A Brief, Incomplete, and Mostly Wrong History of Serverless

Michael Hausenblas
Apr 28 · 1 min read

1993A bunch of folks at the National Center for Supercomputing Applications (NCSA) write the specification for calling command line executables from Web servers. This evolves into CGI or the Common Gateway Interface, which — if you’re a fancy-pants — would call by its royal Internet name RFC 3875.

1996 The Standardization community strikes again: SQL/PSM or Persistent Stored Modules is published as an extension of SQL-92. With this, we can not only make our Web servers less secure and slower, but now also relational databases.

2008 Someone at Google asks themselves what would happen if one could do CGI using Java and Node and decides to give it a try: GAE or Google App Engine is born. Most people ignore the suggested compute paradigm and focus on trying to find ways to make App Engine execute long-running tasks.

2014 AWS launches Lambda, an event-driven, serverless computing platform (without calling it serverless). Initially, people try to ignore it since at around the same time the two new hotnesses Docker and Kubernetes take off. However, eventually, due to the Trinity of Simon Wardley, Corey Quinn, and Paul Johnston the serverless, erm, Function-as-a-Service paradigm wins the day.

Yeah, so, the inspiration for this post comes mainly from this episode of the Screaming in the Cloud podcast. Further, the title has been lifted from the James Iry classic A Brief, Incomplete, and Mostly Wrong History of Programming Languages. You’re welcome.

Michael Hausenblas

Written by

Developer Advocate at AWS | Gopher | Cloud Native Ambassador | Kubernetes | containers