Presentations
This session will explore the history of digital resistance through open-source tools, examine different open government approaches with Decidim. We will delve into real-world use cases from Spain, the USA, and Brazil. We will look how local governments have accelerated democratic engagement through open public digital infrastructure. Finally, we will reflect on how we can contribute to building this infrastructure and shaping the future of digital democracy at our level.
While x86 servers certianly dominate the landscape in datacenters, they are not the only option for running an OpenStack environment. We'll discuss some options for installation, and then examine the concrete example of how a dual architecture enivornment works in the IBM Storage Scale lab.
We all know honeypots can reveal interesting details about threat actors and there tactics, but it’s not every day that a threat actors sends you their own credentials. Operational security is hard. In this session, I’ll share how my team and I developed a simple Flask application to emulate an exposed Docker endpoint, and how an everyday log review led to discovery the X-Registry-Auth header. The header turned out to be a DockerHub token. I’ll take you down the rabbit-hole on how my team and I pivoted for additional research and derived some level of attribution.
This talk explores the transition to conscious, dynamic systems and business models in the era of intelligence. By rethinking how workflows, compute, and context flow seamlessly, we move beyond the band-aid approach of containerized apps forced to orchestrate. Instead, we embrace dynamic, natural, adaptive continuums that form the foundation for equitable, scalable, and sustainable agentic ecosystems. Highlighting the diversity of AI, expanding beyond LLM-generated responses from a single vendor to a butterfly effect triggered by events that transcend the physical world.
At the CELEHS laboratory we are particularly interested by LLM-based embeddings as BGE and BERT. As the number of models increases, we need methods to compare their clinical usefulness. While some R packages exist to leverage GPU capabilities, Pytorch is by far more used for GPU computation. In contrast, R is efficient for data management and visualization. How should one build robust and reproducible pipelines incorporating them both ? My answer is well-designed pipelines with Docker, Makefile, and Elasticsearch. In this talk I will showcase my design approaches to such challenges.
Data flows continuously—and so should your analyses. In this session, learn about the fundamentals of stream processing and how it’s transforming industries like finance, e-commerce, and IoT. We'll dive into commonly used, open-source tools like Kafka, Flink, and RisingWave. Then we'll build out a simple data pipeline, showing how to wrangle endless data streams into manageable, actionable insights instantly. Whether you're starting out, or improving your approach, join us to learn about the potential of real-time data.
In the before times, OpenStack operators met several times a year to network, exchange experience and to bring feedback to the OpenStack development community based on experience of using OpenStack in production at scale. Since the pandemic, and with shrinking travel budgets, in-person meetups have become extremely rare. On the other hand, interest in OpenStack is growing again.
We'll take a fresh, no-holds barred look at the status quo and discuss what works for the community in this new reality.
There was a time when people wrote software for their own needs. Knowing that selling software is hard (and to a very small market compared to today's markets) they shared their software with "Birds of a Feather" to help others who neeed it.
Over time, markets grew and the desire to make money pulled the making of software from being fun to being a (sometimes mind-numbing) job.
Can we Make Software Fun Again? (MSFA)?
A detailed explanation of how pull requests are staged and trickle into the different channels in Nixpkgs and NixOS.
Join us for a presentation discussing the unique considerations of moderating a code collaboration platform. Using diverse case studies, we'll look at how GitHub has refined its approach to developer-first content moderation in response to technological and societal developments as well as notable incidents on the platform. This presentation is a condensed version of the recently published T&S Research Journal article, “Nuances and Challenges of Moderating a Code Collaboration Platform” authored by members of the GitHub Trust and Safety, Legal, and Policy teams.