[ UK Smalltalk User Group ]

 Coypu and Phausto are two Pharo packages offering respectively a DSL and an API that turn the Pharo IDE into a music and sound design environment. They enable on-the-fly music composition, pattern sequencing, and DSP (Digital Signal Processing) programming. Born as a solo project and free, open source alternative to Symbolic Sound Kyma, they have been subsequently funded by the Pharo Association and Inria. Coypu , deeply inspired by Tidal Cycles, handles musical pattern creation and playback across different audio servers. Phausto  provides an interface for programming synthesizers and audio processing via an embedded Faust compiler, with Bloc widgets that make it easy to display and control synthesis parameters. Phausto can also be used to develop audio plugins thanks to its JUCE and Cmajor exporters. Live performances with both tools have demonstrated that Pharo can handle real-time music and sound design reliably, with solid timing and no audio glitches. Domenico Cipriani ...

Diálogo is a desktop tool that lets children aged 10–17 create their own videogames by combining drawn characters with a set of visual icons — no typing required. The icons compose like cards, activating behaviours in the drawn objects and naturally leading kids through concepts such as categories, rules, recursion, and metaprogramming. In this talk I'll show how Smalltalk's live, image-based environment made it uniquely suited for building Diálogo. I'll give a demo of the app, walk through some implementation metrics, and discuss the pedagogical ideas behind the project — including a free 7-class course I've shared online and workshops I've run open to the community at the Faculty of Exact Sciences (UBA). Agustín Martínez is a developer and researcher at the Universidad de Buenos Aires (UBA), and the creator of Diálogo, a programming environment for children built in Cuis Smalltalk. He presented a paper on Diálogo at the Onward! track of OOPSLA, and has given publ...

Smalltalk environments have long been pioneers in developer productivity. With the VAST Platform AI Assistant, Instantiations is extending that tradition by integrating Large Language Models (LLMs) directly into the VAST platform. This new tool is designed to bring AI capabilities directly into your daily workflow. In this session, Johan, Kris, and Mariano will share a user experience report and a live demonstration of the assistant’s current capabilities, powered by Google’s Gemini. You’ll see how it goes beyond simple chat by directly interacting with the live VAST image through the LLM’s “tools” capabilities. This allows investigating source code as it responds to your questions. These examples will demonstrate how the VAST Platform AI Assistant leverages the live nature of Smalltalk to act as a true pair programmer. We’ll also look ahead at our enterprise-focused roadmap, including support for local models to ensure privacy and security, integration with additional LLM providers, a...

Introducing VEO: A Live Visual Smalltalk that reimagines what a Smalltalk programming environment can be. VEO enables direct visual manipulation of any object and the creation of windowless visual interfaces. It is built around four core principles: all objects are visual; a single infinite zoomable and pannable canvas serves as the workspace; code can be written "in the air" anywhere in that space to send, evaluate and animate messages and live expressions; and live expressions can freely combine symbols, literals and live object references. Adrian Soma began his professional programming career in 1987, with a longstanding focus on the fundamental challenges of programming itself — particularly user interface design and interaction. He first encountered Smalltalk in 1996 and has worked with it ever since. An early project that would prove formative was an interactive chess course application, for which he built a custom graphical object system from scratch. That system becam...

 This talk explores what it means to work with AI from inside a live Pharo system rather than treating AI as an external chatbot. I will demonstrate experiments where AI models are invoked directly from the Playground, including a concrete implementation of conversation history (AIAHistory) and experiments across multiple models and providers. Rather than focusing on code generation, the emphasis is on workflows: how different conversation structures affect results, how styles and constraints can be imposed on generated comments, and how AI systems can be inspected for systematic errors and limitations. Examples include documentation support, UML generation (via PlantUML), and experiments in automated paper review. The talk is experimental in nature and rooted in Smalltalk’s strengths: reflection, live objects, and tools that are part of the system rather than bolted on. The goal is not to present a finished framework, but to share concrete insights, failures, and possibilities for...