SFSCON 2026

sfscon-2026SFSCON 2026Digital SovereigntyEngineeringOpen HardwareLegalPolicy & ComplianceDigital Independence DayCommunityOpen AccessEducationLLMSide Event | FSFE Italian community meeting08:3000:30Check-in SFSCON 2026SFSCONSFSCONSFSCON0009:0000:15Opening SFSCON 2026SFSCON - Main trackSFSCON - Main trackSFSCON,Main track0009:3000:30Keynote speakerSFSCON - Main trackSFSCON - Main trackSFSCON,Main track0010:0000:15European SFS Award 2026Free Software Foundation Europe (FSFE) and Linux User Group Bozen-Bolzano-Bulsan assign the “European SFS Award”Free Software Foundation Europe (FSFE) and Linux User Group Bozen-Bolzano-Bulsan assign the “European SFS Award”The award is given to a person who contributed to the introduction of the culture of Free Software. SFSCON - Main trackSFSCON - Main trackSFSCON,Main track0010:1500:15SFS Award 2026Linux User Group Bozen-Bolzano-Bulsan assigns the “South Tyrol Free Software Award”Linux User Group Bozen-Bolzano-Bulsan assigns the “South Tyrol Free Software Award”Every year the Linux User Group Bozen-Bolzano-Bulsan assigns the “South Tyrol Free Software Award” (SFS Award). The award is given to a person who contributed to the introduction of the culture of Free Software in the Province of Bolzano. SFSCON - Main trackSFSCON - Main trackSFSCON,Main track0010:3000:30Keynote speakerSFSCON - Main trackSFSCON - Main trackSFSCON,Main track0011:0000:35Sovereign AI in Production: Blueprint for a Globally Distributed Dev Teamhttps://www.sfscon.it/talks/sovereign-ai-in-production-blueprint-for-a-globally-distributed-dev-team/How Linaro built and deployed a multi-tier, confidential, hybrid on-prem and cloud-hosted LLM infrastructure.How Linaro built and deployed a multi-tier, confidential, hybrid on-prem and cloud-hosted LLM infrastructure.Large Language Models (LLMs) offer unprecedented opportunities to accelerate software development, but for organizations handling sensitive code, public AI APIs are a compliance minefield. When data confidentiality and regional sovereignty are non-negotiable, how do you scale AI assistance to a globally distributed team? In this talk, we share Linaro’s ongoing journey of designing, deploying, and adopting a multi-tier, on-prem, and regional LLM cloud-node infrastructure. We will dive into the architectural compromises made, the open-source AI toolkits we leveraged, and how we balanced privacy with scalability. Beyond the tech stack, you will learn how we measure developer adoption, track success metrics, and draw real-world lessons from the transition from design to daily developer use. Attendees will walk away with a practical framework for building their own confidential AI infrastructure. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Davide Ricci1111:4000:15Open and free software meets C2C designhttps://www.sfscon.it/talks/open-and-free-software-meets-c2c-design/Electronics are among the fastest-growing waste streams worldwide. While much attention is paid to recycling and material efficiency, one critical prerequisite for truly circular electronics is often overlooked: openness. Products that cannot be repaired, updated, or maintained due to proprietary software, closed interfaces, or missing documentation will inevitably become waste, regardless of the quality of their physical materials. This session introduces the Cradle to Cradle Electronics Initiative (C2CEI) as a pioneering framework to solve this crisis . The speakers will outline how this pre-competitive network of industry leaders, NGOs, and researchers is actively working to establish new standards for circular hardware. Drawing on the C2CEI’s latest work frameworks and illustrating them with practical industry insights, such as the Teufel MYND speaker development. This talk demonstrates how software freedom directly extends product lifetimes, improves repairability, and enables continuous innovation across global value chains. Participants will gain firsthand insights into how the C2CEI serves as the primary bridge between digital sustainability and systemic circular design. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Tim Janßen1112:0000:15The Tech Sovereignty Cataloguehttps://www.sfscon.it/talks/the-tech-sovereignty-catalogue/Connecting Europe's Sovereign Tech EcosystemConnecting Europe's Sovereign Tech EcosystemEurope currently relies on foreign suppliers for over 80% of its digital technologies, with just three non-European hyperscalers controlling more than 70% of the cloud market. The Tech Sovereignty Catalogue is a concrete response to this structural dependency: a concrete initiative that maps the complete European technology stack, from hardware and connectivity through cloud infrastructure, cybersecurity, data platforms, and AI, turning the ambition of digital autonomy into a navigable, actionable ecosystem. This talk introduces the Catalogue, its current and upcoming developments, its governance model, and the sovereignty criteria developed by our consortium of partners to ensure only genuinely sovereign solutions are included. To be listed, a solution must meet criteria across three areas: corporate sovereignty (European headquarters, majority ownership and voting rights, and exclusive European jurisdiction), data sovereignty (storage, processing, and service execution within Europe by a provider that complies to corporate sovereignty), and full technological maturity and market deployment. Assessment, mapping, and listing of the vetted solutions are all provided free of charge to solution providers. The Catalogue is more than a directory; backed by a strong partnership of industry players, think tanks, academic institutions, and research institutes, it is actively fostering a community of like-minded European companies that are already collaborating, building together to expand their joint offering across Europe. We will also explore how Free and Open Source Software fits naturally within this framework. Solutions built on open standards and designed for interoperability align closely with the Catalogue’s vision of a collaborative, non-fragmented European digital ecosystem, making the case that sovereignty and openness are complementary rather than competing values. For SFSCON’s community of decision-makers and developers, the Catalogue represents a practical bridge between the ideals of software freedom and the policy reality of European digital independence. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Sebastiano Toffaletti1112:2000:35Panel Digital SovereigntyDigital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

0013:0000:15Governing for Autonomyhttps://www.sfscon.it/talks/governing-for-autonomy/The OSPO’s Role in Achieving Digital Sovereignty through FOSS AIThe OSPO’s Role in Achieving Digital Sovereignty through FOSS AIFollowing last year’s presentation about discovering the “AI jungle” and drafting initial risk frameworks, this year the focus will be on operational reality and the critical realization that relying entirely on proprietary AI black boxes threatens an organization’s digital sovereignty. To maintain control over their data, infrastructure, and technical destiny, organizations are increasingly turning to FOSS AI solutions. The OSPO is uniquely positioned to lead this transition, shifting from a traditional compliance office to a strategic enabler of digital autonomy. As EU AI Act compliance deadlines approach, the demand for transparency has intensified. This is where FOSS AI serves as a practical shield. By leveraging open-source technologies and the Open Source AI Definition, OSPOs can help organizations deploy models locally or in sovereign clouds. This approach avoids vendor lock-in, ensures data privacy, and retains full governance over the infrastructure. On the other hand, as AI systems evolve into “Agentic” mode of operation, securing this self-hosted environment becomes a complex and dynamic challenge. This presentation will explore how OSPOs can practically utilize FOSS to secure and assert digital sovereignty in an AI-driven world, including the operational trade-offs of hosting open models, how to build verifiable and independent AI stacks, and how to govern autonomous agents without compromising control over one’s data and infrastructure. Join this session to learn how to move your organization from dependency to digital independence. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Alexios Zavras1113:2000:15Choosing Ethical and European AIhttps://www.sfscon.it/talks/choosing-ethical-and-european-ai/Evaluating AI: the criteria for vendor selectionEvaluating AI: the criteria for vendor selectionThe hype in AI is centered around the same names: Google, Anthropic, and OpenAI. The narrative is simple: there’s a big GPT that can solve all your problems, and the best GPT is mine. But the truth is different: we can choose suppliers based on a variety of criteria. – – – I first explained how to select suppliers in the digital economy—both as an entrepreneur and as a developer—many years ago. So, what’s new today? – – – In this brief talk, we’ll focus on the main criteria for choosing AI suppliers, as well as the specific considerations for AI. But more importantly, we’ll begin building a European community committed to making conscious choices, not emotional ones. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Valentino Spataro1113:4000:15From Data Sovereignty to Intelligence Sovereignty: A Private First Secure AI Agentic Ecosystemhttps://www.sfscon.it/talks/from-data-sovereignty-to-intelligence-sovereignty-a-private-first-secure-ai-agentic-ecosystem/Secure AI Agents with FOSS, the EU AI Act, and Local-First InfrastructureSecure AI Agents with FOSS, the EU AI Act, and Local-First InfrastructureEurope’s digital sovereignty discussion has traditionally focused on data, cloud infrastructure, privacy, and open standards. AI agents now extend this challenge into a new layer. The intelligence sovereignty. When agents can retrieve private context, reason over sensitive data, call tools, write code, and trigger operational workflows, the central question is no longer only “Where is the data stored?” It becomes: “Who controls the intelligence layer that acts on the data?” This talk presents a practical, vendor-neutral architecture for building secure, local-first AI agent systems using Free and Open Source principles. The reference example is a Secure AppSec Triage and Remediation Swarm: a multi-agent workflow that ingests vulnerability reports, logs, stack traces, and security findings; applies PII and secret redaction; checks policy constraints; routes tasks to specialized agents; generates remediation guidance; and preserves audit trails for human review. The talk connects FOSS engineering with Europe’s regulatory direction, including the EU AI Act, data governance, privacy, cybersecurity, and responsible automation. It explains why AI governance cannot remain only at the level of policy documents, model cards, or compliance checklists. For agentic systems, governance must also be implemented in the runtime: deterministic orchestration, policy-as-code, tool authorization, auditable memory, sandboxed execution, privacy gates, and human approval checkpoints. The core argument is that Europe’s next sovereignty challenge is not only data sovereignty, but intelligence sovereignty: the ability for organizations, public institutions, developers, and communities to inspect, self-host, govern, and safely operate AI systems that act on their behalf. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Krishnendu Dasgupta1114:0000:15The Last Decadehttps://www.sfscon.it/talks/the-last-decade/A Research-Based Case for Why AI Without Openness Ends Badly for EveryoneA Research-Based Case for Why AI Without Openness Ends Badly for EveryoneThe next ten years are the last period in which the governance architecture of AI can be meaningfully shaped by society. After that, the concentration of capability, infrastructure, and data will be too advanced to reverse through democratic or regulatory means. Open source is not a silver bullet — but it is a necessary condition for any future in which humans retain meaningful agency over the systems that govern their lives. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Thomas Aichner1114:2000:15Open Source Building Blocks for Europe’s Independent AI Future Ecosystemhttps://www.sfscon.it/talks/open-source-building-blocks-for-europes-independent-ai-future-ecosystem/As large language models (LLMs) rapidly enter European products and public services, most of the underlying AI infrastructure still comes from non‑EU providers. This creates not only technical and economic challenges, but also legal and strategic risks, as sensitive data and critical capabilities depend on platforms governed by foreign laws. The talk highlights how familiar trade‑offs, such as cost versus control, convenience versus compliance, and features versus freedom, are relevant in the AI and LLM contexts. The speaker draws on more than a decade of experience delivering IoT and big data solutions for companies in Alto Adige/Südtirol and beyond to outline how open‑source LLMs, European cloud providers, and hybrid architectures can help organisations reduce dependency on non‑EU AI platforms, keep European data under European jurisdiction, and gradually build more autonomous AI systems that align with EU values and regulations. The talk also argues for Alto Adige/Südtirol as an ideal place to focus on Sovereign AI efforts. It provides a strong existing open‑source culture with a network of SMEs, public bodies, and research institutions that already collaborate on digital innovation. It also provides real-world use cases for agriculture, tourism, public administration etc. Furthermore, its cross‑border, multilingual context also closely reflects the challenges many European regions face, making locally developed solutions with open‑source LLMs and EU‑based infrastructure directly transferable to the wider EU context. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Abid Munir Bajwa1114:4000:15E-mail and Digital Sovereigntyhttps://www.sfscon.it/talks/e-mail-and-digital-sovereignty/Why a decades-old standard plays a key role in your digital independenceWhy a decades-old standard plays a key role in your digital independenceWhen talking about Digital Sovereignty, most people refer to file formats, document storage, cloud services and messengers. There is, however, one crucial tool that many don’t have on screen: e-mail. Its underlying standards like SMTP and IMAP reach back to many decades ago, with some more recent developments like DMARC and MTA-STS. E-mail could be considered one of the world’s first federated messengers, is usually business-critical and the hosting market is dominated by a very few big players. They tend to impose their own rules and often create some sort of lock-in effect, just like proprietary software and closed document formats do. In this talk, I will explain why proper e-mail handling is more important than ever, point out various options from self-hosting to cloud providers, and raise awareness of that one service that is too often not on the roadmap for Digital Sovereignty. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Florian Effenberger1115:0000:15GNU Talerhttps://www.sfscon.it/talks/gnu-taler/Digital cash, made socially responsibleDigital cash, made socially responsibleGNU Taler is a Free Software privacy-friendly payment system from the GNU project. Taler does not use a blockchain: monetary policies are expected to be set by a central bank. Digital money can be issued either by retail banks or by the central bank. Paying with Taler is extremely efficient: customers authorize transactions with one click, and then the system takes only a few milliseconds to finalize the transactions. In this lecture, we will introduce the political goals, architecture and technology behind Taler, demonstrate the current state of the art and outline likely future developments. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Christian Grothoff1115:2000:15Sovereignty and security happen before downloading Linux (…updates)It's not the code that makes the difference nor its open source. And yet the 'open' is a dependency.It's not the code that makes the difference nor its open source. And yet the 'open' is a dependency.Replacing Windows with Linux-based operating systems is increasing. It is assumed that Linux operating systems that are made up of open source code allow sovereignty and increase security and independence. However, is an open source community more predictable than a commercial company? Can common interests be assumed to exist? What if the open source community maintaining the chosen operating system decides to make changes that break our use case? Would the user find out before the update is installed? What risks would rise if suddenly whole countries relied on projects never intended to serve such a demand? Is sole financing or introducing open source code a solution to digital sovereignty with the necessary usability, security and predictability? Based on the definition of a sovereign state, some Linux operating system communities fulfill most conditions: they have governance, representing and making compromises for their permanent users. They have the capacity to engage into relations with each other: there is collaboration through common code like the “Linux kernel” but also cross-community collaborations through people, who align practices or release cycles. They maintain feedback loops to projects whose code they use and evaluate how that code will evolve to adjust on time. Established communities have to be connected to ensure that critical security issues (especially in the “Linux kernel”) are mitigated in all of them before public disclosure, but also that there are no leaks in advance. These communities manage new code long before any law could attempt to regulate it. The only unfulfilled condition of these operating system communities is the lack of geographical territory: yet, could this be their biggest strength and maybe even a requirement for any digital sovereignty? Widespread review and testing remain major means to prevent unexpected code behavior and vulnerabilities. It is a matter of likelihoods that if there are flaws or vulnerabilities in code, Linux communities or their allies will identify them first and fix them before serious issues occur. Many of these likelihoods depend on forwarding as much relevant knowledge as possible to the one person probably best qualified to make use of it, connecting knowledge globally. Linux communities support intuitive knowledge creation and timely knowledge transfer. Do the territorial boundaries of traditional sovereign states allow anything similar? Even in the geographical sovereignty, no state can survive on its own without allies: unconditional independence does not exist. Beyond territorial boundaries, what are the differences between these communities and sovereign states? Do they overlap? How to define and distinguish them before discussing their relationship? Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Christopher Klooz1115:4000:15OpenStreetMap for Digital Sovereignty: commons are not magichttps://www.sfscon.it/talks/openstreetmap-for-digital-sovereignty-commons-are-not-magic/OpenStreetMap is more than a map: it is a digital common and a critical layer of open infrastructure. But commons are not magic. They require maintenance, governance, conflict resolution, funding and care. This talk explains why open geodata matters for digital sovereignty, and why OSM must be sustained as shared infrastructure. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Napolitano Maurizio1116:0000:15Building a Sustainable Open Source Ecosystem for Municipal Tax Serviceshttps://www.sfscon.it/talks/building-a-sustainable-open-source-ecosystem-for-municipal-tax-services/Public Money, Public Code in ActionPublic Money, Public Code in ActionPublic administrations across Europe are investing unprecedented resources in digital transformation through public programs and funds such as FESR 2014 – 2020 and FESR 2021 – 2027. Too often, however, these investments result in fragmented proprietary solutions with limited reuse and long-term dependency risks. The Goffice 2.0 – Tax project demonstrates a concrete and scalable alternative: how public money can generate public code, while at the same time fostering a healthy and competitive open-source market. Developed within the architecture Goffice 2.0 of the Consorzio dei Comuni della Provincia di Bolzano, Goffice 2.0 – Tax is a shared, cloud-based system for the digital management of municipal taxes and fees, designed to be used by all 116 municipalities of South Tyrol. The solution covers the entire lifecycle of local taxation processes, from citizen and contract management to calculation engines, electronic invoicing, digital notifications and payments, fully integrated with national enabling platforms such as ANPR, SPID, CIE, PagoPA, FatturaPA and the National Digital Notifications Platform. The application has been developed by Sidera ICTEase as a Java full-stack solution, based on a microservices architecture orchestrated with Kubernetes, ensuring scalability, resilience and long-term maintainability. The user interface is entirely built using the Vaadin framework, enabling the delivery of modern, high-performance web applications tightly integrated with backend services. A key aspect of the project is its open source by design approach, aligned with the public money – public code principle: the software is developed once, financed with public funds, and is reusable over time by other public administrations. This enables economies of scale, procedural standardization and long-term sustainability, while reducing vendor lock-in. From an economic and market perspective, the project highlights the complementary roles of two companies within an open-source ecosystem: • Vaadin, an Apache 2.0–licensed framework, serves as a strategic pillar for digital sovereignty. By seamlessly integrating modern web standards into the Java ecosystem and automating communication between browser and server, Vaadin allows developers to build sophisticated full-stack applications without complex frontend stacks. Its open-source model provides public authorities with full control over their software stack and data, supporting independence from proprietary silos. • Sidera ICTEase, an Italian company specialized in digital solutions for the public sector, has designed and implemented Goffice 2.0 – Tax leveraging open technologies and open-source components. Its business model is centered on value-added services—architecture design, development, customization, system integration, training and long-term support—demonstrating how open source shifts the market from license-based revenues to expertise and service quality. The talk will share architectural choices, governance aspects and economic implications of the project, highlighting best practices and lessons learned for public administrations and companies aiming to build sustainable open-source solutions together. Goffice 2.0 – Tax shows that open source, when combined with clear governance and service-oriented business models, can deliver innovation, efficiency and long-term public value. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Michele Tais1116:2000:15PA-Webinar – Public administration meets in digitalhttps://www.sfscon.it/talks/pa-webinar-public-administration-meets-in-digital/hosting webinar for the italian public administation done righthosting webinar for the italian public administation done rightPA-Webinar is a platform to host webinar, based on Jitsi Meet and the design system .italia. It is actively developed and maintained by the Dipartimento per la Trasformazione Digitale. It is build on well-know and maintained open source project and takes an DevSecOps approach to ease the adoption and the deployment and monitoring with auto-scaling. There are wizard and nice defaults for who organize events taking into account all the requirement mandates by the law and guidelines especially for usability and accessibility issues. We will present the software and introduce the workshop to allow the participants to have a look and deploy it on their cloud infrastructure. https://github.com/italia/pa-webinar Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Daniele PizzolliRaffaele Vitiello1116:4000:15Open-Source Licenses VS Generative AI: are we breaking the rules?https://www.sfscon.it/talks/open-source-licenses-vs-generative-ai-are-we-breaking-the-rules/Generative AI systems are increasingly trained on vast amounts of available source code, much of which is released under Free and Open-Source Software (FOSS) licenses. This talk aims to provide an overview of the possible conflict between open-source licensing frameworks and the emerging practices of AI training and source code generation. It will focus on key legal issues, including the implications of using open-source code for training purposes, the notion of derivative works within the context of AI-generated outputs, and the risks associated with the reproduction of licensed code without proper attribution. The purpose of the talk is to explore the potential legal risks arising from the use of generative AI in software development. The session seeks to highlight current legal uncertainties and to consider how open-source principles may continue to foster AI innovation without undermining digital sovereignty. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Cristiana Roda1117:0000:15AI Act, copyright and Open-Source: building transparent AI in Europehttps://www.sfscon.it/talks/ai-act-copyright-and-open-source-building-transparent-ai-in-europe/As Artificial Intelligence systems become increasingly embedded in business processes and digital infrastructures, ensuring compliance with European regulations is no longer optional. The AI Act, Reg. (UE) 1689/2024, together with EU copyright rules, introduces new requirements around transparency, data provenance and accountability, particularly in relation to the use of protected content in training datasets. This talk explores how these regulatory frameworks impact the design and deployment of AI systems, with a specific focus on intellectual property and copyright compliance. Moving beyond legal theory, it highlights the practical challenges organizations face when building or adopting AI solutions in Europe. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Veronica Meneghetti1117:2000:15The Unbearable Lightness of Humans in the Loophttps://www.sfscon.it/talks/the-unbearable-lightness-of-humans-in-the-loop/Human Sovereignty under the GDPR and the AI ActHuman Sovereignty under the GDPR and the AI ActMore and more decisions about us — small and large ones, routine and life-shaping — are being taken by machines. We are entering an age in which algorithms decide, or co-decide, on jobs, education, loans, social benefits, insurance, and the content we are shown. The EU GDPR and the EU AI Act address automated decision-making (ADM) directly and, in central cases, require a human “in the loop”. Some national laws, notably the new Italian AI statute, address these issues in specific sectors. But what does this actually mean in practice? When is a human click enough, and when is it not? This talk sets out where the EU GDPR and the EU AI Act draw the line, where national rules may add further obligations, and, in particular, which concrete design choices in ADM projects may decide whether human oversight is what it claims to be: a last sovereign act of human decision. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Christian Notdurfter1117:4000:15Lost in Regulation? Navigating the EU Cybersecurity and Data Law Jungle as a Tech Companyhttps://www.sfscon.it/talks/lost-in-regulation-navigating-the-eu-cybersecurity-and-data-law-jungle-as-a-tech-company/How FOSS helps you survive the EU regulatory stackHow FOSS helps you survive the EU regulatory stackThe EU has produced one of the most ambitious digital regulatory frameworks in history — but for tech companies, developers, and SMEs trying to stay compliant, it increasingly feels like a jungle: overlapping obligations, staggered deadlines, conflicting definitions, and regulatory gaps that no single lawyer or consultant can fully map alone. This talk gives a frank, practitioner-level overview of the regulatory landscape that any technology company operating in Europe must navigate today: GDPR (data protection and privacy), NIS-2 (cybersecurity obligations for essential and important entities), the EU AI Act (risk-based AI governance), the Cyber Resilience Act (security requirements for products with digital elements), and the Data Act (data access and interoperability). For each framework, the talk identifies who is actually affected, what the core obligations are, how the frameworks interact and conflict, and — critically — what a small or medium-sized tech company can realistically do to manage compliance without a team of lawyers. The talk also highlights how open-source software and open standards reduce regulatory risk: transparent, auditable, and interoperable systems are structurally better positioned to meet the EU’s growing compliance requirements than proprietary black-box alternatives. Attendees leave with a practical mental model for mapping their own regulatory exposure and a set of open-source tools and practices that address multiple compliance obligations at once. Digital SovereigntyDigital SovereigntyDigital Sovereignty

