The Divine Software Engineering Comedy -- Inferno: The Okinawa Files

TL;DR

The paper offers a satirical, narrative critique of the future of software engineering as discussed at FUSE 2024 in Okinawa. It argues that AI-driven tools, rapid tech turnover, and evolving education and research practices threaten coherence in SE practice, calling for longer-term studies and closer industry–academia collaboration. The author outlines three intertwined challenges—unskilled but productive software makers, an unmanageable proliferation of technologies, and a crisis in how we learn and validate knowledge—while stressing the need for human-centric wisdom and responsible maintenance. The piece emphasizes that progress in SE must be grounded in understanding contexts, learning from failures, and building sustainable, accountable practice as the field diversifies into multiple disciplines.

Abstract

In June 2024 I co-organized the FUture of Software Engineering symposium in Okinawa, Japan. Me, Andrian Marcus, Takashi Kobayashi and Shinpei Hayashi were general chairs, Nicole Novielli, Kevin Moran, Yutaro Kashiwa and Masanari Kondo were program chairs, some members of my group, Carmen Armenti, Stefano Campanella, Roberto Minelli, were the tables, can't have a room with only chairs, after all. We invited a crowd of people to discuss what future software engineering has. FUSE became a 3-day marathon on whether there is actually a future at all for SE. This essay is a slightly dark take about what I saw at that event, very loosely based on the discussions that took place, adding some healthy sarcasm and cynicism, the intellectual salt and pepper I never seem to run out of. I listened to the brilliant people who gathered to talk about where we're headed, and distilled three nightmares headed in our direction: software makers who don't know what they're doing, but get the job done anyway, a field moving so fast it can't remember its own lessons, and technologies multiplying like rabbits in Spring. So, let's start. The future, eh? The future of software engineering looks like a car crash in slow motion: you can see it coming but you can't look away. The thing is...