Want to share your screen? See the person you're talking to? Contact us via digital library desk! We will be with you shortly.
Monday-Friday
Want to share your screen? See the person you're talking to? Contact us via digital library desk! We will be with you shortly.
Monday-Friday
Here you can find a list of events previously held or organized by Lib4RI. If available, course materials, screen casts, and presentation slides are included.
Behind the bookshelves - Recordings
Speaker: Jeroen Sondervan, Open Science NL
Geopolitical tensions and emerging threats increasingly shape the global research landscape and its supporting infrastructures, as national security concerns and financial constraints limit the open exchange of knowledge. At the same time, addressing global challenges—from pandemics and climate change to rapid technological transformation—depends on strong international scientific collaboration that remains as open and transparent as possible. This contribution examines how Open Science, with its emphasis on transparency, collaboration, and the reliability of scientific results, can be protected and sustained in these turbulent times, and how it may help increase resilience within the research ecosystem.
Download the slides here and find the referenced blog post here.
Speaker: Anina Köppli, Consortium of Swiss Academic Libraries
In this Coffee Lecture, Anina Köppli, Team Lead Negotiations at the Consortium of Swiss Academic Libraries, provides an insider’s perspective on negotiating with major scholarly publishers in a highly concentrated market.
Drawing on the Swiss big deal negotiations the talk explores strategies such as collective action, clear mandates, data-driven preparation, and credible alternatives. The session asks a central question: when is “no deal” a problem - and when can it be part of the solution?
Download the slides here.
Speaker: Katharina Eggenberger, WSL
Consolidating the legal basis for Open Research Data (ORD) is one of the five measures in the ETH ORD programme. Identification of obstacles, clarification of responsibilities, and development of ORD guidelines as common reference within the research institutes are the key objectives here. The guidelines aim to facilitate and promote the making available of research data as ORD by providing a set of rules to be applied. They offer an orientation to support researchers in their ORD-related decisions, based on the ORD strategy of the respective institution.
Download the slides here.
Speaker: Janina Radny, Lib4RI
AI bots are accountable for a significant share of website and API traffic nowadays. This can results in drastically reduced “real human” interaction for websites and repositories, with at paradox increased traffic and associated costs. Instead of hiding content behind paywalls, Creative Commons (CC) propose a new set of licences that express preferences how works can be used in AI training—emphasizing reciprocity, recognition, and sustainability in machine reuse. In contrast to the well established CC licences, CC signals is not targeting the individual author but rather stewards of content and collections of data. In this coffee lecture we present the idea of CC signals and want to start a discussion with you on applicability and consequences for research data collections.
Download the slides here.
Speaker: Olivia Denk, Swiss academies of arts and sciences
In this coffee lecture, presented by the co-coordinator of CoARA’s Swiss National Chapter, researchers were introduced to CoARA’s mission to reshape research culture. The presentation focused on the initiative’s vision and current activities of its communities, like Working Groups and National Chapters. Additionally, the role of Open Science in transforming research assessment was highlighted.
Download the slides here.
Open Access Week - Recordings
“Who owns our knowledge?” – The motto of this year’s Open Access Week raises a question that has been at the core of the Open Access movement. The rise of AI and large language model (LLM) training and use brings a new urgency to this question. In this talk, David Rosenthal shared his perspective as expert lawyer on this field. Specifically, he focused on how today's copyright law deals with the training of large language models with third party content (what is permitted?), how well copyright law is prepared for fairly balancing the interests of rights holders models and those using AI and what we have to expect from a legal point of view.
Speaker: David Rosenthal counts to the leading experts in Data and Technology law in Switzerland and is partner at the Swiss law firm Vischer. His view on the topic is broadened also by his earlier work as software developer, journalist and legal advisor.
In several blog entries, he and his colleagues focus specifically on the interplay of AI/LLM use with Swiss and EU copyright, data, and privacy laws. In cooperation with the ETH AI center, they also authored a detailed legal analysis about LLM training with regard to the Swiss legal system. On behalf of the Consortium of Swiss Academic Libraries, he played a key role in shaping the AI clause in the publishing agreement with Wiley, signed earlier this year.
His publication on challenges for media houses in an era of 'agentic AI' services even led to discussions in the Swiss National Council on how to amend Swiss copyright law.