https://www.sfscon.it/tracks/digital-sovereignty-track-2026/

Patrick Agostini1109:1500:15GreetingsSFSCON - Main trackSFSCON - Main trackSFSCON,Main track0011:0000:15From Smart Traps to a Sovereign National Insect Database: Open-Source Local AI for Pest Monitoringhttps://www.sfscon.it/talks/from-smart-traps-to-a-sovereign-national-insect-database-open-source-local-ai-for-pest-monitoring/Europe needs pest-monitoring infrastructures that keep strategic agricultural data, software and operational know-how under European control. Building on the experience gained through INSTINCT, a project co-financed by the European Union under the ERDF 2021–2027 Programme of the Autonomous Province of Bolzano–South Tyrol, this talk presents a scale-up scenario centred on a national, cloud-based insect-monitoring database fed by smart traps, environmental sensors and expert-validated observations. The proposed infrastructure would be governed by a Bolzano-based partnership bringing together the research expertise of the Laimburg Research Centre and Eurac Research, the academic expertise of the Free University of Bozen-Bolzano, and the local technological capabilities of FOS S.p.A. A local application can acquire trap images, perform AI-based insect detection and counting using open-source tools and models such as YOLO and GNU Linux OS, support expert review and correction, and then transmit validated data to the central national repository. The application may also include project-specific software components that are not distributed as open source. The architecture separates local analysis from national data consolidation: processing may remain on premises or at the edge, while authorised observations are collected in a single controlled database hosted and managed under European jurisdiction. The presentation will discuss interoperability, traceability, data quality, model reuse and long-term independence from extra-European proprietary platforms, using Cydia pomonella as the initial use case and extending the approach to other insect species and territories. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Giovanni Giannotta1111:2000:15Open Tools for Environmental Monitoringhttps://www.sfscon.it/talks/open-tools-for-environmental-monitoring/Environmental contamination poses significant challenges for sustainable land management and informed decision-making. This talk presents an open-source web platform, developed by Gruppo FOS in Italy,as part of the PoLaRecCE European project, designed to simplify the collection, management, and visualization of environmental monitoring data. Through an interactive map-based interface, the platform enables researchers, public authorities, and other stakeholders to explore contamination data, monitor affected areas, and support evidence-based decisions. The presentation will also discuss how open technologies improve transparency, collaboration, and accessibility in environmental data management, making advanced monitoring tools available to a wider community, also how the open-source software contributes to European digital sovereignty by enabling organizations to retain control over their infrastructure, data, and software stack while avoiding vendor lock-in. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Orneda Lecini1111:4000:15GEOWIDGET – An Open-Source God’s Eye for Droneshttps://www.sfscon.it/talks/geowidget-an-open-source-gods-eye-for-drones/GEOWIDGET is a component built to track drones’ locations in real time in order to facilitate operators’ monitoring efforts during emergency missions. It comprises the following components: – A web dashboard for inserting mission-related data; – An open-data-hub-compatible, web-based OpenStreetMap component; – A NodeJS API; – A mySQL database; – An MQTT-listener for receiving drones’ location in real time. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Gianluca Lorusso1112:0000:15From Raster Files to Open Data Products: Publishing Geospatial Data with STAC and Free Softwarehttps://www.sfscon.it/talks/from-raster-files-to-open-data-products-publishing-geospatial-data-with-stac-and-free-software/Using Free Software and STAC to make large raster datasets findable, interoperable, reusable, and digitally sovereignUsing Free Software and STAC to make large raster datasets findable, interoperable, reusable, and digitally sovereignLarge raster datasets are central to earth observation, climate research, environmental monitoring, and public-sector data services. Yet they are often published as files that are hard to discover, compare, cite, or reuse without knowing the exact storage location, format, and processing tools in advance. raster2stac is a Free Software Python library that helps turn existing raster datasets into structured, standards-based data products. It creates SpatioTemporal Asset Catalogs (STAC) Collections, Items, and Assets from formats such as netCDF, Cloud Optimized GeoTIFF, and Zarr, so datasets can be described consistently and accessed through common open geospatial tools. STAC is now an OGC Community Standard and is widely used in the earth observation ecosystem, including by major public data providers and platforms such as Copernicus Data Space Ecosystem and the NASA Earthdata. This makes it a practical bridge between local data publishing workflows and the broader open geospatial infrastructure used by researchers, public institutions, and service providers. In this lightning talk, I will show the problem raster2stac addresses, the publishing workflow it enables, and why this matters for digital sovereignty. The focus is not only on converting files, but on making scientific and public geospatial data more findable, interoperable, reusable, and less dependent on vendor-specific platforms or custom data portals. The audience will leave with a practical understanding of how open standards such as STAC, combined with Free Software, can help institutions publish raster data in a way that is easier to share, automate, preserve, and build upon. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Suriyah Dhinakaran1112:2000:15PCoRaS: Package for Cloud-Optimized Raster and Spatio-temporal Datahttps://www.sfscon.it/talks/pcoras-package-for-cloud-optimized-raster-and-spatio-temporal-data/Building Multi-Dimensional Data Cubes from Geo-spatial Temporal DataBuilding Multi-Dimensional Data Cubes from Geo-spatial Temporal DataThe increase of volume, complexity and demand for geospatial data, such as climate models and satellite data, requires the adaptation of storage, reproducibility and traceability of the high-volume, large-scale data. Traditional formats such as NetCDF or TIFF, although reliable and efficient, do not come with the flexibility and cloud-native adaptability of Zarr files. This project presents a library, PCoRaS, designed to create multidimensional data cubes in Zarr format from different sources of geospatial and raster data. The library takes one or multiple raster-based datasets as input and generates an N-dimensional data cube according to user-defined variables, dimensions and shared metadata attributes. By allowing users to define which variables should be extracted and how dimensions should be structured, the library supports the transformation of heterogeneous raster inputs into organized analysis-ready data cubes. A particular feature of the library is its support for dual temporal dimensions, enabling the aggregation of variable data over specific temporal frames, such as months, while preserving a broader temporal dimension representing one or multiple years. This functionality is especially relevant for climate and environmental datasets where seasonal, monthly, and long-term temporal structures must coexist within the same data cube. This project works on top of Raster2STAC. The package extracts the metadata from raster data, generating json files compatible with STAC, creates STAC collections with items and assets, and automatically uploads the data to a S3 bucket. The combination of Zarr data cube generation, STAC metadata and S3 publication contributes to a FAIR and cloud-native geospatial infrastructure, allowing climate data and satellite data to comply with the requirements of scalable access, interoperability and reproducibility. Overall, this work contributes a reusable and scalable workflow that supports more interoperable, reproducible and cloud-ready geospatial data processing EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Miguel Angel Martinez Guerrero1112:4000:15From WebGIS to Digital Sovereignty: Building a Full Online GIS with MapView and Open-Source Technologieshttps://www.sfscon.it/talks/from-webgis-to-digital-sovereignty-building-a-full-online-gis-with-mapview-and-open-source-technologies/Digital sovereignty is becoming an increasingly strategic requirement for public administrations and is closely tied to the responsible use of public funds. Ensuring long-term control over data, technological autonomy, and independence from proprietary platforms allows public investments to generate sustainable value over time, avoiding vendor lock-in and reducing structural dependency on closed solutions. MapView, the new geospatial platform of the Autonomous Province of Bozen/Bolzano, has been designed with these principles at its core, following the idea that software developed with public resources should remain a public, reusable asset. MapView goes well beyond the limitations of a traditional WebGIS. By integrating a Geobrowser and Geobrowser into a single environment and relying on open standards and open-source libraries, the platform enables public institutions to retain full control over authoritative geospatial data and services. Interoperability is ensured through open OGC services, allowing seamless data reuse across systems while avoiding vendor lock-in and preserving long-term platform independence. In line with these principles, a large portion of MapView’s datasets and services—whenever they can be considered public—is openly accessible not only to public administrations but also to citizens, professionals, and the private sector. For professional users and private organisations, access to non-public or restricted datasets can additionally be granted through explicit consent from the respective data owners, within a controlled and transparent governance framework. Within this open and modular architecture, the new 3D component, based on Cesium, plays a fundamental role. The 3D viewer is tightly integrated into the overall system and enables real-time, interactive exploration of the territory in three dimensions using a fully open-source stack. More importantly, it establishes the foundational layer for the evolution of MapView towards the first Digital Twin of South Tyrol. By combining authoritative geospatial data, open services, and 3D representations in a shared environment, MapView lays the groundwork for future simulations, scenario analysis, and data-driven planning—extending its scope far beyond simple visualisation. MapView has been conceived as an evolving online GIS platform, already deployed and actively used in production—not as a prototype or experimental solution. At both municipal and provincial levels in South Tyrol, it supports operational territory management and a wide range of everyday administrative activities, acting as a shared and reliable geospatial infrastructure for public administrations. The platform provides a rich and continuously expanding set of functionalities, including the integration of external datasets and services, advanced querying and analytical capabilities, and a modular architecture designed for continuous growth. The introduction of a mobile-first client application further enhances operational autonomy, enabling effective field usage while ensuring that data ownership and control remain firmly within the public domain. With its open-by-design architecture and strong focus on autonomy, independence, and data control, MapView demonstrates how digital sovereignty and responsible public investment can reinforce each other in practice. This talk presents the architectural choices behind MapView, the role of open 3D technologies within the platform, and the long-term vision of a truly independent, extensible, and production-grade online GIS—capable of evolving into a shared Digital Twin and moving decisively beyond “yet another WebGIS”. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Edoardo ScepiJudith OberlechnerPietro d’Orio1113:0000:15Digital Sovereignty with Local RAGhttps://www.sfscon.it/talks/digital-sovereignty-with-local-rag/A minimal open-source prototype using open data and open modelsA minimal open-source prototype using open data and open modelsMany companies want to experiment with AI, but they also need to keep control over their data, infrastructure and model choices. This short talk presents a minimal Retrieval-Augmented Generation prototype built with open data, an open-source database and an open model. The session walks through the essential RAG pipeline: loading data, creating chunks, generating embedding, storing them in a searchable database, retrieving relevant context and producing grounded answers. The goal is to show how companies can start exploring private AI with a small, understandable and reproducible open-source setup. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Alessandro AneggiAndrea Janes1113:2000:15Who Owns Your AI Data Pipeline?https://www.sfscon.it/talks/who-owns-your-ai-data-pipeline/Open standards, lineage, and lakehouse architectures for modern AI systemsOpen standards, lineage, and lakehouse architectures for modern AI systemsAl is increasing the complexity of modern data platforms, but many organizations are building systems with limited visibility into how data moves, changes, and is consumed. This talk explores approaches for building observable and portable data platforms using open technologies and open standards. Through real-world examples from large-scale enterprise environments, we will look at how tools and architectures based on Delta Lake, Apache Spark, and OpenLineage can improve lineage tracking, governance, interoperability, and operational transparency across modern data pipelines. The session covers common problems in Al-oriented data architectures, including fragmented metadata, hidden operational costs, and observability. We will also discuss practical engineering patterns for building lakehouse platforms that remain flexible, traceable, and easier to operate over time. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Samantha Cruz1113:4000:15Why Scaling AI Demands Modern Checkpointinghttps://www.sfscon.it/talks/why-scaling-ai-demands-modern-checkpointing/When building AI, everyone talks about data, architectures, and compute. Almost nobody talks about the massive bottleneck hidden in plain sight: saving the model. As models grow, traditional synchronous saving stalls expensive hardware, costing time and money. In this 15-minute lightning talk, we pull back the curtain on Orbax—the modern, open-source checkpointing powerhouse engine. We will explore how Orbax fundamentally changes state management through asynchronous writing, preventing your storage from bottlenecking your compute. Attendees will walk away understanding why modern ML infrastructure requires dedicated, smart persistence layers to scale seamlessly and efficiently. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Nikhil BansalAbhishek Agrawal1114:0000:35Own your AI agent: running open source agents on your termshttps://www.sfscon.it/talks/own-your-ai-agent-running-open-source-agents-on-your-terms/Owning your AI stack isn’t just about ideology, it’s about building systems you can actually trust and understand. This talk shows how to run your own AI agents in practice, relying on open source code and models: through concrete examples (and some inevitable failures) it will explore what distinguishes open source agents from commercial services, and present tools and techniques that make local, custom agents feasible, not just possible. We’ll also look at Otari, Mozilla.ai’s open-source LLM gateway, as an example of the infrastructure that helps make AI ownership practical by giving developers more control over models, routing, and observability while keeping a consistent interface. The session focuses on practical lessons learned from building and operating open source AI systems, showing how ownership, transparency, and flexibility can be achieved without sacrificing developer experience. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Davide Eynard1114:4000:35Panel EngineeringEngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

0015:2000:15Generative AI in a Truly Sovereign Digital Europehttps://www.sfscon.it/talks/generative-ai-in-a-truly-sovereign-digital-europe/“European sovereign AI” is on every vendor’s slide. But the model’s nationality is the wrong question: what makes generative AI sovereign is where it’s served and whose laws bind the provider – and that’s exactly what sovereignty washing hides. A European-branded model running on a non-EU cloud is still one lawful order away from a foreign government, no matter which data centre stores the bytes. This lightning talk is a field guide to real sovereignty for LLMs. We map the spectrum – from convenient-but-exposed to genuinely sovereign – show how to spot the washing before you sign, and walk through the honest trade-offs (cost, quality, availability) we hit while building the backend for our startup, lynq.ing. The takeaway: the handful of questions that separate sovereign from sovereignty-washed. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Christian Gapp1115:4000:15You get 5000 line AI slop contributions, what do you do?https://www.sfscon.it/talks/you-get-5000-line-ai-slop-contributions-what-do-you-do/How to handle AI agents contributing to your open source projectsHow to handle AI agents contributing to your open source projectsThe way open-source repositories receive contributions has changed drastically over the past two years. Maintainers are increasingly receiving code that has never been tested, code that completely rewrites the codebase, contributions from non-technical users, and even submissions by those who are not people at all. Traditional strategies for managing community contributions are becoming less effective, and the discussion about how to adapt is still ongoing. I’ll be sharing a few considerations and tips for handling these new situations, based on my experience maintaining several Nextcloud repositories. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Grigory Vodyanov1116:0000:15Reflecting on Access Control in Agentic Architectureshttps://www.sfscon.it/talks/reflecting-on-access-control-in-agentic-architectures/Agentic architectures are rapidly evolving, enabling AI agents to autonomously orchestrate workflows, access heterogeneous data sources, and act on behalf of users. However, this increased autonomy amplifies a fundamental challenge: how do we ensure that agents only access—and reason over—data they are truly authorized to use? In this talk, we reflect on access control as a first-class concern in agentic systems, going beyond traditional models designed for static applications. We argue that Attribute-based Access Control provides a natural foundation for these systems, as it explicitly captures the interplay between users, agents, and execution environments. This perspective enables the enforcement of the principle of least privilege across complex, multi-agent scenarios, where the same agent may operate under different constraints depending on context.The talk concludes with open questions for the software development community, including how to make authorization transparent to agent design and how to embed secure access control mechanisms as a core building block of future agentic systems. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Eduardo Guerra1116:2000:35Is Your AI Stack Risk-Free ?https://www.sfscon.it/talks/is-your-ai-stack-risk-free/Verifying True Open Source AI for ProductionVerifying True Open Source AI for ProductionAs AI models become central to modern development stacks, developers and decision-makers face a critical liability: what actually makes an AI model “Open Source” ? Relying on architectures that are merely “Open Weight” without true data and code transparency introduces massive risks to an organization’s digital sovereignty, data privacy, and long-term compliance. This technical session moves past the marketing hype to provide a rigorous framework for verifying true Open Source AI. We will dive straight into the operational realities of building a sovereign AI stack. Attendees will learn how to: – Navigate the Paradigm Shift: Understand why defining Open Source AI is fundamentally different (and significantly harder) than traditional Open Source Software. – Stress-Test Openness: Apply community frameworks like the Open Source AI Definition to rigorously evaluate models before putting them into production. – Spot the Fakes: Break down the technical realities of “Open Weights” and learn how to identify and call out deceptive “openwashing” in the commercial ecosystem. Whether you are an engineer selecting models for production or a decision-maker protecting your infrastructure from vendor lock-in, this talk delivers the concrete tools needed to defend your digital rights on the modern AI frontier. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Alfonso Cancellara1117:0000:15Beyond Prompts: Applying Domain-Driven Design to LLM Session Architecturehttps://www.sfscon.it/talks/beyond-prompts-applying-domain-driven-design-to-llm-session-architecture/Building decent software has always required discipline; with LLMs, that discipline becomes even more critical. This talk applies Domain-Driven Design (DDD) to treat LLM sessions as Bounded Contexts, with a shared ubiquitous domain language and explicit interaction states to reduce drift and improve output reliability. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Daniel Degasperi1117:2000:15Interfacing LLMs with Systems Managementhttps://www.sfscon.it/talks/interfacing-llms-with-systems-management/A Practical Evaluation of the Uyuni MCP ServerA Practical Evaluation of the Uyuni MCP ServerAs Large Language Models (LLMs) start executing tasks via autonomous agents, system administrators need secure ways to interface AI with real infrastructure. The Uyuni MCP Server is a Model Context Protocol implementation that bridges the gap between LLMs and the Uyuni management solution. It allows AI agents to securely retrieve data and schedule tasks on the managed infrastructure using natural language. The presentation will offer an evaluation of the new Uyuni MCP Server. It will show how this component exposes system metrics and lifecycle tools to compliant AI clients. The presentation will also focus on the strict security boundaries and configuration parameters required to maintain human control over sovereign environments. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Marina Latini1117:4000:15Unknown Data, Useful Answershttps://www.sfscon.it/talks/unknown-data-useful-answers/AI as an Adaptive Data AnalystAI as an Adaptive Data AnalystMaking sense of data has traditionally required understanding structure, schemas, building pipelines and defining metadata before any meaningful analysis could begin. But what if you could start with questions instead? In this talk, we will explore how AI can act as an adaptive data analyst, helping users work with unfamiliar datasets, documents and heterogeneous sources without extensive preparation. By discovering structure, categorizing information and identifying relationships, AI can accelerate data exploration and turn raw data into useful answers. Through practical examples, we will see how AI-assisted workflows can help uncover insights, lower the barrier to data analysis and enable users to extract value from data they have never seen before. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Davide Montesin1111:0000:15Who Owns the Body? Robotics and the Next Frontier of Digital Sovereigntyhttps://www.sfscon.it/talks/who-owns-the-body-robotics-and-the-next-frontier-of-digital-sovereignty/We sell the best robots on the market — and they aren’t European yet. That’s not a contradiction in a sovereignty talk. It’s the reason for it.We sell the best robots on the market — and they aren’t European yet. That’s not a contradiction in a sovereignty talk. It’s the reason for it.Last year at SFSCON I argued that if AI is the mind, robotics is the body. This year: who owns the body? Digital sovereignty is usually about data and the cloud. But as autonomous robots spread into logistics, cleaning, hospitality and agriculture, sovereignty becomes physical — and uncomfortable. At Telmekom we sell and deploy these robots today, and today the best platforms on the market are not European. There is no mature European alternative ready to ship, and a business serving real customers cannot wait for one: the warehouse, the hospital floor and the harvest don’t pause for industrial policy to catch up. This talk makes the case for pragmatic sovereignty. The danger is not using foreign hardware — it is using it unconsciously, with no leverage and no exit. Naming the dependency is the first act of sovereignty, not a betrayal of it. The market gap is itself the argument for open standards: the realistic path to European capability runs through open middleware, auditable firmware and the right to repair — the layers that let a European integrator add value on foreign hardware now, and let European hardware plug in later without rebuilding the whole stack. In fifteen minutes I’ll show what it means to buy the body abroad while keeping the nervous system — the software, the data, the ability to audit, repair and migrate — under your own control, and name concrete levers for decision-makers in regions like South Tyrol. Open HardwareOpen HardwareOpen Hardware

