smalltalk were common. I am completely in awe of today's Java garbage collectors.

@kentpitman@climatejustice.social @screwlisp@gamerplus.org @cdegroot@mstdn.ca @ramin_hal9001@fe.disroot.org Your understanding is mostly faulty. The original GC was written by me, and I'm no Lisp GC expert. I was (and still am) an admirer of Lisp. I wrote the code for my whole PhD thesis in Lisp. My admiration for garbage collection started earlier, when I was a big user of Simula in the 70s. But the motivation for GC in Java was different: the motivation was all about reliability and security. A leading cause of security vulnerabilities as always been buggy code. And one of the leading root causes of many long standing, hard to diagnose and fix, bugs has been flakey storage management. Garbage collection goes a long way to increasing system reliability, and hence security. I had always wanted to make GC more mainstream. When you described garbage collection to senior management back in the day, their reflexive judgement was: "bullshit! Lazy engineers just don't want to clean up their mess". But when they see measureable improvements in system robustness, and corresponding decreases in failures, they Notice.