When is software finished? When it passes a test? When it's deployed? Trick question. "Finished" is not really something software can be, because we're always updating, refining, re-releasing, or bug-fixing. Even EOL software sometimes gets security updates. In this talk, I'm going to discuss the opportunity presented by the unfinished nature of everything we make, and how we can use observability, metrics, and conversations to break down the silo between software and… people. Join me for a vision of how we can add value with tools that we already have, and recognize the value of things that are not the new hotness.
Scientific management says that there is a right way to do things, a best way, and that the way to make workers more efficient and standard is to enforce that one way. Large Language Models take the texts they are fed and predict the most likely thing we want next. While both scientific management and LLMs are a revolution in how work gets done, they are also a tribute to homogeneity, mediocrity, and norming. Join me for a far-ranging exploration of industrial design, standardization, and why normal and average are sometimes a deadly myth.
Heidi is an advocate for progressive delivery, organizational transformation, technical communication, and marketing you hate slightly less. She is a nerd about industrial psychology and patterns of progress. One of her favorite hobbies is talking to developers about things they already knew but had never thought of that way before. She sews all her conference dresses so that she's sure there is a pocket for the mic.