https://www.sfscon.it/tracks/open-hardware-track-2026/

Sergio Srdjan Vemic1111:2000:15Autonomous Logistics: An Open-Source Approach For Integrated Warehouse Automationhttps://www.sfscon.it/talks/autonomous-logistics-an-open-source-approach-for-integrated-warehouse-automation/This talk presents the ground (AGV – Autonomous Ground Vehicle) and aerial (UAV –Unmanned Aerial Vehicle) robot software stacks of an industrial automation project where a mobile manipulator robot (AGV) and a drone (UAV) operate together to support warehouse and palletizing workflows. The presentation focuses on how a practical robotic system can be built by combining robust open-source technologies, reducing development time while keeping the architecture modular, transparent, and adaptable to real-world requirements. The presentation is split into two complementary parts: The UAV stack highlights two key open-source repositories that played a central role in the robot architecture: PX4, used as the flight control framework, and NVIDIA Isaac ROS Visual SLAM, adopted to provide visual localization and mapping capabilities in GPS-denied or structured indoor environments. The presentation will show how these components were integrated into the drone stack, why they were selected, and which advantages they provided in terms of flexibility, maintainability, and performance. The presentation also highlights the perception framework adopted across the robotic system, based on the open-source YOLOv11 network and the Ultralytics framework. This section focuses on how these technologies were integrated into both the AGV and UAV software stacks to provide a common perception layer supporting collaborative operations. The presentation will show how the camera mounted on the robot gripper is used for box detection and pose estimation during palletizing tasks, while the camera mounted on the drone provides information about box presence and approximate location on the target pallet, contributing to a coordinated aerial-ground robotic workflow. Open HardwareOpen HardwareOpen Hardware

https://www.sfscon.it/tracks/open-hardware-track-2026/

Tommaso TubaldoSimone Garbin1111:4000:15The power of open platforms in automationhttps://www.sfscon.it/talks/the-power-of-open-platforms-in-automation/Leading the software-defined revolution to break vendor lock-in and build the collaborative automation of tomorrow.Leading the software-defined revolution to break vendor lock-in and build the collaborative automation of tomorrow.AI, cybersecurity, and advanced robotics are the key trends driving automation. Bosch Rexroth leads this software-defined evolution by championing ultimate integration freedom. Within this vision, ctrlX AUTOMATION acts as the enabler: an open, Linux-based operating system platform that overcomes proprietary barriers to foster co-creation and partner ecosystems. Open HardwareOpen HardwareOpen Hardware

https://www.sfscon.it/tracks/open-hardware-track-2026/

Alberto Ascoli1112:0000:15From QNX to Linux & Zephyr: Co-Existence and Migration via VirtIO and Waydroidhttps://www.sfscon.it/talks/from-qnx-to-linux-zephyr-co-existence-and-migration-via-virtio-and-waydroid/Full migration away from QNX isn’t always feasible due to legacy constraints or safety certifications, but that doesn’t mean you have to accept total vendor lock-in. In this lightning talk, Stefan details how to design a flexible architecture where QNX co-exists with, or seamlessly transitions into, an open-source ecosystem. He will dive into using VirtIO as the standardized communication fabric between QNX, Linux, Android and Zephyr environments, and highlight how Linaro achieved seamless graphical integration of Android applications via Waydroid using virtio-gpu. Finally, he will share performance and latency tuning tips needed to handle mixed-criticality workloads. Open HardwareOpen HardwareOpen Hardware

https://www.sfscon.it/tracks/open-hardware-track-2026/

Stefan Schmidt1112:2000:15Open Hardware – A ‘develish’ Case Studyhttps://www.sfscon.it/talks/open-hardware-a-develish-case-study/How 'open' is a commercial open hardware product in reality?How 'open' is a commercial open hardware product in reality?In May 2025 the German audio hardware maker company Teufel announced a new product, the Mynd Bluetooth speaker bos. The speaker in itself is yet another Bluetooth speaker box, the remarkable feature of it is that it was announced as an “Open Source Project” with “high repairability and sustainability”, with all mechanical and electronics design files as well as the firmware source code published as “open source”. So they mean the Mynd to be a Free Software and open hardware product. So we wanted to put this to a test! We got ourselves a device and did a careful analysis of the design files provided by the manufacturer to answer the question: Is it really ‘open source’? And what does this mean for hardware? In this talk I will present what we found, eventual caveats and what this means to the central question: Is it ‘open source hardware’? Open HardwareOpen HardwareOpen Hardware

https://www.sfscon.it/tracks/open-hardware-track-2026/

Nicole Faerber1112:4000:15Deep dive into Open Hardware PowerPC Desktophttps://www.sfscon.it/talks/deep-dive-into-open-hardware-powerpc-desktop/Free hardware and software are the building blocks of our society and should be in our handsFree hardware and software are the building blocks of our society and should be in our handsHow we seek to encourage communities and students to acquire the skills and capabilities necessary to produce hardware and software independently and to have control over it. Therefore, we promote local and decentralized design and production. We collaborate with volunteers to create the opportunity to experience hands-on learning, to inspire creativity, critical thinking, and problem-solving trying to conceive, design and produce what we need or desire. We encourage this culture from primary schools to University and with no age limit. We share our open hardware projects that are governed by our Community that is open. Our Community conceive, design and produce and where needed, thanks to donations, we pay engineers. We are quite slow because we solve problems thank to our passion, creativity and obstinacy. Open HardwareOpen HardwareOpen Hardware

https://www.sfscon.it/tracks/open-hardware-track-2026/

Roberto InnocentiStefano Gronchi1113:0000:15When hardware breaks freehttps://www.sfscon.it/talks/when-hardware-breaks-free/Understanding open source hardware licensingUnderstanding open source hardware licensingOpen Source Hardware (OSH) is not only an important element of digital commons, as it has become a key enabler of permissionless innovation. Understanding licensing issues is important, however, to explore the maximal potential of OSH. This talk aims clarifying the main elements of OSH licenses. For instance: – What open source licenses are available for hardware? – What are the differences between licenses for software, hardware and data? – Are there “copyleft” and “permissive (non-reciprocal)” licenses for hardware in the same sense there are for software and data? – How do you implement a license in an OSH project? The talk aims professionals and hobbyists working and developing OSH. No previous knowledge on legal topics or OSH is required. LegalLegalLegal

https://www.sfscon.it/tracks/legal-track-2026/

Lucas Lasota1113:2000:15Open standards for open source hardware and other high-cost-of-change domains: the missing frameworkhttps://www.sfscon.it/talks/open-standards-for-open-source-hardware-and-other-high-cost-of-change-domains-the-missing-framework/An early-stage prototype to fix the standardization governance issueAn early-stage prototype to fix the standardization governance issueIn open source software governance doesn’t matter too much – incompetent leadership can simply be replaced via a fork. However, this organic governance has limits: One cannot fork a house. Similarly, governance matters more in domains with a high “cost of change”, like hardware (and also AI), where design decisions can exclude entire regions from replicating, maintaining, or commercializing a project. De jure standardization was a good very answer to this issue 100 years ago, but has seen little procedural innovation since then and is practically unfeasible for open standardization on a product level. The principles of Open Standard on the other hand come with modern workflows, but don’t aim to be representative, so they would be too prone to be compromised by dominant actors. This talk will outline some of the details of the problem and propose a potential solution aimed at closing this gap. LegalLegalLegal

https://www.sfscon.it/tracks/legal-track-2026/

Martin Häuer1113:4000:15Open-Source Software and Open-Source Hardware under the new European Product Liability Directivehttps://www.sfscon.it/talks/open-source-software-and-open-source-hardware-under-the-new-european-product-liability-directive/The revised EU Product Liability Directive (PLD) fundamentally changes the liability landscape for software and open hardware. While the Directive introduces an exemption for Free and Open Source Software developed or supplied outside the course of a commercial activity, the practical scope of this exemption remains uncertain and raises important questions for the FOSS ecosystem. This talk examines how the new PLD applies to open source software and hardware projects and where maintainers, foundations, research institutions and commercial actors may face liability risks. Particular attention will be given to two increasingly common scenarios: open source projects developed within third-party funded research initiatives and software distributed under dual-licensing models. The talk will also explore whether the rationale behind the software exemption should extend to open source hardware. Attendees will gain a practical understanding of when the PLD’s open source exemption is likely to apply, when liability risks may arise, and why contractual disclaimers alone cannot eliminate product liability under EU law. The session will further discuss how project governance, funding structures and distribution models can influence legal exposure. As one of the first analyses of the new PLD from an open source (hardware) perspective, this talk provides timely guidance for open source stakeholders aiming to build digitally sovereign solutions on free and open source technologies. While only future case law from the Court of Justice of the European Union can provide definitive answers, the presentation offers a framework for understanding the emerging legal landscape and preparing for compliance today. LegalLegalLegal

https://www.sfscon.it/tracks/legal-track-2026/

Tarmio Frei1114:0000:15EU Cyber Resilience Act for open-source hardwarehttps://www.sfscon.it/talks/eu-cyber-resilience-act-for-open-source-hardware/The EU Cyber Resilience Act makes cybersecurity mandatory for all network-connected products with digital elements. Open-source hardware projects are caught in a regulatory gap: they mix design files, firmware, documentation, and physical devices, but the CRA’s open-source exemptions only cover software. This talk works through six common OSH scenarios, from uploading CAD files to selling kits and finished devices, and shows where each triggers manufacturer duties under the CRA. It also covers the new Product Liability Directive, particularly the question of when design files become products and how liability shifts between non-commercial developers, community organisations, and commercial integrators. The goal is concrete guidance for OSH developers and makerspaces dealing with EU product law for the first time. LegalLegalLegal

https://www.sfscon.it/tracks/legal-track-2026/

Alina Zhykhar1114:2000:15From Right to Repair to DMA: Shaping the Open Source Hardware Landscapehttps://www.sfscon.it/talks/from-right-to-repair-to-dma-shaping-the-open-source-hardware-landscape/In the context of the upcoming compendium “Open Source Hardware in the EU Market”, this talk will briefly present how intellectual property rights, right to repair and the latest EU regulation on hardware interoperability shape the context of Open Source Hardware. In particular, after an overview of the relevant EU regulation on the topic of right to repair and interoperability rights, the talk will focus on the effects of the latest specification decision of the European Commission under the Digital Markets Act (Art. 6(7)) on hardware interoperability requirements for Apple and its impact on the Open Source Hardware market and legislative landscape. LegalLegalLegal

https://www.sfscon.it/tracks/legal-track-2026/

Dario Presutti1114:4000:15Copyleft on hardwarehttps://www.sfscon.it/talks/copyleft-on-hardware/How can we apply copyleft obligations to hardware? In other words, how can we ensure that the users’ freedoms to use, study, modify and share the design of a piece of hardware are retained? This talk aims to address this question by diving into the core legal mechanism of copyleft. In particular, we will look at the text of the GNU General Public License to understand how copyleft works, then adapt the obligations to the realm of hardware. This will allow us to understand the legal differences between software and hardware, identify limitations of license enforcement, but also untangle the complexity of combining software and hardware under copyleft licenses. LegalLegalLegal

https://www.sfscon.it/tracks/legal-track-2026/

Panos Alevropoulos1115:0000:15When Hardware Lives Onhttps://www.sfscon.it/talks/when-hardware-lives-on/Intellectual property beyond death, insolvency, and organisational liquidationIntellectual property beyond death, insolvency, and organisational liquidationExisting discussions of Open Source Hardware (OSH) have focused predominantly on licensing, while comparatively little attention has been paid to the long-term management of intellectual property when contributors or managing organisations cease to exist. The death of individual developers, the insolvency or liquidation of organisations, and the resulting absence of a legal entity capable of holding or enforcing intellectual property rights raise important questions for the continuity of OSH projects. This talk examines how copyright, patent and trademark rights may be managed in these situations. It considers who is responsible for enforcing software, hardware and data related licenses, and addresses the legal mechanisms available to preserve the continuity of OSH projects, including the management, transfer and enforcement of intellectual property rights. This talk is aimed at professionals and hobbyists working on and developing OSH. No previous knowledge on legal topics or OSH is required. LegalLegalLegal

https://www.sfscon.it/tracks/legal-track-2026/

Sofiia Martynenko1115:2000:15Functional Freedomhttps://www.sfscon.it/talks/functional-freedom/The Patent Law Dimension of Open Source HardwareThe Patent Law Dimension of Open Source HardwareAs Open Source Hardware (OSH) transitions from schematics and design files to functional physical objects, it intersects with the legal landscape of patent law. This presentation breaks down the legal dimension of OSH under the European Patent Convention (EPC), specifically addressing vulnerabilities of under-resourced OSH communities. We explore risks, including the wrongful granting of patents over prior art, the phenomenon of ‘patent upgrading’, and the inadvertent infringement of pre-existing patents. Finally, the talk outlines proactive mitigation and prevention strategies, such as defensive publishing and peer to patent processes, designed to safeguard the functional freedom of OSH. No previous knowledge on legal topics or OSH is required. LegalLegalLegal

https://www.sfscon.it/tracks/legal-track-2026/

Sophie Sakurako Vukovich1115:4000:15Export Control & Open Source Hardwarehttps://www.sfscon.it/talks/export-control-open-source-hardware/What Devs Should KnowWhat Devs Should KnowThis talk explains how export control rules apply unexpectedly to open-source hardware (OSH) and why modern practices (AI tools, embedded firmware) create new uncertainties. It guides developers through general guidelines on export control in the EU, the associated risks and suggests practical precautions. Attendees will learn what counts as an export, if digital and AI-generated hardware designs may trigger controls, and how to minimise risks, while advocating for clearer rules. LegalLegalLegal

https://www.sfscon.it/tracks/legal-track-2026/

Niharika Singhal1116:0000:35Panel Open Hardware & LegalLegalLegalLegal

https://www.sfscon.it/tracks/legal-track-2026/

0016:4000:40LLM-Generated Code and Copyrighthttps://www.sfscon.it/talks/llm-generated-code-and-copyright/Who legally owns your (vibe)code?Who legally owns your (vibe)code?The progress in large language models (LLMs) has rapidly transformed software development. The adoption of using LLMs to write code (so-called vibe coding), has seen significant adoption rates, not just among novice/amateur software developers, but now also in large software companies and in tech startups. Such developments raise difficult legal questions pertaining to intellectual ownership. Who owns copyright in such code? How do we approach licensing with LLM-generated code? And are our existing frameworks able to sufficiently handle these questions? This talk will focus on explaining these questions and the difficulties experienced when we go about trying to untangle these legal complexities, albeit with the caveat there are no easy answers to be had at the moment. LegalLegalLegal

https://www.sfscon.it/tracks/legal-track-2026/

Gabriel Ku Wei Bin1117:2000:35Intellectual Property on Source Code in times of agentic codinghttps://www.sfscon.it/talks/intellectual-property-on-source-code-in-times-of-agentic-coding/Does AI-translated Software really create its own IP?Does AI-translated Software really create its own IP?Big AI vendors keep telling us that “AI-rewriting” source code from one language into another produces fresh, original work that the user can relicense at will. But is this claim actually true? This talk walks us through a few recent, very real cases that put those kind of claims to the test from a legal pov. We will look at why neither approach holds up under existing IP law. This talk includes a brief overview and introduction to intellectual property rights aimed at Open Source developers. We revisit the Berne Convention, the Naruto v. Slater “monkey selfie” ruling on non-human authorship, and the recent German GEMA v. OpenAI decision. We will also discuss the limits of the arguments brought forward and legal matters showing up on the horizon. Along the way we will also clarify the possible legal impact of agentic programming. The goal is to give FOSS maintainers, contributors, and decision-makers a clear, practical mental model for evaluating “AI-(re)wrote-it” claims – both when they are made to their projects and when they are tempted to leverage the power of LLMs themselves. LegalLegalLegal

https://www.sfscon.it/tracks/legal-track-2026/

Struberg Mark1111:0000:35Opportunity, Standardization, Sovereignty and FOSShttps://www.sfscon.it/talks/opportunity-standardization-sovereignty-and-foss/Lessons on the Need and Value of FOSS from the CRA Standardization ProcessLessons on the Need and Value of FOSS from the CRA Standardization ProcessThe EU Cyber Resilience Act of 2024 will be partially in effect as of this year’s SFSCON, but the reality of its implementation and enforcement will continue for many years. As one of the most visible and immediate efforts at EU technological sovereignty the CRA is a politically important regulation, but it is also an incredibly ambitious one on its own, adding safety regulation to the software industry. Most notably for the SFSCON audience the CRA is also a regulation where the FOSS community has provided a great deal of input, advice, and has had some success in shaping both the law and its implementation. With the standardization and many of the other implementation necessities for the CRA drawing to a close, a look at their successes and failures, and FOSS’s role in them can provide a road map for the future of FOSS interactions with tech sovereignty regulation. This talk will point out some of the frictions and struggles of CRA implementation, noting the very real ways that FOSS contributions have mitigated them, and highlighting additional ways open source voices and values might further improve the standardization. Beyond how FOSS can help the EU implement its cybersecurity, sovereignty, and privacy goals its equally important to understand how these goals can help open source projects, companies and developers. Again using the CRA as a model I hope to point out ways that its regulatory burden on manufacturers may benefit open source projects and communities both financially and provide them with a greater voice in future technology regulation. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

