Spies and Secret Codes

Elementary School Workshop
OSSO Collective
May 23 — 24, 2024

Before establishing λLang. as an R&D platform, I was invited to give a workshop at OSSO aimed at elementary school students. Drawing from my research interests, the educational activities I had been developing with the OSSO collective and the goals outlined for Portuguese elementary school curricula, I designed and conducted a series of two one-day workshops entitled "Spies and Secret Codes", focusing on the contextual and, thus, constructional nature of language.



The workshops were rooted in the theme of “espionage” and the “transmission of secret messages,” with the underlying aim of exploring, through games and play, the constructive and contingent nature of systems of meaning and representation. The activities sought to expand upon the collective mechanisms of commitment, negotiation, and inference that form the broader practice of language, emphasizing the foundational importance of context, domain, and data models. Inspired by Goodman and Dewey, the focus was shifted from asking, "What does it mean?" to "When does it mean?".



Although the term 'code' was used in the title to create an engaging and thrilling premise for the participants, the workshop dealt exclusively with ciphers, focusing on syntactic games. The workshop was structured as a treasure hunt, where each stage contained an enigma that revealed the path to the next stage, enciphered using a historical protocol. These protocols — Alphabetical Index, Scytale, Atbash, Masonic Cipher, Caesar’s Cipher, and Morse Code — were introduced and explained in general terms before participants attempted to solve the cipher independently.



The workshop dossier and materials are available for download [here].


A Skeptical Techno-Logical Introduction to AI

Graduate Workshop
School of Arts — Portuguese Catholic University
June 20 — 21, 2023

In 2023, responding to requests from former MA students and PhD candidates at the School of Arts — UCP, I designed and conducted a two-day workshop delving into the techno-logical structures underlying artificial intelligence. The workshop followed a progressive trajectory, beginning with the foundational arithmetical concept of equivalence and culminating in the linear algebra mechanisms that underpin multilayer perceptrons.

Designed to be accessible to participants without prior knowledge of mathematics, the workshop began with a simple yet profound question: what does it mean to say 'x = y'? This exploration of identity was not only an appropriate and fruitful starting point for art students but also encapsulated the core conceptual program of artificial intelligence. From this foundation, we moved progressively through fundamental notions such as functions, recursion, regular languages, and models. The first day concluded with hands-on exercises involving finite state automata and Markov chains.



The second day was dedicated to deeper analysis and experimentation with single and multilayer perceptrons. Using diagrams and whiteboard calculations, we tackled key concepts in artificial intelligence, including activation functions, matrix multiplication, and gradient descent. The designs were then implemented and tested using simple Java programs, providing participants with a practical framework for understanding these foundational ideas.


The workshop’s bibliography can be found [here].