Arty Starr
Lone Star Software Symposium: Austin
Austin · July 27 - 29, 2018
Developer Experience Researcher, University of Victoria
Arty Starr has been a software engineer for 20+ years and is a researcher and pioneer of new theories of the developer's cognitive experience that help us recognize how developers' experiences are changing with AI. Her theories of developer momentum and troubleshooting help us see software development through an ecological lens, recognizing it as a living activity in motion and the living capacities of care and cognition that support this motion. Her aim is to bring heart, human-centric values, and agency to the craft of software development so we can build toward the type of world we want to live in. With richer perception and by recognizing what matters, we can light a path to a future of thriving habitats where humans and AI can collaborate well and thrive together.
Arty is the author of Idea Flow, a speaker, thought leader, and artist who loves mixing ideas across disciplines into a fusion of creativity to help make the world a better place through her projects.
Presentations
Engineering Feedback Loops
Science is a systematic process of engineering feedback loops that help you learn. This learning isn't arbitrary. In science, we're on a mission to discover useful clues that help us understand a system. Whether that system is a product we're trying to design, a process we want to improve, or an application we're trying to understand – the art of engineering feedback loops still applies.
Test Harnessing Legacy Infrastructure
In the world of legacy code, we often end up inheriting a tangled ball of mess with a lack of automation, and no clear surfaces for testing. Yet still, under these circumstances, we're expected to safely make changes without regressions. Where do we start? How do we tackle this challenge? How do we get a handle on re-architecture?
We'll start this discussion with a first-hand use case and example – tackling the re-architecture of an 800k line JBoss application with near-zero unit tests. Ugh. The only option on the table was Selenium. UGH.
Let's talk about alternative strategies. How have you tackled similar situations? How could we build a data-driven regression framework without going through the UI?
Test Automation Disasters
How much time have you spent automating tests, then discovered the tests were mostly useless, brittle, time-consuming, and hard to understand? How many times have you deleted everything, started over, and came with a substantially better strategy for next time?
In this session, we'll dig into one disaster story at a time, explore the nuts & bolts, then distill lessons learned and improvement ideas.
Putting Out Fires with Gasoline
Once upon a time, it was just me and my app – the days when all I had to know was “get data, put on screen.” Fast forward ten years later, and what the hell happened? The level of complexity that we deal with in modern software development is insane.
Are we really better off than we were 10 years ago, or have we just been putting out our fires with gasoline?
Theory of Reality Construction
Ready Player One.
Welcome to the future of Augmented Reality. All physical limitations of what can exist around you, have now been lifted. In the world of AR, we can visualize anything we can dream… so dare to dream big.
In this session, we'll break down the Matrix into a buildable software architecture model – the algorithm of Life() expressed as a probabilistic Magnet Operating System. Next, we become Software Shaman responsible for shaping the metaphors and schemas that define the Oasis… welcome to the world of Reality Construction.