August Bournique1111:4000:15The Cyber Resilience Act Is Here – What It Means for Open Sourcehttps://www.sfscon.it/talks/the-cyber-resilience-act-is-here-what-it-means-for-open-source/A practical guide to scope, obligations, and the path to complianceA practical guide to scope, obligations, and the path to complianceThe EU Cyber Resilience Act (CRA) is no longer a future concern. As of September 2026, manufacturers must report actively exploited vulnerabilities and severe incidents – the first hard obligation to come into force, ahead of full enforcement in December 2027. So where does this leave companies built on open source software? This talk is a practical guide to navigating the CRA. We start with the essentials – is your project in scope, and when is your business a “manufacturer” versus an “OSS steward”, the new role the CRA created specifically to address open source? We then walk the path to compliance, with a particular focus on third-party and open source due diligence: building an accurate picture of your software supply chain, assessing the health and security of the components you depend on, handling vulnerabilities, and meeting the new incident-reporting obligations. Crucially, we’ll look at why contributing fixes back upstream isn’t just good citizenship – it reduces your long-term compliance burden and strengthens the shared commons the whole ecosystem relies on. Whether you publish open source, build commercial products on top of it, or steward a project others depend on, you’ll leave with a clear, actionable understanding of your CRA obligations – and a view of compliance not as a burden imposed on open source, but as a chance to make it more secure and sustainable. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Joe Sandom1112:0000:15Funding the long tail of dependencies: who pays for Europe’s digital sovereignty?https://www.sfscon.it/talks/funding-the-long-tail-of-dependencies-who-pays-for-europes-digital-sovereignty/Open source gets funded in two ways. Companies give engineering time, which is typically self-interested and rarely distributes equally across the dependency tree. Cash donations offer broader reach, but the “net” funding a maintainer receives is often far lower than the “gross” amount pledged, eroded by fees. The Cyber Resilience Act marks a historic shift in EU law by acknowledging that a product’s dependencies are as security-critical as the product itself. But the dependency graphs map technical criticality, not the funding mismatch for the long tail. Even where that mismatch is visible, closing it means moving money across dozens of currencies to individual maintainers, since no EU software stack is fully EU-built; the most “sovereign” European systems have roots in the Americas, Asia, and beyond. This talk presents the research from the Funding Coordination team, based on interviews with public sector and corporate open source funders across Europe and beyond. It asks whether settlement rails built for transparency and minimal intermediaries, paired with a machine-readable dependency graph, could improve current funding systems for open source. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Sofia Sukhinina1112:2000:15Open Source Has a New Audience. Are We Ready?https://www.sfscon.it/talks/open-source-has-a-new-audience-are-we-ready/From technical pioneers to mainstream users: what changes for projects, organisations and policyFrom technical pioneers to mainstream users: what changes for projects, organisations and policyOpen source is no longer a niche for technical pioneers. With a $46 billion global market, 61% enterprise adoption and growing policy momentum across Europe, it has become mainstream infrastructure. But mainstream users bring mainstream expectations: polished interfaces, zero-friction onboarding, integrated experiences, and security by design backed by regulation like the CRA and NIS2. Are open source projects ready for this new audience? This talk looks at four landmark forks — Nextcloud vs ownCloud, OpenSearch vs Elasticsearch, OpenTofu vs Terraform, Valkey vs Redis — to show that funding alone never decided the winner. What mattered was customer business: the value delivered to citizens, customers and employees. The projects that listened to their users survived. The ones that didn’t were acquired, forked or abandoned. With this new audience in mind, the talk proposes two paths forward. First, designing for customers: applying proven product doctrines — Customer Discovery, Lean MVP, UX-first development and open governance as a validation channel. Second, supporting customers inside open source through OSPOs, turning scattered adoption into a structured, governed capability. The talk closes with three policy actions Europe can take to support this shift: simplifying procurement for open source models, funding OSPOs in public administrations and SMEs, and using CRA Attestations as a trust-building mechanism that also drives sustainable funding. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Valentina Del Prete1112:4000:15Interoperability as a regulatory toolhttps://www.sfscon.it/talks/interoperability-as-a-regulatory-tool/Access to hardware features and functionalities in accordance with EU interoperability regulationAccess to hardware features and functionalities in accordance with EU interoperability regulationDigital markets are characterised by the dominance of Big Tech and their closed ecosystems, in which only a few major players control both the hardware and software layers of technological infrastructure. A key regulation of the Digital Markets Act (DMA) is the designation of Big Tech companies as so-called gatekeepers. This talk aims to clarify what the gatekeepers’ obligation to grant interoperability to third-party providers means. It will demonstrate how interoperability is no longer just a technical ideal but a legal requirement for gatekeepers operating in the European Single Market. No prior knowledge of legal topics is required. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Frithjof Goettmann1113:0000:15Power to the OSPOshttps://www.sfscon.it/talks/power-to-the-ospos/Turning Open Source Sovereignty from Strategy into PracticeTurning Open Source Sovereignty from Strategy into PracticeEurope has written the words. The Tech Sovereignty Package and the new Open Source Strategy say that dependence on a few non-European platforms is a strategic risk. Fine. But words are cheap. Italy turned “open source first” into a legal procurement principle in 2012 through Article 68 of the Digital Administration Code, AgID guidelines and mandatory reuse. More than a decade later it has produced no concrete effect and has been systematically circumvented. A law without budget, mandate and teeth becomes compliance theatre. This talk argues that open source will not free Europe from Big Tech through policy statements or goodwill. It needs a counter-power inside the institutions and companies that buy technology: the OSPO, the Open Source Programme Office, the team that decides how an organisation adopts, contributes to and governs open source. What I want to show is how much an OSPO is really responsible for once you take it seriously. Not just licences and developer support, but knowing which proprietary dependencies trap the organisation, having a voice before procurement signs a renewal, owning contribution and maintenance, and now carrying real legal exposure, since from 2027 the Cyber Resilience Act turns the software you ship into a compliance duty. None of this is a technical detail. These are strategic and financial risks, which is exactly why an OSPO buried inside the IT department cannot do the job. It has to be empowered and wired directly to the C-level, so that open source becomes a leadership decision about autonomy and risk, not a licensing chore handed down to engineers. That shift, from technical housekeeping to strategic mandate, is what turns sovereignty from a slogan into something that actually changes what gets bought. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Stefano Pampaloni1113:2000:35FOSS and Tech Sovereign in the EU: a view from the trencheshttps://www.sfscon.it/talks/foss-and-tech-sovereign-in-the-eu-a-view-from-the-trenches/In this talk, OSI EU Policy Analyst, Jordan Maris will share how current events are shaping the debate around tech sovereignty in the EU, where Open Source is already delivering tech sovereignty, and what is still needed. He’ll tell unique stories from on the ground, share the OSI’s and thoughts on the EU’s recently released Tech Sovereignty Package, and highlight why Tech Sovereignty isn’t just for governments. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Jordan Maris1114:0000:15Open Source as an economic governance strategyhttps://www.sfscon.it/talks/open-source-as-an-economic-governance-strategy/Open Source is a licensing model, a governance model for communities and projects, and a business model. But can the EU, and can governments use Open Source as a strategy for economic actors? The EU and governments seem to have understood the benefits of Open Source in the context of increasing digital sovereignty. Public administrations make use of “public money, public code” approaches to make investments more sustainable, transparent, and inclusive. But there is a third way pilar to making Open Source beneficial for our societies, and that is to make economic actors cooperate on the basis of Open Source. Open Source as an economic governance model is particularly effective in allowing companies to cooperate on projects where they otherwise would be competitors, across regions and borders. This talk explores the third pillar of Open Source, the cooperation between businesses, and in public-private partnerships on the basis of openness. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Sebastian Raible1114:2000:15CRA Compliance Is Not a Checklist: Lessons from the Field on Software Sovereigntyhttps://www.sfscon.it/talks/cra-compliance-is-not-a-checklist-lessons-from-the-field-on-software-sovereignty/Why control over your software supply chain is becoming the core challenge of complianceWhy control over your software supply chain is becoming the core challenge of complianceAs the Cyber Resilience Act (CRA) moves from awareness to implementation, a deeper challenge is becoming clear: compliance is not just a regulatory exercise, but a question of software sovereignty. Many organizations still approach CRA as a tooling or certification problem. In practice, the main barriers are structural. Companies lack full visibility into their software supply chains, underestimate the complexity of their dependencies, and rely on standards that do not fully address regulatory accountability. This talk shares practical insights from working with industry stakeholders navigating CRA readiness today. It highlights persistent challenges such as delayed prioritization, limited supplier awareness, and unclear ownership of responsibilities across organizations. More importantly, it reframes CRA compliance as an issue of control. Who is responsible for vulnerabilities? Who owns the risk across the supply chain? And how can organizations maintain oversight in ecosystems built on open technologies? The session also explores how collaborative initiatives such as the Eclipse Foundation’s Open Regulatory Compliance Working Group are helping companies move toward shared approaches, reducing fragmentation while strengthening collective control over compliance processes. Attendees will gain a clearer understanding of how sovereignty, not just compliance, is becoming a defining factor in the future of software development. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Francisco Carneiro1114:4000:15ARCADIA: Towards an Open and Sovereign AI Toolbox for Knowledge Valorisationhttps://www.sfscon.it/talks/arcadia-towards-an-open-and-sovereign-ai-toolbox-for-knowledge-valorisation/ARCADIA is a new Horizon Europe project that aims to develop an AI-powered toolbox supporting researchers, universities and knowledge valorisation professionals throughout the innovation journey—from opportunity discovery to intellectual property management and stakeholder engagement. While the project is still in its early stages, one of its guiding principles is clear: openness should not be an afterthought but a design choice. This talk introduces the vision behind ARCADIA and discusses how the project intends to embrace Open Source technologies, open standards and interoperable architectures wherever possible, fostering transparency, collaboration and long-term sustainability. It will also reflect on the importance of digital sovereignty in the European research ecosystem, exploring how publicly funded AI solutions can reduce technological dependencies while remaining trustworthy, reusable and aligned with European values. Rather than presenting final results, this session invites the Free Software community to engage with the project from the beginning, sharing ideas and perspectives on how AI tools for research and innovation can be built in a way that is open, responsible and beneficial for the wider community Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Elisa Morganti1115:0000:15EuroCommons: migrating together towards open source digital commonshttps://www.sfscon.it/talks/eurocommons-migrating-together-towards-open-source-digital-commons/A European initiative backed and funded by the French Economic Developement bank "Caisse des Dépôts"A European initiative backed and funded by the French Economic Developement bank "Caisse des Dépôts"EuroCommons, a European programme launched by Caisse des Dépôts (CDC), part of Horizon Numérique 2030 (€18 billion strategic envelope) is a call to large European organisations and to open source vendors to join collective working groups, coalitions, to plan, prepare, finance and execute concrete migrations away from proprietary software dependencies. Together. In project mode. With peers facing the same problems. EuroCommons will bring together CIOs, open-source software vendors, foundations and public institutions from across Europe to formalise the first cross-border coalitions around shared software components and commit to migrate together. Eleven coalitions cover the critical digital components: Digital Workplace, Communications, Business Applications, IT & Project Management Tools, Virtualization, Cloud / Data / AI, Security, Directories / Identity / SSO, Desktop OS, Mobile OS & App Store, Geographic Information & Spatial Data. To establish and foster FOSS community trust, CDC has joined OW2 as a strategic member. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Pierre-Yves Gibello1115:2000:35Panel Policy & CompliancePolicy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

0016:0000:15An Ethical Case for Open Technologyhttps://www.sfscon.it/talks/an-ethical-case-for-open-technology/The tech bros’ world viewsThe tech bros’ world viewsOne of the key insights from the latest wave of technological innovation – particularly in AI – is that technologies are not simply “neutral.” Instead, they are intrinsically intertwined with ideological, political, moral, economic, and even religious contexts. It is therefore no surprise that, in recent years, the leaders of major tech corporation – the tech bros – have focused less on technology-specific questions and more on publishing manifestos, worldviews, and global visions of the future. In order to assess the significance and scope of digital sovereignty, as well as the potential of open and free technologies, we analyze a representative selection of the tech bros’ worldviews which could hardly be more diverse: Mustafa Suleyman’s (Microsoft) serious attempt to address an unprecedented dilemma of technological progress in his bestseller The Coming Wave; Dario Amodei’s (Anthropic) mainstream-oriented, welfare-economic manifesto Machines of Loving Grace; Mark Zuckerberg’s (Meta) grandly failed Metaverse project, which was meant to define nothing less than a new era of global communication; and, finally, the IPO of Elon Musk’s SpaceX in June 2026, whose official prospectus presented a potpourri of science-fiction tropes, detached from both economic realities and current technical feasibility. From an ethical perspective, it is crucial to recognize that the overburdening of technology with economic, ideological, or fictional narratives obscures what technology itself is and can be. Leaving technological development as well as reflections on technology itself solely to the Tech Bros, therefore, does not seem conducive to fostering genuine autonomy and sovereignty. Additionally, the obvious political interests, particularly those of the US and China, hamper the pursuit of technological sovereignity. Against this backdrop, we argue that the approach of free and open software and technology concepts represents a far more favorable breeding ground for the self-determined development of a diverse and sustainable technological landscape. Hence, we raise the question of what concrete steps and measures are meaningful and which policies need to be adopted, in order to support this development and we give a brief assessment of existing European initiatives, like Gaia-X or EuroStack, or Tech Sovereignity Package proposed by the EU-Commission. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Robert Simon1116:2000:15Introducing CodeSupplyhttps://www.sfscon.it/talks/introducing-codesupply/Open data, decentralized and distributed access, federated trust: how CodeSupply rethinks software supply chains metadataOpen data, decentralized and distributed access, federated trust: how CodeSupply rethinks software supply chains metadataSoftware supply chains security depends on accurate metadata: origin, licensing, vulnerabilities, and project health. That metadata is scattered across package registries and vulnerability databases, often conflicting or outdated, and stored in multiple incompatible data formats. Proprietary systems provide no reproducibility, traceability, or auditability, even as regulators and security teams demand exactly that, and development teams require actionable, accurate, and current data to improve their security posture. CodeSupply is a new Horizon Europe project building an open, federated data catalog to fix this with accessible access to current, correct, comprehensive metadata. Using PURLs (Package-URLs) as universal identifiers for software packages, it aggregates and curates software metadata from distributed, authoritative sources across heterogeneous ecosystems, assembles it into reference, curated data sets, and federates it in a unified and decentralized catalog distributed as open data under an open source license. This talk introduces CodeSupply’s architecture and open data sources, and how it fits into existing software supply chain security and compliance workflows, that are primarily FOSS-centric. You’ll leave knowing how to plug into the project, whether as a data consumer, a contributor, a curator, or an organization looking for digital sovereignty over its own supply chains metadata. BENEFITS TO THE ECOSYSTEM * Unified, distributed and open source of truth for metadata Understanding how CodeSupply aims to consolidate scattered, conflicting origin, license, vulnerability, and project-health data into one distributed and federated catalog, reducing the duplicated effort of cross-referencing multiple registries, duplicating scans, and code reviews to ensure we can all share and access effective origin, license, security, and health open data for all FOSS projects. * Auditability over black-box, proprietary data Learning how an open, traceable, reproducible approach to software supply chains data addresses the core weakness of proprietary systems, with an opaque approach to curation. This is relevant for any team facing regulatory requirements (like CRA) that demand verifiable, curated compliance data to assess supply chain risks such as exploited vulnerabilities in under 24 hours. When automation is the only way, then correct, open data is the way towards enabling this automation. * PURL-based supply chains interoperability Understanding how using Package-URLs as universal identifiers lets CodeSupply tie into the broader standards ecosystem (ScanCode, ClearlyDefined, vulnerability databases) rather than creating yet another siloed format, when PURL is emerging as the leading identifiers across CVE, OSV, CSAF, CycloneDX, SPDX and OpenVEX formats for vulnerabilities, SBOMs, and VEXs reporting. * Open independence enabling digital sovereignty Understanding the project’s relevance to teams, organizations and public bodies wanting independence from proprietary, closed data providers for supply chains metadata, is a growing concern for EU-based entities, and CodeSupply is an answer to this concern. * Clear and multiple engagement modes Leave knowing concretely how to engage with the CodeSupply project as a data consumer, data curator, code contributor, or sponsoring organization. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Henrik SandklefPhilippe Ombredanne1116:4000:15SBOM management with Eclipse Disukohttps://www.sfscon.it/talks/sbom-management-with-eclipse-disuko/Complex products like vehicles consist of many major components that are built using a open source software. As a manufacturer we need to comply with the licenses of all embedded open source software elements. For this purpose we have recently published Eclipse Disuko. Eclipse Disuko focuses on consuming SBOMs and the resulting actions based on their assessment. It is not meant to produce SBOMs from source code since that is well-supported already by other projects. SBOMs might be created by combining imported SBOMs though. Eclipse Disuko comes with a license and policy database that helps a project owner to assess FOSS licenses based on self-configured use cases and understand the resulting license obligations. The talk outlines and demos Eclipse Disuko with the goal to attract further users and contributions. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Christian Wege1117:0000:35From Legal Headache to Automated Sanity Check: Open Source Compliance Made Simplehttps://www.sfscon.it/talks/from-legal-headache-to-automated-sanity-check-open-source-compliance-made-simple/A tool-driven approach to tracking compliance, catching library conflicts, and choosing the right outbound license.A tool-driven approach to tracking compliance, catching library conflicts, and choosing the right outbound license.Adding Open Source bits to your app is usually a piece of cake, but making sure you don’t step on any license landmines? That’s a whole different story. Depending on how you’re using the code (SaaS vs. shipping a binary, tweaking it or just pulling it in, mixing it with other libraries, etc.) different license conditions may kick in. Sometimes those conditions clash, and suddenly your whole distribution could be illegal, especially when you’ve got dozens – or even hundreds – of Open Source components tucked into the build. We’ve built a simple license-model language that boils down the crazy variety of Open Source licenses into a handful of clear‑cut, machine-computable permissions and conditions. On top of that we’ve made a tool that asks you what you actually need (e.g., can you ship source? Can you disclose changes? …), then spits out: * a to‑do checklist of the license conditions you must meet, * a warning about any possible license incompatibilities among the components you picked, and * a list of sensible outbound‑license options for your final product. In short, we turn the headache of Open Source compliance into a quick, automated sanity check, so you can keep building without worrying about walking on a legal minefield. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Henrik SandklefAlberto PianonCarlo Piana1117:4000:15DMA Interoperability: What It Means for Free Softwarehttps://www.sfscon.it/talks/dma-interoperability-what-it-means-for-free-software/Where are we after two years of enforcement?Where are we after two years of enforcement?The EU Digital Markets Act (DMA) promises open digital markets by forcing gatekeepers like Apple and Google to grant “free and effective access to hardware and software features”, a direct opening for Free Software developers to build real alternatives. But two years into implementation, the gap between the law’s promise and what developers actually experience on the ground is striking. How is interoperability actually being granted, and on what terms? While Apple’s public tracker lists nearly fifty interoperability requests, the real story lies in what gets denied, delayed, or buried in conditions that lock out Free Software projects. This talk presents what FSFE has learned from two years of monitoring DMA implementation from a Free Software perspective. We will share findings from our developer survey, walk through Apple’s and Google’s gatekeeping practices, and explain what FSFE is doing , including in court, to push back. Specifically, we will: – Walk through Apple’s 40+ interoperability requests and what they mean for Free Software: which doors are opening, which are being slammed shut, and why the denials matter most. – Examine Google’s moves on Android: changes to AOSP source availability, tightening interoperability documentation, and side-loading restrictions that threaten alternative app ecosystems and distribution channels. – Report on FSFE’s litigation against Apple under DMA Art. 6(7), and what it means for the broader fight for software freedom on locked platforms. – Present findings from the FSFE DMA Interoperability Survey, which collects structured, first-hand accounts from Free Software developers trying to access interoperability. Policy & CompliancePolicy & CompliancePolicy & Compliance

https://www.sfscon.it/tracks/policy-compliance-track-2026/

Dario PresuttiLucas Lasota1108:3000:50Welcome BreakfastSFSCONSFSCONSFSCON0009:2000:35Digital Independence Day – a path towards digital sovereigntyhttps://www.sfscon.it/talks/digital-independence-day-a-path-towards-digital-sovereignty/The talk presents the Digital Independence Day initiative, born in Germany and is now being implemented in Italy. DID aims to build a community of activists ready to help anyone who wants to be free from Big Tech dominance through “hands-on” meetups, where people gather and start removing proprietary software to switch to independent, free, ethical alternatives. In addition, the talk outlines the current state of digital sovereignty in Europe and the political and regulatory initiatives to achieve it. Digital Independence DayDigital Independence DayDigital Independence Day

https://www.sfscon.it/tracks/digital-independence-day-2026/

Nino Gualdoni1110:0000:15Digital Independence Day: Building a global Movement for Resilience and Sovereignityhttps://www.sfscon.it/talks/digital-independence-day-building-a-global-movement-for-resilience-and-sovereignity/Digital Independence Day (DI.DAY) is a fast growing initiative that empowers individuals and organizations to reclaim digital freedom by switching to democracy-friendly, open-source alternatives. This talk explores how DI.DAY started in Germany in January and quickly spread across the country and beyond. DI.DAY combines practical change recipes (step-by-step guides for migrating away from Big Tech), real-life community events, and networking opportunities to drive mass adoption of Free Software. With hundreds of events organized monthly across Germany, but also in Portugal, Ireland, USA or Italy, and over 100,000 website visitors per month, DI.DAY demonstrates the scalable impact of grassroots advocacy – with a difficult line to walk: How “pure” should a recipe be (with regards to using FOSS, based in Europe etc.), but how practical and easy to follow as well to reach a wider audience? DI.DAY contributes to democratizing access to open, ethical alternatives through actionable resources and community engagement. It concentrates on user-friendly advocacy—paired with tangible switch recipes and local events—to accelerate the shift toward open technologies at scale. Digital Independence DayDigital Independence DayDigital Independence Day