Recording and slides for the talk AI & Copyright - friends or foes? © 2025 by David Rosenthal are licensed under CC BY-NC-ND 4.0
Panelists:
Guest (University of Zürich): Prof. Dr. Torsten Hothorn
Professor of Biostatistics at the Epidemiology, Biostatistics and Prevention Institute, Co-Editor-in-Chief for the Diamond OA Journal of Statistical Software
Eawag: Prof. Dr. Juliane Hollender
Associate Editor for Environmental Science & Technology (hybrid)
Empa: Dr. Mateusz Wyrzykowski
Managing Editor for RILEM Technical Letters (Diamond OA)
PSI: Dr. Anuschka Pauluhn
Editor for ISSI Bern , books & public outreach journals for astrophysics & space science
WSL: Dr. Eckehard Brockerhoff
Co-Editor-in-Chief for New Zealand Journal of Forestry Science (Diamond OA), Associate/Subject Editor for other journals incl. Biodiversity and Conservation & New Zealand Entomologist
As our contribution to Open Access Week 2023 and as part of WSL Distinguished Lectures, Lib4RI hosted a lecture by Matthias Egger, President of the National Research Council of the SNSF.
Abstract
In recent years, Open Access (OA) publishing has experienced rapid growth, witnessing the emergence of numerous new publishers and journals. In this talk, I will explore the benefits, challenges, and potential pitfalls of OA publishing from the perspectives of both an active researcher and a research funder. I will briefly revisit the well-rehearsed arguments for OA and discuss some challenges and abuses, such as peer review quality or predatory practices. Finally, I will conclude with a discussion of best practices and alternative publishing models.
The slides and video of Matthias Egger's talk “Open access: the good, the bad and the ugly?” are only available within the four Research Institutes! Please connect to the server via cable or VPN.
Love Data Week - Recordings
Speaker: Felix Moerman, Lib4RI
Data management plans (DMPs) are an essential tool for preparing the collection, storage, and processing of research data. Yet the creation of DMPs can often feel cumbersome. Differences of expectations by funding agencies can result in duplicated efforts, and tracking changes to the DMP can be laborious.
To facilitate this process at the ETH-domain research institutes, Lib4RI has prepared a unified DMP template on Data Stewardship Wizard. This demonstration aims to show how this template can be used to a) more easily track changes in the DMP, b) allow for easy export to multiple formats/funder requirements, and c) minimize work effort for creating DMPs.
Speaker: Corsin Battaglia, Empa
I will present an overview of our efforts in developing Aurora, an automated robotic battery materials research platform, designed to accelerate the validation of novel battery materials in battery cells. Our platform is powered by automated workflow and data management. I will describe how we structure, annotate, and link our data in compliance with FAIR data principles, employing the BattINFO ontology and how we leverage Bayesian optimization to define a series of experiments. I will conclude by summarizing how we leverage the multimodal self-learning capabilities of large language models in our research.
Speaker: Chase Núñez, Lib4RI
In the fast-paced world of scientific research, where data is often vast and experiments must move quickly, managing and versioning data effectively can become a major challenge. This talk will provide practical, scalable, and sustainable strategies for versioning research data, balancing the need for speed with the imperative of long-term reproducibility and reusability. We will explore the core principles of Research Data Management (RDM), including how to preserve raw data, record transformations, and leverage tools like Git Large File Storage (LFS) for large datasets. Through real-world examples, we will demonstrate how to integrate version control into your everyday workflows, ensuring that your research remains accessible, reproducible, and ready for future discovery. Key topics will include how to maintain an immutable record of raw data, how to document data transformations, and the benefits of using institutional and external repositories for publishing datasets. By the end of the session, you will leave with a checklist, practical commands, and a clear strategy for improving the versioning of your own research data.
Speaker: Séverine Duvaud, Swiss Institute of Bioinformatics
Galaxy is an open, user‑friendly platform for reproducible data analysis. This talk introduces Galaxy Swiss, a national instance designed to support researchers across Switzerland. After a brief overview of the Galaxy project and its mission, we will walk through the interface and demonstrate a simple analysis workflow. The session concludes with practical guidance on how to register, access training resources, and join the Swiss Galaxy community.