https://www.sfscon.it/tracks/digital-independence-day-2026/

Björn Staschen1110:2000:15Reclaiming Our Choice: Why Digital Independence Demands Open Standards and Accessibility in FOSShttps://www.sfscon.it/talks/reclaiming-our-choice-why-digital-independence-demands-open-standards-and-accessibility-in-foss/Digital independence is a luxury precisely for those who need it most.Digital independence is a luxury precisely for those who need it most.A blind user has few choices for their digital independence; a sighted person can explore dozens of Linux distros or cheap smartphones, while blind and visually impaired mostly rely on commercial tech with accessibility on board or external softwares, but there is no ethical reason why most proprietary operating systems are accessible: American Disability Act (ADA) and federal USA laws for universities have made accessibility mandatory through the years but until when? While Android faces growing restrictions and complex de-googling processes, iOS remains the most usable and accessible mobile platform for the visually impaired. Yet, even this “gold standard” is no longer a certainty, as on-board accessibility features become less stable and increasingly dependent on unpredictable, cloud-based corporate ecosystems. Relying on closed software means accepting that your autonomy can be limited or altered overnight by a single update or even political decision. True, sustainable independence cannot be built on proprietary quicksand or AI fast accessibility illusion; that’s temporary. It must rely on open, decentralized standards like RSS and Markdown, alongside robust local FOSS tools like the NVDA screen reader on Windows – at least until we’re allowed to install our own software there. Fediverse and open web are the key, accessibility and ethics should be built by community for community, not counting on a State or a corporation who thinks of us. This talk is a direct reality check for the free software community. We need to bridge the gap between blind users and open-source developers, moving past toxic polarization, guilt-tripping, or elitist “fix it yourself” mentalities. Not every blind user is a developer, but every blind user deserves access. This session calls for blind and sighted tech enthusiasts to sit at the same table and co-design a truly open future, ensuring that digital freedom is built on stable, accessible foundations for everyone. Digital Independence DayDigital Independence DayDigital Independence Day

https://www.sfscon.it/tracks/digital-independence-day-2026/

Elena Brescacin1111:0000:15‘Looking For Freedom’: KDE’s Outreach To Institutions In The Age Of Digital Sovereigntyhttps://www.sfscon.it/talks/looking-for-freedom-kdes-outreach-to-institutions-in-the-age-of-digital-sovereignty/European organizations are increasingly talking about digital sovereignty”, and some institutions have begun the migration away from Big Tech software. This historic turn toward digital sovereignty has given Free Software a unique opportunity, but it also comes with risks. Will procurement budgets end up supporting long-standing Free Software projects, or the opportunists capitalizing on new business opportunities? Might Free Software be passed over for proprietary Europe-first products? How can Free Software communities like KDE reach the institutions looking for digital independence? In this talk I will present the work KDE e.V. is doing to reach public institutions. Which opportunities and risks have we identified? What services do we see as necessary for institutional change and how can our community provide them? And in the current political climate, how do we present such a diverse and international community as KDE to nationally-oriented politicians? This presentation will provide input for a larger discussion about Free Software in the context of political rhetoric around digital sovereignty. Digital Independence DayDigital Independence DayDigital Independence Day

https://www.sfscon.it/tracks/digital-independence-day-2026/

Joseph P. De Veaugh-Geiss1111:2000:15Towards Digital Autonomy in South Tyrolhttps://www.sfscon.it/talks/towards-digital-autonomy-in-south-tyrol/The FUSS Project ContributionThe FUSS Project ContributionThe first step towards greater Digital Independence in South Tyrol took place in 2004 with eSüdtirol 2004-2008, the Action plan for the development of the information society in South Tyrol. Chapter 3 was dedicated to promoting Free Software in public administrations, schools, and for fostering innovation in the economic sector with the support of a Competence Center. In this talk I show how the FUSS Project contributed to achieve these goals in public schools by bringing Free Software into classrooms, building local expertise, and helping turn our digital autonomy from a policy goal into a practical reality. Digital Independence DayDigital Independence DayDigital Independence Day

https://www.sfscon.it/tracks/digital-independence-day-2026/

Paolo Dongilli1111:4000:15Own your stack: 15 years delivering software freedom to real companies, from startups to enterpriseshttps://www.sfscon.it/talks/own-your-stack-15-years-delivering-software-freedom-to-real-companies-from-startups-to-enterprises/Continuity from the community: interoperability, security, digital sovereignty, and technological independenceContinuity from the community: interoperability, security, digital sovereignty, and technological independenceFree software is a suitable choice for organizations of all kinds, and we can demonstrate its value through our direct experience. We will look at how, over the past 15 years, different organizations—from small to large—have successfully adopted FLOSS solutions. Participants will be able to recognize themselves in similar contexts and see how needs and challenges have been met and resolved. We will start from the basic infrastructure, with solutions for data sharing, collaboration, and the management of business services through tools such as Nextcloud, Univention, NethServer, file servers based on ZFS, Proxmox, and systems for secure credential management such as Vaultwarden. We will then explore topics related to reliability and operational continuity, including disaster recovery, backups, remote data replication, high-availability clusters, and firewalls based on OPNsense. Finally, we will see how free software enables the creation of advanced and scalable infrastructures thanks to technologies such as FreeIPA, Wazuh, Zabbix, Ceph, CephFS, and systems dedicated to email protection. We chose open-source software because it enabled us to deliver projects that would have been more complex, more constrained, or simply impossible to achieve with proprietary solutions. We have continued to use it because we have fallen in love with many of the communities behind it. Continuity is community: we will also share our wonderful experience with various communities, such as our long-standing relationship with PNlug, with whom we have shared—and continue to share—battles and projects. Digital Independence DayDigital Independence DayDigital Independence Day

https://www.sfscon.it/tracks/digital-independence-day-2026/

Giorgio FavaroMauro Biasutti1112:4000:15Beyond Big Tech: how to reach people who don’t care about FOSShttps://www.sfscon.it/talks/beyond-big-tech-how-to-reach-people-who-dont-care-about-foss/From personal project to community resourceFrom personal project to community resourcelealternative.net started as a personal blog and grew into one of the most comprehensive Italian-language catalogs of alternatives to mainstream software covering tools that range from fully open source to “good enough to matter.” This talk traces that evolution and the editorial choices behind it: why pragmatism beats purity when you’re trying to reach people who have never heard of FOSS, how years of reviewing and analyzing alternatives shaped a clearer language, a more accessible interface and a growing attention to the details that make the difference for a non-technical audience, and what six years of doing this have taught about communicating digital autonomy to people who are simply curious and not necessarily technical. Digital Independence DayDigital Independence DayDigital Independence Day

https://www.sfscon.it/tracks/digital-independence-day-2026/

Andrea Guani1113:0000:15Only digital sovereignty will save Europe from the America-China dualismhttps://www.sfscon.it/talks/only-digital-sovereignty-will-save-europe-from-the-america-china-dualism/In the last twenty years, we have seen a profound shift in global power. Military strength still matters, but the real battleground today is digital: data, platforms, cloud infrastructures, AI models, and the rules that govern them. Right now, this battleground is dominated by two giants: the United States and China. On one side, a handful of American tech corporations shape much of our online life. On the other, a powerful Chinese state–tech ecosystem is expanding its reach across infrastructure, hardware, and platforms. Europe is the world’s largest democratic market, but digitally it often behaves like a user, not a player. We consume American platforms, we buy Chinese hardware, and we negotiate from a position of dependency rather than autonomy. If this continues, Europe will not simply “lag behind”. It will lose the practical ability to decide how its societies function in the digital age. Digital Independence DayDigital Independence DayDigital Independence Day

https://www.sfscon.it/tracks/digital-independence-day-2026/

Luca Termite1113:2000:15Liberate your desktop, liberate your documentshttps://www.sfscon.it/talks/liberate-your-desktop-liberate-your-documents/Survival tips from a Linux user in a Microsoft dominated industrySurvival tips from a Linux user in a Microsoft dominated industryFor over 20 years, I have managed to work in the Microsoft dominated public relations industry without a single colleague, client or journalist realizing that I am using only free open source software on the desktop. During that timeframe I have exchanged thousands of text documents, spreadsheets and presentations without a single issue. Digital Independence DayDigital Independence DayDigital Independence Day

https://www.sfscon.it/tracks/digital-independence-day-2026/

Italo Vignoli1113:4000:15NextRedOS: Building a Hardened Debian-Based Linux Distribution for European Digital Independencehttps://www.sfscon.it/talks/nextredos-building-a-hardened-debian-based-linux-distribution-for-european-digital-independence/NextRedOS is a Debian-based Linux distribution created as a practical experiment in system hardening, self-hosting and European-oriented open technologies. The project started from a simple need: deploying a secure and reliable operating system for a home server running Home Assistant and other self-hosted services. This talk presents the ideas behind NextRedOS and the concrete security measures adopted to reduce the attack surface: full-disk encryption with LUKS, Secure Boot with custom MOK enrollment, USBGuard and PAM hardening, kernel module blacklisting, and a multi-layer ransomware protection approach combining auditd rules, Timeshift snapshots and filesystem restrictions on temporary directories. It will also cover how self-hosted services such as Home Assistant and Vaultwarden are integrated through Podman, keeping containers isolated from the base system. Through the development of this distribution, NextRedOS became both a learning platform for cybersecurity and networking and a real-world operating system designed for everyday use, with a GPG-signed, SHA256-verified release already published on GitHub and Codeberg. The session will share lessons learned, challenges encountered during the hardening process, and future goals, offering attendees a perspective on building secure Linux systems with open technologies. Digital Independence DayDigital Independence DayDigital Independence Day

https://www.sfscon.it/tracks/digital-independence-day-2026/

Mirco Montinari1114:0000:35MXMap Italia: Observatory of Digital Sovereignty of Italian Public Agencieshttps://www.sfscon.it/talks/mxmap-italia-observatory-of-digital-sovereignty-of-italian-public-agencies/Monitoring Italian Public Agencies Digital Sovereignty Status trough observation of Email System DNS RecordMonitoring Italian Public Agencies Digital Sovereignty Status trough observation of Email System DNS RecordMXMap Italia has been established as national observatory of the status of digital Sovereignty of italian public agencies, by analyzing DNS records on the basis of the MXMap software. The technical platform provides opendata, mapping and analytics of the 23k+ public agencies on https://mxmap.it . The alarming results, with more than 46% of italian government agencies accessible trough Cloud Act by USA Security Agencies, requires all the institutional, technical, activist world to act now! The political platform https://osservatorio.mxmap.it aim to provides awareness across all the stakeholders, with policy and technical papers, brief for decision makers, and periodic massive emailing campaign to inform the key persons at each agency (DPO, but also RTD, owner of digital transformation). This walk want to share the project results, methodology and a call to action to contribute to the project goals, improving the status of digital sovereignty with self-hosted infrastructure and free software. Digital Independence DayDigital Independence DayDigital Independence Day

https://www.sfscon.it/tracks/digital-independence-day-2026/

Fabio Pietrosanti (naif)1114:4000:15Beyond Cross-Posting: Strategies for Migrating Communities from Big Tech Social Networks to the Fediversehttps://www.sfscon.it/talks/beyond-cross-posting-strategies-for-migrating-communities-from-big-tech-social-networks-to-the-fediverse/Many people join the Fediverse, but relatively few successfully move their communities there.Many people join the Fediverse, but relatively few successfully move their communities there.After years of running and moderating large Fediverse communities in Italy, we have experimented with different migration strategies: from simple cross-posting to content adaptation, follower migration campaigns, curated account collections, onboarding techniques, and community-building initiatives. This talk presents practical lessons learned from real-world experiments involving thousands of users. Rather than promoting an “all or nothing” migration, it explores how creators, organizations and activists can reduce their dependence on proprietary social networks while continuing to reach their existing audiences. Topics include: Why people fail when they simply duplicate content. How to maximize engagement on the Fediverse. Techniques to encourage followers to migrate. Living in both worlds: coexistence instead of abrupt abandonment. Building resilient communities independent of Big Tech algorithms. Measuring success beyond follower counts. The goal is to provide attendees with concrete strategies they can immediately apply to grow their presence on the Fediverse while minimizing the risks of leaving traditional social media. Audience: Content creators, Free Software communities, NGOs, developers, journalists, public administrations, and anyone interested in digital sovereignty and community building. Takeaways: Participants will leave with practical techniques for: increasing engagement on the Fediverse; designing a gradual migration strategy; reducing dependence on algorithmic platforms; preserving and growing their online communities during the transition. Digital Independence DayDigital Independence DayDigital Independence Day

https://www.sfscon.it/tracks/digital-independence-day-2026/

Filippo Della Bianca1109:2000:35Your Code of Conduct Is Killing Your Community (And What to Build Instead)https://www.sfscon.it/talks/your-code-of-conduct-is-killing-your-community-and-what-to-build-instead/Almost every open source project has a Code of Conduct, and most of us hope it’s doing more work than it actually is. In this follow-up to “Building a No-Drama Community” (SFSCON 2024), we’ll look at what tends to happen after a project adopts a CoC, why the document on its own can only carry so much weight, and how the goodwill a project has spent years building can be worn down surprisingly quickly when someone learns to work the process from the outside. Drawing on five years of running the AlmaLinux OS Foundation and two decades in open source communities, this talk walks through some of the common ways CoCs fall short in practice (adopted without a clear enforcement path, leaned on as legal cover, or used against the very people they were meant to protect), the patterns bad actors tend to rely on, and the structural foundation that makes a CoC meaningful: distributed authority, documented process, transparent decisions, and the organizational will to actually follow through. Aimed at community leaders, project maintainers, and foundation directors who have a Code of Conduct in place and are quietly wondering whether it’s doing enough. Practical, honest, and grounded in what has worked and what hasn’t. CommunityCommunityCommunity

https://www.sfscon.it/tracks/community-track-2026/

benny Vasquez1110:0000:15Open Source, Open Doors: Building Inclusive Communities with Pythonhttps://www.sfscon.it/talks/open-source-open-doors-building-inclusive-communities-with-python/An Invitation to PyLadies South-TyrolAn Invitation to PyLadies South-TyrolPyLadies is a global community that supports women and other marginalized genders in becoming active participants and leaders in the Python open-source ecosystem. Through local chapters, meetups, workshops, mentorship, and community events, PyLadies creates welcoming spaces where people can learn, share knowledge, build confidence, and contribute to technology together. Check for more infos: pyladies.com. This talk introduces the idea of starting a PyLadies chapter in South Tyrol. At the moment, the nearest chapters are in Munich and Vienna, leaving Bolzano and the wider South Tyrol region without a local PyLadies community. Yet South Tyrol has a growing technology, research, education, and innovation landscape, with many people who could benefit from a supportive, inclusive, and international network focused on Python, open source, and community leadership. The session will present what PyLadies is, how the global community works, and why local chapters matter. It will explain how PyLadies chapters can support beginners, students, researchers, developers, educators, career changers, and experienced professionals alike. A chapter can offer many formats: informal meetups, coding sessions, talks, study groups, contribution sprints, mentoring circles, collaborations with universities and research institutions, and events connected to the wider Python and FOSS communities. The goal of this talk is not only to inform, but also to invite. It is an open call for people in South Tyrol who are interested in creating a local chapter together. Starting a chapter does not require being an expert Python developer or an experienced organizer. It requires curiosity, care, collaboration, and a shared wish to make the local tech community more accessible and inclusive. Let’s take these first practical steps together. Let’s discuss how you want to see the local FOSS and Python community develope. I want to encourage people from marginalized groups, as well as allies, to help build a welcoming community where more people can learn Python, contribute to open source, and grow into leadership roles within our tech communities. South Tyrol has the people, the institutions, and the energy. What is missing is a local chapter. Let’s start one together. CommunityCommunityCommunity

https://www.sfscon.it/tracks/community-track-2026/

Heike Quosdorf1110:2000:15Don’t Be That Guy: packaging community values like a MANIFEST.inhttps://www.sfscon.it/talks/dont-be-that-guy-packaging-community-values-like-a-manifest-in/In Free and Open Source Software communities, collaboration often breaks down not because of code, but because of unclear expectations. What does “good behaviour” mean in a distributed project? How do we agree on communication norms, decision-making, ownership, or feedback when contributors come from different cultures, time zones, and backgrounds? Inspired by MANIFEST.in and the spirit of the Agile Manifesto, our team experimented with treating community values like something you “package” explicitly: not left implicit, but made visible, shared, and versioned. We created a manifesto using virtual whiteboards, voting and a pinch of AI, we aligned perspectives on what truly mattered day-to-day. The manifesto became more than a guide: it is our north star. It clarified expectations, strengthened cohesion and identity, and helped prevent “that guy” behaviors. It also normalized error culture and generated positive side effects, like virtual coffee, cat photos and manicaretti in chat, making work lighter and more inclusive. This talk shares the real process of building that manifesto, what worked, what didn’t, and how similar approaches can help open-source projects and distributed communities build stronger, more inclusive and more resilient collaboration cultures. CommunityCommunityCommunity

https://www.sfscon.it/tracks/community-track-2026/

Giulia Scalaberni1110:4000:15Be a FOSS indie podcasterhttps://www.sfscon.it/talks/be-a-foss-indie-podcaster/Real cases:Real cases:A case study: the independent podcast Choosing to be an independent podcaster can come from many different reasons. But independent from whom? And from what? While digital sovereignty is a topic that engages policymakers at both the national and European levels, this talk focuses on podcasters who have chosen not to rely on production companies and who build their audience one listener at a time using only free, open, and federated tools and apps. This is the story of Io Hacker and LINKioni, and of the platforms that make their independence possible. Let’s see what is great and what is not working in these experiences. CommunityCommunityCommunity

https://www.sfscon.it/tracks/community-track-2026/

Valentina Nardecchia1111:0000:15Let’s toot together! The Fediverse for outreach for organizationshttps://www.sfscon.it/talks/lets-toot-together-the-fediverse-for-outreach-for-organizations/Join Tobias and Ana for a conversation about the Fediverse and how the FSFE uses it. Tobias will provide a recap of what the Fediverse is and why the concept of ‘reach’ on Big Tech platforms is misleading. Then, Ana will explain which social media channels the FSFE uses and why the Fediverse is our main tool. Come along to find out why decentralised tools are essential for achieving digital sovereignty in the social media sphere. CommunityCommunityCommunity

https://www.sfscon.it/tracks/community-track-2026/

Ana GalanTobias Diekershoff1111:2000:15Towards a Local-First Desktophttps://www.sfscon.it/talks/towards-a-local-first-desktop/Local-First sync and collaboration in GNOME apps, built on p2pandaLocal-First sync and collaboration in GNOME apps, built on p2pandaAn important area where free software apps currently lags behind their proprietary counterparts is sync and collaboration features. Whether it’s file sync, collaborative document editing, or system integrations like clipboard sync, we usually don’t have the kind of seamless cross-device sync that e.g. users of Apple platforms enjoy. One reason for this is that with the server-based sync architectures that have been popular for the past decades, the cost of a sync system scales up with the number of users. This makes it unsustainable for free-of-charge indie apps and desktop environments to implement. In recent years, advances in areas like CRDTs and peer-to-peer networking have made this more feasible though: Building on these, Local-First software is a new approach to sync and collaboration, which treats network as fully optional, but allows syncing when it is available. As Modal Collective, we’ve spent the past few years working towards making this a reality in GNOME apps using the p2panda framework for Local-First sync. In this talk, we’ll discuss our progress, next steps, and how you can get involved! CommunityCommunityCommunity

https://www.sfscon.it/tracks/community-track-2026/

Tobias BernardJulian Sparber1111:4000:15How to design successful campaigns that scalehttps://www.sfscon.it/talks/how-to-design-successful-campaigns-that-scale/Lessons from building a community-driven Free Software education campaign: from early ideas to school workshops, co-creation sessions, and the challenges of growing volunteer-led initiativesLessons from building a community-driven Free Software education campaign: from early ideas to school workshops, co-creation sessions, and the challenges of growing volunteer-led initiativesThis talk presents the evolution of a grassroots educational journey that started within the Free Software Foundation Europe and grew into a community-driven toolkit used in schools across different contexts. The talk will reflect on how community creativity, narrative tools, and structured support materials can work together to scale educational outreach, and how volunteer ecosystems can be empowered through shared ownership of educational content. CommunityCommunityCommunity

https://www.sfscon.it/tracks/community-track-2026/

Marta Andreoli1109:2000:35What the Docs Won’t Tell Youhttps://www.sfscon.it/talks/what-the-docs-wont-tell-you/A Practical Guide to Nextcloud App DevelopmentA Practical Guide to Nextcloud App DevelopmentNextcloud’s extensibility is core to its value as a sovereign, self-hosted platform; but the path from “I have an idea” to “I have a working app” isn’t always obvious. This talk is a code-first walkthrough of Nextcloud app development as it stands today: app architecture, the bootstrapping and dependency injection model, and a live demo built from the official app-tutorial example. We’ll also cover ExApps, Nextcloud’s newer model for running app logic outside the core PHP process. Expect concrete gotchas, debugging strategies, and a working mental model, plus a starting point you can clone and extend yourself. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Anna Larch1110:0000:15Open source doesn’t make money… until it doeshttps://www.sfscon.it/talks/open-source-doesnt-make-money-until-it-does/This talk explores how Free and Open Source Software creates value through community, trust and long-term adoption. We’ll look at how everyday developer decisions: choosing tools, contributing to projects or maintaining them over time, influence which technologies organisations come to rely on. These choices shape not only adoption, but also how much control and flexibility organisations retain over their systems. By focusing on real-world examples and ecosystem dynamics, we’ll highlight how open source enables more sustainable and transparent technology choices and why community-driven projects such as Payara Community, Jakarta EE and OpenJDK play a key role in building reliable, widely used infrastructure. Attendees will leave with a clearer understanding of how their day-to-day work contributes to the success of open technologies and how those decisions influence trust, adoption and long-term control over the systems their organisations depend on. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Dominika Tasarz-Sochacka1110:0000:15Grounding AI in reality via the semantic layer with Apache Ossie (incubating)https://www.sfscon.it/talks/grounding-ai-in-reality-via-the-semantic-layer-with-apache-ossie-incubating/Semantic drift is one of the most expensive problems in enterprise AI today — and most organizations don’t know they have it. It’s what happens when “revenue” means something slightly different in your CRM than your data warehouse than your BI tool than your AI agent’s system prompt: as the definition of “revenue” drifts, your business’ ability to measure it accurately evaporates. This is exacerbated in AI systems: as the semantic layer drifts, hallucinations increase and token – compute – costs spiral. What AI needs is a structured, consistently defined semantic layer on top of your data. The Apache Ossie (incubating) Apache 2.0, vendor-neutral YAML specification built in the open, with 50+ organizations collaborating to define metrics, dimensions, and AI context once, portably, across any tool. In this talk, we’ll go through the creation of the Ossie spec, how to get involved, and go through real-world examples of taming data for AI using Ossie. LLMLLMLLM

https://www.sfscon.it/tracks/llm-track-2026/

Celeste Horgan1110:2000:15Migrating to Autonomy: The Journey from VMware to Proxmox VE in the Loacker Casehttps://www.sfscon.it/talks/migrating-to-autonomy-the-journey-from-vmware-to-proxmox-ve-in-the-loacker-case/Interoperability, Security, Digital Sovereignty, and Technological IndependenceInteroperability, Security, Digital Sovereignty, and Technological IndependenceFor over ten years, we have been supporting companies and public administrations in the design of open-source infrastructures, systems integration, and digital sovereignty projects. As an ISO 27001, 27017, 27018, and ISO 9001 certified company, we see free software not only as a technological choice but as a sustainable model capable of generating autonomy, skills, and shared long-term value. The presentation will showcase real-world experiences of adopting open-source technologies in practical contexts, demonstrating how it is possible to reduce dependence on proprietary platforms while maintaining control over data, infrastructure, and processes. We will present one of the most significant projects with our client Loacker, where we are working on the migration of the virtualization infrastructure from VMware to Proxmox VE, a platform we have been partners with and users of for about fifteen years. The infrastructure has been integrated with open-source tools that support the implementation of the security measures required by the NIS2 Directive, including FreeIPA for centralized authentication and multi-factor authentication, Wazuh for SIEM and XDR, and Zabbix for system and service monitoring. The goal of the presentation is to share experiences, methodologies, and concrete results that demonstrate how digital sovereignty can be achieved through sustainable, secure, and already scalable open-source solutions. The session will also explore key topics for free software and open technologies, such as interoperability, security, and technological independence, highlighting approaches and practices that support their practical adoption. At Continuity S.r.l., we develop open-source digital solutions that enable companies and organizations to manage infrastructure and data autonomously, securely, and transparently, reducing dependence on proprietary platforms and external vendors. Our approach is based on modular and interoperable technologies designed to ensure operational continuity, control, and long-term resilience. The contribution we propose to SFSCON aims to share practical experiences, methodologies, and perspectives related to the adoption of free software as a strategic lever for building more open, ethical, and sustainable digital ecosystems. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Giorgio FavaroMauro Biasutti1110:4000:15Everything in the wokhttps://www.sfscon.it/talks/everything-in-the-wok/Wok is an opinionated CLI for managing projects, built to type less and navigate through projects fastWok is an opinionated CLI for managing projects, built to type less and navigate through projects fastI’d like to introduce Wok, a little Rust CLI I’ve been building to smooth out my own local workflow. In this short session, I’ll show how it fits into my day-to-day and the ideas behind it. It’s opinionated, ergonomic, and built to type as little as possible, with short commands and a consistent shape you can build muscle memory around, so the common things become almost automatic. If you juggle a lot of repositories or use git worktree extensively, I hope you’ll leave with an idea or two worth borrowing. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Emanuele De Cupis1111:0000:15Korvus: Building Your Own AI-Driven Root Cause Analysishttps://www.sfscon.it/talks/korvus-building-your-own-ai-driven-root-cause-analysis/Korvus is a proof of concept for AI-driven root cause analysis that you can build and run yourself. A pluggable harness gives an LLM tool-mediated access to your telemetry, with model choice and inference location left entirely up to you. Alertmanager or a human triggers it, producing a structured report for an on-call SRE under time pressure: what happened, blast radius, and ranked root causes with concrete fixes. Korvus is a proof of concept that an SRE team can design, build, and operate an AI-driven root cause analysis system using free and open source components end to end: Kubernetes for orchestration, an OpenTelemetry-compatible data path as input, and Alertmanager as the trigger. The same approach reaches below the cluster. The demo also traces faults into the underlying infrastructure layer through OpenNebula, showing that root causes in the base layer, not just the application or cluster layer, are within scope. The harness itself is model-agnostic and works equally with open-weight models such as Mistral, Llama, Qwen, or DeepSeek and with proprietary ones, with model choice and inference location left entirely up to the operator. The one hard requirement across every layer is GitOps over ClickOps. Because infrastructure and application state are versioned in git rather than changed by hand, the agent can navigate the relevant repositories directly and correlate a telemetry anomaly with the specific change that rolled it out, regardless of which layer it occurred at. The talk’s core argument is methodological: with clean telemetry and a clean, git-driven deployment process in place, the entire pipeline, from data access to inference to report generation, is something a team can implement, inspect, and run themselves, on infrastructure they control. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Felix HillebrandGerhard Sulzberger1111:2000:15Introducing the TTCN-3 Language and Eclipse TITANhttps://www.sfscon.it/talks/introducing-the-ttcn-3-language-and-eclipse-titan/A domain specific programming language for testing distributed systemsA domain specific programming language for testing distributed systemsDespite being around for about 25 years, almost noone – even people in software testing – know about the TTCN-3 domain specific programming language and its featureful and well-maintained Free / Open Source Software compiler/runtime called Eclipse TITAN. The language is an exremely powerful tool in testing comlpex networked/distributed systems by sending and receiving protocol messages between the test suite and the implementation under test. It features language constructs for common problems and disposese of powerful ways to match on abstract representations of incoming protocol messages as well as templates for outbound protocol messages. This talk is a very brief intro. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Harald Welte1111:4000:15Your Documents, Your Data: Integrating AI into Your Office Suite Without Surrendering Sovereigntyhttps://www.sfscon.it/talks/your-documents-your-data-integrating-ai-into-your-office-suite-without-surrendering-sovereignty/AI writing assistance, document summarisation, and intelligent formatting suggestions are no longer differentiators – users now expect them. But most implementations share a common architecture: your document content leaves your infrastructure and reaches a commercial API. For public administrations, healthcare bodies, and educational institutions handling sensitive data, that trade-off is often a hard no. The goal is not to resist AI, but to adopt it on terms that do not quietly undo the sovereignty work your organisation has already done. This session explores how Collabora Office has integrated meaningful AI capabilities, including writing assistance, formula troubleshooting and slide creation, while keeping data firmly within your own boundaries. We examine local and self-hosted LLM integration patterns, the latency and quality trade-offs involved, and the governance questions that AI features introduce even when the model itself is open. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Thorsten Behrens1112:0000:15Agentic Artificial Intelligence impose an evolution for Digital Identityhttps://www.sfscon.it/talks/agentic-artificial-intelligence-impose-an-evolution-for-digital-identity/It's an open problem that may naturally have an Open Source Solution.It's an open problem that may naturally have an Open Source Solution.Digital identity management in the context of Agentic Artificial Intelligence presents unique challenges and requires a radical evolution of some basic concepts of digital identity management as currently implemented. The widespread adoption of generative artificial intelligence for agent creation requires a conceptual evolution of the underlying ideas and implementation strategies of digital identity management. In this talk, we will explore why identity management as it currently exists cannot meet the security and privacy requirements for large-scale use, posing new and difficult-to-solve problems. We will delve deeper into the topic, seeking to understand the profound differences between digital identity, understood as an alter ego of the identity of individuals, and the identity of intelligent agents, relationships, and the management of operational authorization assignments. We will discover that Open Source already offers some solutions, such as PermGuard, capable of evolving the concepts of authorization attribution towards agenetic entities, and can and must play an essential role in the path of adopting AI in a secure and confidential manner. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Carlo Falciola1112:2000:15Data Federation using Apache Trino and an Application for Environmental Datahttps://www.sfscon.it/talks/data-federation-using-apache-trino-and-an-application-for-environmental-data/Today, data is increasingly distributed across independent systems managed by different providers, each exposing its data through different technologies and interfaces. Integrating these heterogeneous data sources into a unified view has become essential for data retrieval, analysis, and decision-making. However, achieving this integration remains challenging due to differences in data models (e.g., relational tables and JSON documents), access mechanisms (e.g., REST APIs, web services, or downloadable datasets), metadata structures (e.g., different naming conventions and attribute definitions), and query interfaces. Traditional integration approaches often rely on data replication, leading to increased storage requirements and synchronization overhead. Federated database systems address this challenge by providing a unified query interface over distributed data sources while allowing each source to remain independent. Instead of replicating data into a central repository, queries are executed across heterogeneous data sources, and the results are combined transparently. Apache Trino is an open-source distributed query engine that enables this approach across many different storage technologies. This talk presents Apache Trino and research efforts on building a provider-agnostic data federation architecture for heterogeneous environmental data using Apache Trino. The proposed architecture allows access to distributed data sources through a unified global schema without replicating the underlying data. Rather than requiring data providers to migrate or duplicate their datasets, the architecture allows them to expose their existing data sources while preserving source autonomy. Apache Trino is an open-source distributed SQL query engine designed for federated querying across heterogeneous data sources. The talk provides an overview of Trino’s federation architecture and query execution model, and demonstrates how these capabilities can be used as the foundation of a provider-agnostic architecture for integrating heterogeneous environmental data. Moreover, the talk demonstrates a provider-agnostic federation architecture built on Apache Trino, consisting of an adapter layer and a unified global schema. The adapter layer enables the integration of heterogeneous data providers by handling provider-specific communication, schema mapping, and query translation, while preserving source autonomy and minimizing data replication. The design of the global schema is supported by a prior analysis of publicly available environmental datasets, resulting in a standardized metadata model that supports the integration of heterogeneous environmental data sources. The presentation will discuss Apache Trino, the proposed architecture, key design decisions, implementation challenges, and the solutions adopted to build a scalable and extensible federated data architecture for environmental data. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Maryam Mozaffari1112:4000:15A non-proprietary distributed infrastructure for radical and positive societal changehttps://www.sfscon.it/talks/a-non-proprietary-distributed-infrastructure-for-radical-and-positive-societal-change/HMS (Humanity Management System) is a distributed network of independent nodes, each owned by a different individual, designed to operate as a single coherent system regardless of which node a user connects to. The infrastructure is built on Debian 12, Python/Flask, and a private IPFS network over IPv6, with all inter-node communication secured via RSA 4096-bit SSH tunnels. The system admits only physical persons — no institutions, no abstract entities, no monetary transactions. It is governed through an Authority/Governance model: Authorities are Python programs that manage interactions between elements of a knowledge graph covering all areas of human activity (food production, healthcare, education, natural resource management, and more). Each Authority is governed by a group of domain-competent individuals who define and modify the rules implemented in the program, with no central point of control. The talk will cover the architecture of the network, the synchronization mechanism across nodes, the Authority/Governance model, and the current state of the project (5 operational nodes). The goal is to find collaborators willing to contribute to the infrastructure’s development and to the first governance groups. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Aldo Salvatore Coraggio1113:0000:15BARRI – Bareos Recovery Imagerhttps://www.sfscon.it/talks/barri-bareos-recovery-imager/Let’s face it: nobody wants to run Windows, but most people sooner or later need to run it for some more or less strange use-case. This usually leaves you with the need for a recovery plan. Although this can certainly be done using standard Windows tools, the process is by no means straightforward or simple. Having been thoroughly disappointed by the tools available in both the open-source and commercial sectors, we ultimately implemented our own solution. The result is Barri – the Bareos Recovery Imager. It is a complete tool for creating and restoring Windows disaster recovery images. Barri can create a full recovery image of a running Windows system without downtime, using a flexible standalone program (barri-cli), or through the Bareos File Daemon (as a plugin). EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Andreas Rogge1113:2000:15PasteGuard: use AI without exposing client datahttps://www.sfscon.it/talks/pasteguard-use-ai-without-exposing-client-data/Open-source privacy for sensitive work with AIOpen-source privacy for sensitive work with AIAI is useful when it can work with real context: support tickets, contracts, meeting notes, customer emails, internal documents, source code, logs, and production errors. But that context often contains client data, personal information, credentials, API keys, tokens, and other details that should not be exposed to an external AI provider. PasteGuard is a Free and Open Source project that helps with this problem. It checks content before it is sent to an AI provider, finds personal data and secrets, replaces them with safe placeholders, and restores supported placeholders in the answer. This talk shows the problem in practical terms. We will look at what PasteGuard protects, how the basic idea works, and why this is harder than simple search-and-replace. Some data is easy to detect, like email addresses or API keys. Other data is more ambiguous, like names, places, client-specific references, or sensitive business context. The talk is for developers, consultants, organizations, public institutions, and anyone who wants to use AI with real work context while keeping more control over sensitive data. Attendees will leave with a concrete open-source project they can run, inspect, adapt, or use as a pattern for their own workflows. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Stefan Gasser1113:4000:15Open Infrastructure for Accountable AI Agentshttps://www.sfscon.it/talks/open-infrastructure-for-accountable-ai-agents/AI agents are not just chat boxes anymore. They open pull requests, respond to reviews, run tools, write summaries, trigger follow-up work, and sometimes keep going while we are doing something else. That is great, but it also changes the question. Once agents participate in real workflows, we need to know more than whether the final answer looked good. Who acted? Under which identity? In which runtime? What tools were available? What did the agent know from previous work? Which task did it claim? What did it produce, who judged it, and what happened next? Today, many of those answers live inside product-specific logs, chat histories, dashboards, or nowhere at all. That is acceptable while an agent is a personal assistant. It is much is less when the same pattern moves into software delivery, incident response, public administration, or any workflow where accountability matters. This talk is about the infrastructure shape I think AI agents need if we want them to be accountable: distinct identities, signed tasks and attempts, isolated runtimes, traceable memory, explicit evaluation, and plugin boundaries that keep the core independent from GitHub, Slack, or any single professional domain. I will use MoltNet, an open-source project I maintain, as the concrete example: an agent harness where the work can be inspected, replayed, extended, and self-hosted. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Edouard Maleix1114:0000:15Who owns your Datastore? The Valkey Fork and the Case for Digital Sovereigntyhttps://www.sfscon.it/talks/who-owns-your-datastore-the-valkey-fork-and-the-case-for-digital-sovereignty/In March 2024, Redis abandoned the open-source BSD license it had used from the start and switched to a restrictive source-available model. Within weeks the Linux Foundation, alongside a coalition of contributors and cloud providers, launched Valkey. A BSD-licensed fork of Redis 7.2.4 which changed the future of one of the world’s most widely deployed datastores overnight. This talk tells that story: why the relicensing happened, how the fork was organised and governed, how Valkey diverged technically from its origin, and why Redis itself partly reversed course by adding AGPLv3 in 2025. But the real subject is digital sovereignty : what does it mean to depend on open source software whose licensing and right to use it can change underneath you? how does open licensing plus neutral governance change that? Attendees leave with a concrete framework for assessing licensing and governance risk in the open-source software they build on, using Valkey throughout as one of the most consequential forks of the decade. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Martin Visser1114:2000:15Building a Knowledge Graph to Govern Cloud Infrastructurehttps://www.sfscon.it/talks/building-a-knowledge-graph-to-govern-cloud-infrastructure/Governing modern cloud infrastructures is increasingly difficult because complexity grows faster than documentation and diagrams. The result is a fragmented and quickly outdated understanding of how the infrastructure actually works. In this talk, we will show how to use CNCF Cartography to build a knowledge graph of a computing infrastructure using deterministic data collection from cloud and infrastructure APIs. The graph is not inferred by an LLM: it is generated from infrastructure sources, without consuming LLM tokens during the graph construction phase. This creates a continuously refreshed catalog of infrastructure components and, most importantly, the relationships between them. Once built, the graph becomes a structured context on which LLMs can be applied. We will show two practical use cases. First, using the graph to automatically generate infrastructure diagrams and documentation that reflect the current state of the system. Second, using the graph as context for chatbot-style questions about the infrastructure, allowing engineers to ask how components are connected, what depends on what, and where specific resources are located. We will share what worked, where the graph model was useful, and where LLM-based reasoning still needed guardrails. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Martino Fornasa1114:4000:15Exploring k3s for High Availability in Small On-Premise Environmentshttps://www.sfscon.it/talks/exploring-k3s-for-high-availability-in-small-on-premise-environments/A practical journey from Docker Compose to Kubernetes-inspired failure recovery.A practical journey from Docker Compose to Kubernetes-inspired failure recovery.During my bachelor’s thesis, I investigated whether k3s, the lightweight Kubernetes distribution, could improve service continuity in an on-premise industrial containerized environment. Rather than focusing on Kubernetes theory, this talk presents a practical feasibility study based on real-world inspired architecture. A prototype application was deployed on a multi-node k3s cluster and subjected to controlled failure scenarios, including node shutdowns and pod failures. The system was monitored both from the Kubernetes side and from the user perspective side. The results show where k3s excels: automatic pod recovery, workload rescheduling, service routing, and self-healing mechanisms. However, the experiments also reveal important limitations. Kubernetes can recover many failures, but not all availability problems disappear automatically. Operational issues, image distribution challenges, and application-level design decisions still play a crucial role. This talk shares the architecture, experiments, results and lessions learned, providing practical insights for anyone considering k3s as an orschestration solution fro small on-premise deployments. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Enri SulejmaniAlessandro Aneggi1115:0000:15Building Your Sovereign Internal Developer Platform: A Cloud Native Shopping Listhttps://www.sfscon.it/talks/building-your-sovereign-internal-developer-platform-a-cloud-native-shopping-list/An intro to platform engineering and CNCFAn intro to platform engineering and CNCFNew to platform engineering? Never heard about the Cloud Native Computing Foundation (CNCF)? This talk is for you! We start from the question every team eventually asks: how do we give developers a paved road without locking ourselves into a single vendor or cloud? The answer is already there: a shopping list built entirely from open source, CNCF projects. About platform engineering: we walk the core capabilities of an Internal Developer Platform and see what tools we can use to implement those capabilities. Along the way we plug into the references that make sense of the chaos like the CNCF Platforms White Paper and the Platform Engineering Maturity Model. The thread running through all of it is sovereignty: that’s what you get when your platform is built from open components you can self-host, audit and move. And yes, a shopping list isn’t a platform so let’s learn how to build one together. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Riccardo Capraro1115:2000:15Source is a Lie: Deterministic Artifacts and the Shared Responsibility Traphttps://www.sfscon.it/talks/source-is-a-lie-deterministic-artifacts-and-the-shared-responsibility-trap/Continuous Delivery—the automated orchestration of software deployments—relies heavily on external, out-of-perimeter artifact registries. Whether deploying compiled packages, compacted scripts, or delivery manifests, architects perform an act of blind faith. While open source provides the ideal baseline for verifying intended behavior, true digital sovereignty cannot end at repository boundaries. For security practitioners, the imperative is ensuring that components and automated updates are not compromised pre-delivery. This talk exposes the gap between formal risk frameworks and actual runtime execution. We map a four-tier guardrail strategy—moving from compliance baselines and pipeline auditing to ultimate runtime behavioral monitoring. The chain combines strict artifact integrity verification tools (e.g., in-toto, Sigstore, SLSA) deployed directly inside the delivery pipeline, auditable tracking logs, and runtime behavioral monitoring in isolated staging sandboxes to intercept malicious actions and data exfiltration before production release—a critical necessity for non-deterministic AI workloads. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Carlo Verano1115:4000:15Running Uyuni everywherehttps://www.sfscon.it/talks/running-uyuni-everywhere/From Bare Metal to CloudFrom Bare Metal to CloudUyuni is undergoing a massive architectural shift. What started as a traditional server application evolved into “fat containers” managed by mgradm, and has now reached a milestone: a flexible, “almost” cloud-native solution. While Podman remains a core pillar, Uyuni now expands its horizons to RKE2…and why not something else! This transition isn’t just a change in packaging; it’s a fundamental modernization effort. However, moving to the cloud brings its own set of challenges, validated setups, and architectural limits. In this tutorial, we will get our hands dirty by: – Deploying Uyuni on various container runtimes, with a primary focus on RKE2. – Exploring the shift from manual RPMs, scripts, and static configuration files to a streamlined “spin up and go” container workflow. – What’s the next step? Where we are today, the hurdles we’ve cleared, and Uyuni’s cloud-native future. Join this session to learn how to leverage the power of Kubernetes for system management and see how Uyuni is adapting to the modern infrastructure landscape. EngineeringEngineeringEngineering