Speaker: El Knappe, Lib4RI
Ever had code that worked perfectly on your machine but broke when a colleague tried to use it? In this short introductory talk, we'll explore how virtual environments and Docker containers help you package your code neatly for sharing. No deep dives into configuration files or container orchestration—just the core concepts you need to understand why these tools matter and when you might reach for them. Perfect for anyone curious about making their code more shareable and reproducible.
Speaker: Moushumi Ulrich-Nath, Lib4RI
Date: Monday, 10 February 2025
Abstract: Do you want to improve and transform your research data management? Join us for a webinar introducing the Data Management Campus!
What is it?
Interactive eLearning modules and open training resources on Open Research Data (ORD) and Research Data Management (RDM)
Why join?
Learn about our first module on Data Publication and Long-Term Preservation.
Discover how to access, contribute, and shape the future of research data excellence.
Whether you’re a curious researcher or a data management enthusiast, come explore our collaborative ETH Domain initiative designed to support you throughout your research data journey!
Speaker: Janina Radny, Lib4RI
Date: Tuesday, 11 February 2025
Abstract: Quarto is an powerful tool for data analysis and publishing, offering seamless integration with R, Python, and Julia. It enables the creation of dynamic, reproducible reports with embedded code, visualizations, and rich narratives all in one document. With Quarto, you can easily publish interactive documents or websites, facilitating a smooth collaboration process and instant sharing of your work. The platform ensures a clean workflow, allowing for integration with version control tools like Git and making your analysis fully traceable. Additionally, Quarto is open-source, free to use, and supports a wide variety of output formats, including HTML, PDF, and slides, giving you flexibility in presenting your results. In this coffee lecture we give a brief first look, using R and RStudio.
Speaker: Stefanie Hauser, Empa
Date: Wednesday, 12 February 2025
Abstract: In 2020, openBIS was introduced at Empa, making it accessible to all departments and available for testing. This platform, provided by ETH Scientific IT Services (SIS), combines an electronic laboratory notebook (ELN) with a laboratory information management system (LIMS). Despite careful preparation, we encountered unexpected barriers, showing that changing existing systems requires more than tools—it demands effort, education, and sometimes a shift in mindset. In this presentation, I’ll highlight the challenges we encountered and the strategies we use to overcome them, paving the way for digitalized laboratory workflows.
Speaker: Elisabeth Capon Garcia, Swiss Data Science Center
Date: Thursday, 13 February 2025
Abstract: Data-centric research requires sharing not only the data, but also the code and computational environments needed to process and analyze the data. In this talk, we will present Renku, an open-source platform that empowers transparency, collaboration, and reusability in all types of data-centric research. In Renku projects, researchers connect their assets, namely data and code with computational environments, all in one place. These assets are designed to be reusable from the start, so they can be shared, reused and traced between projects. Since Renku projects are interactive in the browser, they are easily shared with anyone to make collaboration smoother. Overall, Renku supports the research lifecycle of data-centric projects, from day-to-day work to demonstration and publication.
Speaker: Federico Cantini, Lib4RI
Date: Friday, 14 February 2025
Abstract: Tracking the changes you make to your code and uniquely versioning it is paramount to reproducibility. Keeping multiple branches of your code and being able to easily jump from one to another is useful for exploring different data processing or collaborating with others. Git does this for you easily and reliably. Whether you write complex software or simple scripts in your language of choice (R, Python, etc…), whether you are a lone code writer or part of a team, Git will make your life easier.
Please note: The videos of these coffee lectures are only available within the four Research Institutes! Please connect to the server via cable or VPN.
Speaker: Reza Ali Rezaee Vahdati, PSI
Date: Monday, 12 February 2024, 13:00-13:15
Abstract: The talk will introduce open science and data management strategies that enable transparency, research integrity, and knowledge transfer within the context of the Circular Economy and the ReMade@ARI project. It will detail the development and implementation of a comprehensive Data Management Plan aligned with FAIR principles, as well as strategies for open access publishing. The talk will also expand on how research infrastructures integrate local data policies with open science principles.