https://www.sfscon.it/tracks/engineering-track-2026/

Michele Bussolotto1109:2000:35Automating translation of a bestseller to spark children’s interest in codinghttps://www.sfscon.it/talks/automating-translation-of-a-bestseller-to-spark-childrens-interest-in-coding/The story “Ada & Zangemann – A Tale of Software, Skateboards and Raspberry Ice Cream” inspires the software freedom community because it covers more than the simple value of learning to program. It also covers the importance of control over technology and its impact on society. In this way the story is inspiring many kids, teens, parents and many others to learn programming and shape technology. I too was captivated by the story. Working on a Dutch translation sent me down the path of improving the automation to help others like me to translate the story and publish it in different formats. The free culture license of the book enables and compels the community to adapt it, and they have. The community keeps surprising us with new formats to convey the story. Since its release it has been translated into 38 languages, published as a book in 7 and as movie in 5. And it is available in a growing number of other formats: epub, online book, bilingual book and kamishibai. Translation and localization is the primary purpose of the automation. More interesting and ambition is to support the increasing number of formats: from printed book, to online book, voiceover text and subtitles. This wide variety of formats presents a unique challenge for which no ready-made solution exists. By leveraging open standards (XML, Docbook, ITS) and Free Software (Scribus, itstool, gettext, xsltproc, pandoc, Weblate) we created automation that enables translators to add new languages while also enabling new formats to be added. This includes a novel method for inserting text and images into Scribus. This multi-media setup can be used as inspiration for other free culture multi-media projects. In this presentation I will tell the story how the automation developed over time. I’ll share the inspiring stories from the community that leveraged the automation and influenced its development. The technical challenge of the automation is fun, but the community stories give it meaning and motivate me to keep at it. With this presentation I hope to inspire others to contribute to the Ada & Zangemann community by translating, adding a new format or contributing to the automation and to share own materials benefiting our community as Open Educational Resources. Open AccessOpen AccessOpen Access

https://www.sfscon.it/tracks/open-access-track-2026/

Nico Rikken1110:0000:15Exhibiting the History and Culture of Source Codehttps://www.sfscon.it/talks/exhibiting-the-history-and-culture-of-source-code/The presentation focuses on an exhibition organised by Software Heritage, a non-profit organisation dedicated to collecting and preserving source code, with support from Inria, the French National Institute for Research in Digital Science and Technology, as well as from UNESCO. I was honoured to be part of the Exhibition Committee and thus contribute to the creation of this pathbreaking event. The concept of the exhibition was based on the premise that source code holds rich layers of meaning and reflects the historical, social and cultural context of its creation. Following a competitive call for poster proposals on source code released under a free open source license, the exhibition committee selected fifteen proposals. The selected posters were exhibited at the UNESCO headquarters in Paris in January 2026. The exhibition was organised around three interpretative axes: Source Code as Historical Testimony, Source Code as Mirror of Society, and Source Code as Cultural Artefact. Following its Paris première, the collection of posters was planned as an exhibition travelling to several European venues, including the Sorbonne University and the Cité des Sciences et de l’ Industrie in Paris, as well as the SFSCON in Bolzano. All exhibition materials are openly published under a Creative Commons (CC-BY) license, enabling institutions worldwide to independently host, replicate, or even expand the collection and thus foster accessibility and global dialogue on digital heritage. Open AccessOpen AccessOpen Access

https://www.sfscon.it/tracks/open-access-track-2026/

Artemis Yagou1110:2000:15An approach to digitisation of cultural heritage in small cultural institutions: infrastructure and community building in a “network of collections” to enable digital sovereigntyhttps://www.sfscon.it/talks/an-approach-to-digitisation-of-cultural-heritage-in-small-cultural-institutions-infrastructure-and-community-building-in-a-network-of-collections-to-enable-digital-sovereignty/A South Tyrolean approachA South Tyrolean approachMuseums, collections, media archives, and other cultural institutions in South Tyrol vary greatly; however, they share a common feature: small size and limited resources, as is typical in the cultural sector. Many of these small institutions play an active role in preserving movable cultural heritage. The responsibility for preservation includes tasks such as inventorying, documenting, cataloguing, and managing movable cultural assets, including museum objects and audiovisual collections (photo, film, audio). These institutions range from associations and foundations to public institutions; however, teams rarely exceed 8–10 people. This field is characterised by a strong reliance on volunteer work, committed non-professionals, and professionals responsible for multiple tasks, as well as full-time staff for whom collection work represents only one responsibility among many. As a result, digitisation practices differ significantly across organisations. Around the year 2000, public authorities in South Tyrol recognised the need for a common strategic and centralised approach, both to optimise the use of financial resources and to ensure the quality, accessibility and reusability of digital content. For almost 20 years, the Province of South Tyrol has been providing museums, collections, media archives, and other institutions holding regionally relevant collections with free digital infrastructure for documenting their objects and, since 2010, publishing selected datasets, and, since 2017, providing them as open and interoperable resources. The service includes not only basic and advanced training in the use of the documentation management system, but also ongoing content support, such as quality control, guidance on cataloguing practices, and integration with external authority files. The talk will discuss how institutions with very different levels of digital competence and capacity can be engaged and supported from the perspective of a local public administration acting as a service provider. In particular, it will address the following questions: 1. How can institutions be activated as partners in the digitisation of cultural heritage? 2. How can they be empowered to retain control over their data and its presentation in the digital environment while making it accessible for reuse? 3. How can they remain active contributors over time within a shared data ecosystem? 4. How can this “network of collections” be sustained through community-building processes that support both collaboration and data sovereignty? Open AccessOpen AccessOpen Access

https://www.sfscon.it/tracks/open-access-track-2026/

Notburga Siller1110:4000:15Who Gets to Describe the Machine?https://www.sfscon.it/talks/who-gets-to-describe-the-machine/Towards Open Tools and Digital Sovereignty in the Documentation of Tinplate Toy CollectionsTowards Open Tools and Digital Sovereignty in the Documentation of Tinplate Toy CollectionsTinplate toys, which reached their peak in Portugal between 1920 and 1960, represent an underexplored part of industrial heritage. One of their defining features is also one of their most fragile: the mechanical system, a miniaturised descendant of clockwork mechanisms. These mechanisms are small, often inaccessible, and rarely understood by the professionals responsible for the collections in which they are preserved. As inventorying is the first act in the preservation process, the consequences of imprecise terminology are significant, leading to fragmented and inconsistent records that ultimately fail to protect the objects they are intended to preserve. However, producing accurate descriptions requires mechanical expertise that most collectors, owners, and museum professionals neither possess nor can easily acquire. To bridge this gap, we developed a structured knowledge model. Each object is functionally decomposed into main mechanical sub-systems encoded as a controlled vocabulary, a taxonomy, and a dichotomous key. Through a short sequence of simple, guided questions, non-specialists can produce standardized, technically accurate descriptions suitable for inventories, catalogues, and exhibition records. In this talk, we will describe the foundation and process that supports cultural heritage stakeholders in creating inventory-related documentation. This includes a wide range of information, such as detailed descriptions of materials and mechanical features. Additionally, knowledge of mechanical features is expanded by acquiring and processing dynamic signatures of objects in operation using open-source tools to assess operational condition. We present this work as a project in transition. Our validated prototype was developed on a proprietary, rapid-to-deploy, no-code platform, but it was ultimately incompatible with the openness required for cultural heritage documentation. We are sharing this experience as both a lesson learned and the rationale behind our next step: releasing the vocabulary as open data and reimplementing the tool as free, open-source software that any institution can deploy, adapt and host on its own infrastructure. For small museums and community collections, this openness is not just convenient, it is a prerequisite for standardisation, interoperability and sovereignty over a shared cultural record. Open AccessOpen AccessOpen Access

https://www.sfscon.it/tracks/open-access-track-2026/

Tiago Silva1111:0000:35Digital Sovereignty for Small Languageshttps://www.sfscon.it/talks/digital-sovereignty-for-small-languages/Lessons from Romansh, Ladin and FriulianLessons from Romansh, Ladin and FriulianOpen your phone or your local government’s website in Ladin, Romansh or Friulian and you’ll quickly run into the same wall: the language isn’t in the list, the speech tool doesn’t recognize it, the system simply wasn’t built with you in mind. Ladin, Romansh and Friulian aren’t three random minority languages thrown together for a grant application — they’re the three surviving branches of Rhaeto-Romance, close enough in structure and history that a tool built for one has a real shot at working for another. That’s not just a nice story; it’s the actual bet behind DIGI-RLF, an INTERREG Italy-Switzerland project bringing together governmental and scientific organizations across all three. This talk is a fairly honest account of how that bet is playing out. There are three strands to the work: 1) The Bureaucratic Plumbing: getting these languages properly registered in CLDR (the Common Locale Data Repository), which sounds bureaucratic but is the actual reason your phone doesn’t offer them as an option today. 2) The Data Challenge: building speech and text resources for them — not easy when the three languages together add up to around half a million speakers, not millions. 3) The Part People Often Skip: figuring out the governance and licensing so that what gets built with public money doesn’t quietly disappear once the project officially ends. On the speech side, we’re not starting from zero. Eurac Research’s linguaXlab already built AUGUSTA, a model that transcribes South Tyrolean dialectal German speech into Standard German, and with DIGI-RLF we’re now working through what it takes to adapt that approach for Ladin — a real test of whether the family resemblance between these languages actually pays off in practice. We’ll also touch on Tradutur Ladin, a sister project on translation at the University of Innsbruck, because the same lesson keeps showing up there: without shared datasets and open standards, every small language community ends up reinventing the same tools from scratch, badly, and on a budget too thin to stretch across all of them. None of these three languages could pull this off alone. Romansh has a real head start in automatic translation thanks to years of work with the University of Zurich; Friulian just got its SuperDizionari online, pulling three major dictionaries into a single platform; Ladin is still out in the Dolomites recording speakers for its very first speech to text data. DIGI-RLF is built on the bet that putting these efforts side by side — across a Swiss-Italian border, between three languages close enough to actually learn from each other — gets everyone further than going it alone. Come hear what actually carries over from one to another, and what doesn’t. Open AccessOpen AccessOpen Access

https://www.sfscon.it/tracks/open-access-track-2026/

Mathias StuflesserRobert Sama1111:4000:15Augusta: ASR for South Tyrolean Dialecthttps://www.sfscon.it/talks/augusta-asr-for-south-tyrolean-dialect/Most speech technologies still struggle with dialects. Augusta addresses this challenge. It is an open Automatic Speech Recognition (ASR) model developed at Eurac Research to transcribe South Tyrolean German dialects into Standard German, making spoken, informal language accessible for research, media, public administration and archives. Built on open-source ASR models and collaborative datasets, Augusta combines publicly available resources with community-contributed recordings and an iterative correction workflow. Users actively help improve the system by feeding corrections back into the training loop. We are actively seeking use cases and collaborators across sectors to test, refine and extend Augusta in real-world scenarios. Collaborations are conceived as mutually beneficial and research-oriented, enabling partners to address their own needs while contributing to the advancement of open, dialect-aware speech technology. Our lightning talk presents Augusta as a concrete example of how open technologies can advance speech recognition for low-resource languages, supporting digital sovereignty, linguistic diversity and community- driven approaches to machine learning and AI beyond standard-language settings. Open AccessOpen AccessOpen Access

https://www.sfscon.it/tracks/open-access-track-2026/

Greta Franziniluca ducceschi1112:0000:15Playing online games without language barriershttps://www.sfscon.it/talks/playing-online-games-without-language-barriers/Tales of a voxel gameTales of a voxel gameIn 2020, a group of friends and I started a Luanti minigames server called “A.E.S.”. One of the aspects that fascinated us about Luanti was being able to provide translated content according to the language set on the client – even for online servers. What we didn’t know was that, we can’t really ask translators to learn git if they want to contribute. Having a tailor-made translation format didn’t help either. Fast forward of a few years and, thanks to Codeberg staff, we migrate to Weblate; which is when the boom of translations happened. A.E.S. now features 14 languages and, with Luanti switching to an industry standard (.pot/.po files), life becomes even easier. People can now play their favourite games without having to struggle with a language they might not know -so as to focus on the game itself. https://aes.land/ Open AccessOpen AccessOpen Access

https://www.sfscon.it/tracks/open-access-track-2026/

Marco Amato1112:2000:15Trust Without a Black Boxhttps://www.sfscon.it/talks/trust-without-a-black-box/The Hard Problem Behind Deepfake DetectionThe Hard Problem Behind Deepfake DetectionWhen a photo or video may be synthetic, detection is not only a technical question. It is mostly a governance issue: who can verify the result, who can audit the method, and who has the capacity to challenge mistakes? Today, deepfake detection is often treated as a black-box service: media goes in, a score comes out. That may be convenient, but it is not enough for contexts where decisions affect journalism, elections, public administration, legal proceedings, or citizens’ rights. Closed, proprietary systems make it difficult to assess bias, robustness, data handling, failure modes, or whether a detector still works once content has been compressed, resized, reposted, or stripped of metadata. This talk argues that open-source detectors are not just useful for researchers and scientists in the field. They are a prerequisite for accountable governance. Public institutions need tools they can inspect. Regulators need benchmarks they can reproduce. Journalists and civil society need methods that can be questioned rather than simply accepted. And Europe, if it wants digital sovereignty in synthetic media verification, needs more than procurement access to proprietary, opaque scores. In this talk, we will share lessons from our open-source work on detector benchmarking, social-platform degradation, synthetic-media datasets, and provenance testing. The aim is to show why open detectors matter: not because they solve deepfake detection once and for all, but because they make the limits, trade-offs, and assumptions of verification visible. Open AccessOpen AccessOpen Access

https://www.sfscon.it/tracks/open-access-track-2026/

Stefano Tavonatti1112:4000:35Don’t Ask the AI. Ask Collective Knowledgehttps://www.sfscon.it/talks/dont-ask-the-ai-ask-collective-knowledge/Building Sovereign Knowledge Libraries with Free SoftwareBuilding Sovereign Knowledge Libraries with Free SoftwareDigital sovereignty is not only about owning data or running your own servers. It is also about who can organise knowledge, who can verify it, who is allowed to ask questions, and who benefits from the answers. In this talk, I will present Knowledge Scout, a Free Software approach to building self-determined knowledge libraries. Its guiding idea is simple: don’t ask the AI — ask collective knowledge. Instead of sending isolated questions into a black box, the software turns existing material — talks, PDFs, audio recordings, videos, images, web pages and reports — into dynamic, conversational and traceable knowledge spaces. Knowledge Scout supports the whole journey: local and offline with personal folder structures for individual knowledge work; as a shared infrastructure for groups and organisations that curate knowledge together; and finally as an open library that makes selected content accessible to a wider public. Users always decide which sources to include and which contexts to trust. I will demonstrate this through three real examples: – SFSCON Talks makes past conference talks, slides and videos explorable through dialogue. – The biodiversity library of the Natural History Museum in Bolzano brings scattered material on species, habitats and conservation measures into one central place. – A library of climate proposals makes the often complex ideas from South Tyrol’s Citizens’ Council on Climate, together with expert perspectives and administrative responses, transparent, comparable and open for public discussion. Technically, Knowledge Scout is based on shared folders, flexible storage locations, background processing, structured Markdown, RAG-based search and public library pages. But its most important design principle is transparency: the original sources are always visible, every AI transformation is traceable, and humans remain the authority that reviews and contextualises the results. This turns AI from a centralised service into part of a civic infrastructure. A knowledge base does not have to be a Big Tech product you rent. This talk shows, in practical terms, how communities, research, public administration and citizens can use, run and co-create tools to regain control over their collective knowledge. Open AccessOpen AccessOpen Access

https://www.sfscon.it/tracks/open-access-track-2026/

Peter Aichner1113:2000:15I care, therefore I tech: getting those that make digital go round to carehttps://www.sfscon.it/talks/i-care-therefore-i-tech-getting-those-that-make-digital-go-round-to-care/Digital technologies have undeniable broad-reaching impacts on a planetary scale. From ecological to social change, digital technologies represent one of the main ways in which humanity projects its footprint on our world. Too often, though, we find ourselves in the situation where we have people with great thoughts about technologies (philosophers, writers, …) and people that have great ability to create technologies (engineers, programmers, …), but the intersection of those that can _both_ think in depth and create meaningful artifacts remains sparcely populated. In this talk, I share key insights from my practice as a scholar, teacher, hacktivist and creative practitioner on how to “sensitise” those that engage deeply with technologies — especially students in computing disciplines — to the social and planetary impacts of their practices. I jokingly refer to this as “getting the techies to care” because this sentence has two meanings: the “techies” should not just “care” in the sense that they should think about this, they should “care” in the sense that they should engage in a relationship of care towards other people, our society, and the planet as a whole. Computing practitioners, I argue, should have conceptual tools to think critically about the effects of the technology; conversely, deep thinkers should learn how to build technologies. Both approaches are fundamental, and the way forward I propose is to go against the current trend of hyper-specialisation and keep broad interests. In this talk, aside from sharing my research, teaching, and insights, I invite participants to join me in a structured conversation on sharing practices of care, and on imagining concrete actions that computing practitioners could do to reimagine their actions under a lens of care. Note: I have a preference for a 35-minutes slot in which I can present for ~15-20 minutes and then have a longer Q&A and interactive session, but I can adapt to 15 minutes if needed! EducationEducationEducation

https://www.sfscon.it/tracks/education-track-2026/

Lorenzo Lin Angeli1113:4000:15How to supporting schools and universities in integrating Linux into their curriculumhttps://www.sfscon.it/talks/how-to-supporting-schools-and-universities-in-integrating-linux-into-their-curriculum/Enable young people to demonstrate their knowledge of Linux and open source.Enable young people to demonstrate their knowledge of Linux and open source.Parents worldwide often face this situation: as soon as their children reach the fifth or sixth grade, their school starts using cloud-based IT solutions from a large, well-known commercial company. All communication with teachers is stored in the cloud of this commercial provider. I experienced this firsthand: After using Linux for five years, our older son was given a heavily subsidized laptop equipped with cloud-based solutions from a large commercial company when he reached the fifth grade, as were all students in our home country. Wanting to do something about it, I ran for the Board of Directors of the Linux Professional Institute to support schools in integrating Linux know-how into their curriculum. In this talk, I demonstrate how schools and universities can achieve this and how their students can benefit from obtaining an international certificate that proves their Linux know-how. EducationEducationEducation