Speaker: Stuart Dennis, Eawag
Date: Tuesday, 13 February 2024, 13:00 - 13:15
Abstract: There is a common misconception that High Performance Computing is only for “big data”, huge datasets that require hundreds or thousands of CPUs, and using HPC is too complicated for “normal’ data.
In this coffee lecture I will dispel this notion and show that HPC could be useful for your data-loving needs, and might even help you to time-travel.
Speaker: Martins Zaumanis, Empa
Date: Wednesday, 14 February, 13:00-13:15
Abstract: Poor data charts are hindering the effective transfer of knowledge in academia. On the contrary – clear charts can help your research papers get noticed and your presentations - understood. Martins Zaumanis will show his 8-step method for creating self-explanatory data visualizations. Download a cheat sheet summarizing the 8 steps here.
Speaker: Matthias Rösslein, Empa
Date: Thursday, 15 February, 13:00-13:15
Abstract: In any HorizonEurope projects Data Management Plans (DMPs) and FAIR Data treatment are now mandatory. Hence each project partner has to contribute, update the DMP on a regular basis, and implement essential parts of FAIR data treatment. We explain, what has to be done, so that you can budget the work accordingly - otherwise you will have to it for free.
Speakers: Mathias Bavay, SLF and Ionut Iosifescu, WSL
Date: Friday, 16 February, 13:00-13:15
Abstract: The data life cycle for FAIR (Findable, Accessible, Interoperable and Reusable) environmental data timeseries is complex. For example, before such timeseries can be made accessible in EnviDat (www.envidat.ch) as FAIR data, they need to be properly understood (in terms of which parameters and which units), be quality controlled, converted into file formats suitable for sharing and reused and enriched with metadata that accurately provide all the necessary information. In this context, the use of MeteoIO (an open-source meteorological data processing library) and NEAD (the non-binary environmental archive data format) can add value to the data by standardisation, metadata enrichment, quality control and potential corrections in a fully reproducible and documented way. This is the main idea behind the Data Life Cycle Integrated System for Sensor Data at WSL/SLF project (DLCIS).
AI tools & LLMs - Recordings & material
As the demand for information and better understanding of AI tools, their capabilities, limitations and implications for academia and the publishing domain continues to surge, Lib4RI is excited to announce a lecture series for end of 2023 and early 2024!
The rise of AI technologies presents both exciting opportunities and challenges for academia and the publishing world. While these tools can enhance research, streamline content creation, and automate certain tasks, ethical concerns and questions of academic integrity have arisen. There are ongoing debates surrounding issues like authorship, plagiarism detection, and the impact of automated content generation on traditional publishing practices. Our lecture series will address these important concerns and provide insights on how to navigate this evolving landscape.
In this presentation, Mark Cieliebak sheds light on recent developments in Generative AI, with a strong focus on Large Language Models (LLMs) such as ChatGPT. This covers topics as broad as text summarization, essay generation, or chatbots. He will introduce the underlying technologies, present several showcases, and assess the potential of LLMs in future research and development. The core algorithms will be explained on a level that is also suitable for a non-technical audience.
The slides and video of Mark Cieliebak's talk "Language Models. Fundamental technologies and applications" are only available within the four Research Institutes! Please connect to the server via cable or VPN.
The second lecture in this series by Dr. Sampoorna Rappaz will specifically explore the role of LLMs in academia, addressing crucial themes such as: the impact of LLMs on academic writing, the ethical and practical considerations when integrating LLMs into academic work, and the opportunities and challenges presented by this transformative technology.
The slides and video of Sampoorna Rappaz' talk "What you need to know when using AI for academic writing" are only available within the four Research Institutes! Please connect to the server via cable or VPN.
In this talk, Sandra Marmy-Brändli will elaborate on the potential copyright issues researchers may face when using AI tools in their research. First, she will look at the question of whether AI-output can be protected by copyright law and if so under what conditions. Secondly, she will take a deeper look at potential copyright obstacles for training AI tools with copyright protected material, as well as potential defense strategies. Lastly, she will review the potential risks of infringing third party copyrights by using output created by AI tools.
The main aim of the presentation is to provide an overview over the legal situation in Switzerland; however, she will also shed light on the situation in the EU and briefly review some of the most recent ongoing AI cases from around the world, namely the pending AI law suits in the United States.