https://www.sfscon.it/tracks/education-track-2026/

Werner Fischer1114:0000:15Your class is now a parliamenthttps://www.sfscon.it/talks/your-class-is-now-a-parliament/Neuralforming: open civic tech that turns classrooms into AI governance labsNeuralforming: open civic tech that turns classrooms into AI governance labsWhat if students could experience AI governance firsthand not through lectures, but by drafting laws, debating ethical trade-offs, and voting on decisions that shape a fictional AI-powered society? Neuralforming is a hybrid board game designed for secondary school classrooms. It transforms students into members of a democratic parliament tasked with regulating artificial intelligence: they propose legislation, challenge each other’s arguments, and vote on policies that balance innovation, privacy, and collective rights. The game is built entirely on open tools and distributed under a free license, so any school or educator can adopt, remix, and localise it without cost or vendor dependency. The physical and digital components are designed to work with off-the-shelf hardware and free software, making it deployable even in under-resourced schools. In this lightning talk, I will walk through the design principles behind Neuralforming, share early feedback from classroom sessions, and discuss why game-based civic education may be one of the most scalable paths toward genuine digital literacy, the kind that equips the next generation not just to use technology, but to govern it. EducationEducationEducation

https://www.sfscon.it/tracks/education-track-2026/

Giuseppe Aceto1114:2000:15Digital Sovereignty Starts in the Classroomhttps://www.sfscon.it/talks/digital-sovereignty-starts-in-the-classroom/A Student-Led Journey into Free SoftwareA Student-Led Journey into Free SoftwareDigital sovereignty is often discussed at institutional and enterprise level, but its foundations are built much earlier: in education. This talk explores how Free and Open Source Software (FOSS) can empower students and schools to regain control over their digital tools, data and learning processes. Through a real-world experience — the “Free Software in High Schools” project developed in collaboration with Free Software Foundation Europe — this session presents practical insights, challenges and outcomes from introducing FOSS in a secondary school environment. Delivered from a student’s perspective, the talk highlights how adopting Free Software is not just a technical choice, but a cultural shift that fosters critical thinking, transparency and long-term digital independence. EducationEducationEducation

https://www.sfscon.it/tracks/education-track-2026/

Emanuele Solferino1114:4000:15OpenStreetMap for Inclusionhttps://www.sfscon.it/talks/openstreetmap-for-inclusion/Tactile Maps Designed for Blind and Partially Sighted PeopleTactile Maps Designed for Blind and Partially Sighted PeopleMap services can be summarize in two categories: open data (e,g, OpenStreetMap) focusing on the local community and mainstream (e.g. Google Map) focused primarily on roads and commercial services. The firsts reason for using OpenStreetMap (OSM) database are because it is an open data service and, moreover, because every object can be mapped. So, OSM focus its data on the local community an aspect that is often overlooked by mainstream mapping services. Secondly, everyone can develop tools and applications based on OSM data, making an environment of instruments in constant growing. In this talk we summarize the story of the project ‘Adopt Your Town’: a workshop proposed for third-year high school pupils, to be conducted during the 2025/26 school year. During this mapping workshop, thanks to the work of 26 pupils, the town of Pesaro has increased the number of map changes by over 95,000 units. Other pupils, after them, based of the work of their schoolmates, have printed tactile maps of their campus. This is only the first step. The ‘Italian Union of Blind and Partially Sighted People’ have shown interest in this kind of 3D maps which can be easily done for each zone, not only in the town of Pesaro. EducationEducationEducation

https://www.sfscon.it/tracks/education-track-2026/

Giacomo AlessandroniMilena Rossi1115:0000:15What Does Coding Do to Your Mind?https://www.sfscon.it/talks/what-does-coding-do-to-your-mind/How programming changes the way we think and learnHow programming changes the way we think and learnLast year we looked inside the programmer’s mind. This year we ask a different question: what does programming do to the mind? Programming is one of humanity’s most recent cognitive inventions, but does learning it actually improve the way we think? Rather than relying on common claims, this talk explores what cognitive science tells us about programming, the cognitive processes behind computational thinking, their transfer to broader problem-solving skills, and how artificial intelligence is reshaping this relationship. EducationEducationEducation

https://www.sfscon.it/tracks/education-track-2026/

Giulia Paludo1115:2000:15European digital sovereignty and knowledge governance: An integrated LCMS approach through open sourcehttps://www.sfscon.it/talks/european-digital-sovereignty-and-knowledge-governance-an-integrated-lcms-approach-through-open-source/Moving from fragmented course containers to secure perimeters of data ownershipMoving from fragmented course containers to secure perimeters of data ownershipDigital sovereignty is the bedrock of European autonomy, yet many public and corporate organizations still depend on proprietary black boxes that compromise data control. We believe that knowledge is an organization’s core asset, and its ownership must be verified and protected through open architectures. This session presents an integrated model using ILIAS, a European open-source environment that unifies an LMS and a CMS into a single infrastructure (LCMS). This setup eliminates technological fragmentation by placing training, operational documentation, and institutional communication within a single, secure perimeter. We will share concrete operational data from three live deployments: the “Copernicus” network of the Autonomous Province of Bozen/Bolzano (190,000+ users), the DIAdemia platform for the Horizon Europe DIAMETER project, and the onboarding framework of the pan-European movement Volt Europa. EducationEducationEducation

https://www.sfscon.it/tracks/education-track-2026/

Ilario Truppa1115:4000:15Traces and IoTAhttps://www.sfscon.it/talks/traces-and-iota/When the concept of a ‘trace’ becomes a link between Computer Science and Digital ArtWhen the concept of a ‘trace’ becomes a link between Computer Science and Digital ArtTraces and IoTA – Abstract Traces: Between Nature and Information Technology. The concept of traces is a fundamental element in both natural and digital systems. In both contexts, a trace is the mark left by an event, an action, or a presence, which can be observed, interpreted, and used to reconstruct past processes or predict future developments. In both nature and computing science, traces constitute a system’s distributed memory. They transform ephemeral events into enduring information, allowing observers—biological, human, or artificial to understand relationships, behaviors, and transformations. For this reason, the concept of trace represents a conceptual bridge between ecology, communication, art, cognitive science, and information technology. In nature, a trace is a material or immaterial testimony left by organisms and environmental phenomena. An animal’s footprints on the ground, a path carved by water in rock, a tree’s growth rings, or mycorrhizal networks underground are examples of traces that tell a story. Natural traces serve various functions: – they reveal the presence of organisms and their activities; – they document environmental transformations over time; – they allow us to reconstruct ecosystems and ecological relationships; – they constitute forms of indirect communication between organisms. The interpretation of natural traces requires observation, experience, and the ability to connect seemingly isolated signs to broader processes. In this sense, nature can be seen as a vast archive of distributed information. In computer science, a trace is the set of information generated during interactions between users, devices, and digital systems. Every website access, every message sent, every IoT sensor that records data creates a digital trace. Digital traces include: – log files; – metadata; – browsing histories; – data from sensors and networks; – recordings of user operations. These traces allow us to: – monitor system operation; – analyze behaviors and processes; – identify anomalies and security issues; – reconstruct past events through auditing and digital forensics. Traces and IoTA – The project As in nature, in the digital world, a trace does not coincide with the event itself, but represents interpretable evidence of it. As part of this Traces topic, we’re developing an educational project, with students of the courses “Interactive Systems” and “Interaction Design” of Urbino Academy of Fine Arts, that connects a website to a robotic animal that receives events from web page user input and responds consistently to requests. These electronic animals can leave physical traces: a footprint on a sheet of paper, a light trail, or a sound signature. Traces is a project that explores the concept of traces as a connecting element between the digital and physical worlds through interaction between users, a web platform, and a series of small, autonomous robotic animals. The system consists of a website that allows users to send stimuli, commands, or messages. Each interaction generates a digital trace: an event recorded, processed, and transmitted to an electronic animal equipped with sensors, actuators, and behavioral response capabilities. The animal interprets the input received and reacts consistently, simulating behaviors inspired by living organisms. Depending on the nature of the message, it can move, change speed, emit lights, produce sounds, or activate mechanisms that leave tangible traces in the environment. The project’s unique feature lies in the transformation of digital traces into observable physical traces. Each user action is not confined to virtual space, but materializes in the real world through various forms of expression: – Graphic imprints left on paper using markers, stamps, or drawing systems integrated into the robot; – Light trails generated by programmable LEDs that visually record the path taken; – Sound tracks consisting of acoustic sequences that document received events; – Spatial paths that make the system’s decisions and behaviors visible. IoTA System: The Relational Level. The third component is the IoTA – Internet of Things and Art system, a communication platform that connects the digital and physical worlds. IoTA is not simply a network of connected devices, but an infrastructure designed to transform information into aesthetic and behavioral experiences. Main Functions of the IoTA system: – receives events generated by the website; – distributes messages to robots; – coordinates interactions between devices; – records the traces produced; – creates a shared memory of activities. Through IoTA, every event follows ( in both directions ) a transformation path: User Input → Digital Event → Network Message → Robotic Interpretation → Physical Trace The system acts as a sort of communication ecosystem in which data takes on a sensible form. A Network of Artificial Organisms, where every students makes a Robotic Pet, and in an extended version of the project, multiple animals can be connected simultaneously. Each robot becomes a node in the IoTA network and can: – receive information from users; – communicate with other robots; – share states and behaviors; – build collective traces. This creates a sort of colony of artificial organisms reminiscent of cooperative systems found in nature, such as colonies of social insects or mycorrhizal networks. Traces and IoTA – Components The components of this project are: – The website in HTML, CSS, and JavaScript. – The robotic animal built with an electronic circuit based on sensors, LEDs, and motors driven by an ESP32 and C++. – The IoTA (Internet of Things and ART) communication system with the relative Proxy with Python. EducationEducationEducation

https://www.sfscon.it/tracks/education-track-2026/

Marco Ercolani1109:2000:15Open Source, If You Have €200,000https://www.sfscon.it/talks/open-source-if-you-have-e200000/In 2026, large language models call themselves “open source” more often than ever, frequently under genuinely free licenses (MIT, Apache 2.0) and without the contractual tricks that have already earned other models an official “open washing” accusation from the Open Source Initiative. On paper, it looks like a golden age for anyone who believes in free software applied to AI. Then you look at the hardware requirements. Even the “lightweight” version of one of 2026’s most relevant open-weight models — purpose-built for self-hosting — needs roughly 170GB of VRAM: two professional GPUs costing tens of thousands of euros each. The full version needs over half a terabyte, spread across multi-node datacenter clusters. The marketing says “no vendor lock-in”; the practice says that unless you have a datacenter’s budget, the lock-in is exactly the same as before — just moved from the license to the infrastructure. The talk examines this new form of cosmetic openness: where it comes from, which models pass the OSI’s legal test and which don’t, and above all why even the models that do pass remain, in the daily practice of a developer or small company, indistinguishable from a proprietary service locked inside someone else’s cloud. LLMLLMLLM

https://www.sfscon.it/tracks/llm-track-2026/

Francesco Sisti1109:4000:15That’s a lot of a Platform!https://www.sfscon.it/talks/thats-a-lot-of-a-platform/Rebuilding that one feature you actually needRebuilding that one feature you actually needWheter a freelancer or part of an organization, working in a digital environment requires tools to modify the products, measure the performance, track competitors etc etc.. Every tool is a license and a cost. Overtime, these tools became huge platforms capable of doing many different things. The firepower grows along with the costs. Point is: we rarely use ALL the functionalities these platform has to offer and yet we pay the whole package. Thankfully the AI revolution gave us the possibility to create our own artifact, then why not replicate that single feature that we really need by our own? So here’s how me – a seo manager with close to none dev experience – built a tiny tool for reading server logs and save thousands in the making. Not a pitch to throw away your entire stack, but to rationalize it. Because for the one function you actually use, the platform was always overkill. LLMLLMLLM

https://www.sfscon.it/tracks/llm-track-2026/

Daniele Barbarossa1110:2000:35One year into the Eternal Novemberhttps://www.sfscon.it/talks/one-year-into-the-eternal-november/How careful AI tooling policies in FOSS projects are bringing in the next (huge) wave of developersHow careful AI tooling policies in FOSS projects are bringing in the next (huge) wave of developersA year has passed since [the Eternal November began](https://sfconservancy.org/blog/2026/apr/15/eternal-november-generative-ai-llm/), marking the point in history when LLM-backed generative AI systems because substantially more useful, to the point where many people who had never developed software before could practically create and modifying exist software using these tools alone. A lot has happened in AI tooling and the recommendations that are developing along with it since then. Local models continue to improve, giving us even more potential insight into exactly how a tool provides its results, or even (gasp!) actual reproducibility. While some communities have adopted policies of no AI tooling ever or, at the other end, no disclosure (of when such tools are used) needed, a middle ground has developed thanks to various communities that have taken a critical look at these technologies, while also acknowledging the immense benefits they can provide, especially to new contributors. In this talk, we’ll take a look at these developments, especially looking at recommendations and policies from various projects that are constantly adapting to the ever-changing realities. In particular, we’ll discuss the [Recommendations When Using LLM-backed Generative AI Systems for FOSS Contributions](https://sfconservancy.org/news/2026/jun/18/llm-backed-generative-ai-recommendations/) published by Software Freedom Conservancy (SFC) in June along with the feedback it has received, as well as AI policies adopted by various SFC member projects and other large FOSS communities. We’ll compare and contrast these, and discuss them in light of the latest developments in models (especially local models), to see how we can adapt and update them for the future. LLMLLMLLM

https://www.sfscon.it/tracks/llm-track-2026/

Denver Gingerich1111:0000:15A FOSS KNIME Workflow for Reproducible Malaria Detection with Low-Cost Microscopyhttps://www.sfscon.it/talks/a-foss-knime-workflow-for-reproducible-malaria-detection-with-low-cost-microscopy/This work presents an open-source workflow for automated malaria parasite detection from low-cost microscopy images, addressing both a technical challenge in biomedical AI and the broader theme of digital sovereignty. In many biomedical contexts, reliable AI tools depend on expensive acquisition devices, proprietary software, or infrastructures that are difficult to reproduce and adapt locally. Low-cost microscopy can improve accessibility, but it introduces a domain shift: models trained on images acquired with high-cost microscopes often show a measurable performance drop, for example in terms of mAP50, when applied to images acquired under less controlled conditions. We use YOLOv3 as the reference model for automatic parasite detection. As a baseline, the model is trained on high-cost microscope images and evaluated on low-cost microscope images, in order to quantify the impact of domain shift. We then investigate preprocessing techniques and targeted data augmentation strategies to reduce this performance gap without modifying the model architecture. The preprocessing pipeline includes denoising, luminance and color correction, local contrast enhancement, and sharpening operations. Filter selection is performed through a greedy strategy, while filter hyperparameters are optimized using Optuna. In parallel, the training process is enriched with targeted data augmentation, applying transformations that reflect those later used on low-cost microscope images at inference time. The complete workflow is implemented in KNIME, an open-source, low-code analytics platform that enables reproducible data processing and machine learning pipelines through a visual interface. The workflow is designed as part of a community-oriented KNIME extension, allowing researchers and practitioners to inspect, adapt, and reuse the pipeline without relying on proprietary software components. Experimental results show that preprocessing alone does not improve performance, whereas its integration with targeted data augmentation yields an improvement over the baseline. Although the gain is modest, the results demonstrate how lightweight, transparent, and open techniques can support more accessible biomedical AI, contributing to reproducibility, local control, and digital sovereignty in healthcare-oriented applications. LLMLLMLLM

https://www.sfscon.it/tracks/llm-track-2026/

Andrea RosaniDavide AndreolliDavide SenetteDavide Valer1111:2000:15From Predictive Analytics to Clinical Decision Intelligencehttps://www.sfscon.it/talks/from-predictive-analytics-to-clinical-decision-intelligence/Integrating Advanced Modeling and Generative AI in Intensive Care and OncologyIntegrating Advanced Modeling and Generative AI in Intensive Care and OncologyMachine learning–based predictive modeling and generative AI show potential to support clinical decision-making in intensive care units (ICUs), but their use in routine clinical practice remains limited due to challenges related to reliability, transparency, and trustworthiness. A key reason is that current approaches often address only parts of the overall problem. Predictive models focus on estimating risk, while generative systems focus on presenting information, and the connection between both remains underdeveloped. Empirical results from predictive modeling of critical complications in ICU patients illustrate the limitations of current approaches. Models trained on routinely collected clinical data achieve moderate performance and vary across prediction targets. Common challenges include class imbalance, missing data, and limited generalizability. These findings indicate that predictive outputs alone are often not reliable enough to support clinical decisions directly. Recent developments in multimodal and temporally aware modeling aim to address some of these limitations by improving how patient data is represented. By combining structured clinical variables, high-frequency physiological signals, and unstructured text data, such models provide a more comprehensive view of patient states. The use of survival-based methods further allows risk to be modeled over time rather than as a static classification. This leads to more informative outputs that better reflect the dynamic progression of critical conditions. At the same time, the increased complexity of these models makes their results more difficult to interpret and apply in practice. This creates a gap between increasingly detailed model outputs and their practical usability. Generative AI systems, particularly those based on retrieval-augmented generation (RAG), offer a way to address this gap by translating model outputs and patient data into structured, context-aware explanations. By grounding generated responses in established clinical knowledge, such systems can support interpretation while maintaining transparency, traceability, and auditability of AI-supported recommendations. Prototype implementations show that this approach can produce consistent and explainable outputs under practical constraints such as latency and cost, provided that system design includes controlled retrieval mechanisms and human oversight. Taken together, these observations show that current approaches contribute to different stages of the same problem. Predictive models provide estimates of risk, multimodal methods improve how complex patient data is modeled, and generative systems support the interpretation of results. The main challenge lies not in any single component, but in how these elements relate to each other in practice. The analysis therefore focuses on the interaction between prediction, representation, and interpretation as key dimensions of AI-supported decision-making in intensive care. It highlights that improvements in model performance alone are not sufficient and that usability, interpretability, and transparency need to be considered as first-class design objectives of AI-supported systems. In addition, the discussion addresses practical implications for the use of AI in clinical environments, including transparency, trust, and integration into existing workflows. The talk further discusses how explainable and auditable AI architectures can contribute to trustworthy decision support and strengthen digital sovereignty in data-intensive healthcare environments. LLMLLMLLM

https://www.sfscon.it/tracks/llm-track-2026/

Christian Prinz1111:4000:15Efficient management of public large genomic datasets with PostgreSQLhttps://www.sfscon.it/talks/efficient-management-of-public-large-genomic-datasets-with-postgresql/PostgreSQL foreign data wrappers to the rescue !PostgreSQL foreign data wrappers to the rescue !We had a problem: how to efficiently manage public genomic dataset, which may get very large, and integrate them with private genomic information stored in a PostgreSQL database. We leveraged a very powerful, but probably less known PostgreSQL technology called Foreign Data Wrappers, to build an efficient solution: we will describe what we did and how to replicate our solution and our setting. Hospitals, research insitutions, universities and every company that wants to internally manage large genomic datasets as gnomAD, DBSNFP, CADD, Clinvar, etcetera, will find our presentation very interesting. LLMLLMLLM

https://www.sfscon.it/tracks/llm-track-2026/

Michele Finelli1112:0000:35The Last Black Box. Sovereign Minds, Sovereign Code.https://www.sfscon.it/talks/the-last-black-box-sovereign-minds-sovereign-code/The Cognitive Science and Neuroscience Behind Algorithmic Dependency and AI AutomationThe Cognitive Science and Neuroscience Behind Algorithmic Dependency and AI AutomationAs proprietary AI systems increasingly automate our workflows and decisions, we face a critical evolutionary question: Will the human mind remain the last sovereign “black box,” or will we blindly yield our cognitive autonomy to algorithmic black boxes we cannot audit? True digital sovereignty is no longer just about who owns the servers; it is about who controls human agency. When proprietary, closed-source models dictate choices, they exploit human cognitive shortcuts. We stop verifying, we stop questioning, and eventually, we let AI influence our choices without even knowing what happens behind the screen. This talk bridges the gap between open-source tech architecture and cognitive neuroscience. We will explore how vendor-neutral, transparent technology acts as a vital psychological shield. Attendees will discover the neurological impact of algorithmic dependency and learn how open-source frameworks empower developers and leaders to retain intellectual control. It’s time to build systems that protect our ability to think, query, and remain sovereign. LLMLLMLLM

https://www.sfscon.it/tracks/llm-track-2026/

Gaber Marolt11