Open Research Data Projects

One of the main forms of ORD in the programming languages research community is two-sided mechanized language specifications (the definition in a proof assistant of the semantics of a language). These mechanized specifications have many benefits: they can be extracted to an executable reference implementation, and used by both implementers (to verify compilers, interpreters and optimizations) and by users (to guarantee the correctness of programs in that programming language).

We propose to contribute the first open, two-sided, mechanized specification of JavaScript regular expressions (regexes). The lack of such mechanization is harming the research community: previous work has mechanized other parts of the JavaScript language but not regexes, and as a consequence researchers use paper-only semantics for JavaScript regexes. These paper semantics are neither executable nor reusable and often incorrect. We will translate to Coq the part of the open ECMAScript standard that describes JavaScript regexes; extract this mechanization to a reference implementation in OCaml; and validate our mechanization with Coq proofs.

Our project will provide a solid foundation for JavaScript regex research to build upon our Coq mechanization, including proving the correctness of regex optimizations, detecting regexes with security issues (ReDOS), or proving the correctness of entire regex engines. Our project will allow the open JavaScript community to test their regex engines.

Brain-Score is an established platform which curates a diverse set of neural and behavioral measurements from neuroscience experiments and facilitates its use in modeling the brain's visual system. By making experimental data accessible to the modeling community in the form of quantitative benchmarks, Brain-Score allows modelers to evaluate computational hypotheses on a broad range of biological data without having to know the details of each experiment. In this proposal, we aim to broaden the scope of Brain-Score model comparisons from the presentation of static images to video inputs. This will enable the modeling of a critical axis of brain processing in visual cortex that has not yet been explored.

Specifically, we will:

* Contribute new software to Brain-Score to enable the platform to work with temporal data. This involves defining a unified interface for how to provide models with video input, and adding candidate video models from the machine learning community.

* Curate published temporal datasets for Brain-Score. Without Brain-Score, even these public data are often difficult to use for model testing.

* Curate new primate recordings from experimental collaborators (MIT DiCarlo lab) for Brain-Score such that they are accessible for model evaluations. These are among the first electrode recordings in the visual ventral stream where the stimuli are short ecological video clips.

A Speckle connector for Open Cascade Technology will enhance software and data interoperability within the architecture, engineering, and construction industry. The project aims to bolster the use of free software in the sector, by uniting the capabilities of two open-source ecosystems in a sector where proprietary tools are currently very dominant. On the one hand, Speckle open-source connectors enable seamless collaboration across diverse AEC software; it ensures accurate and efficient collaborative workflows between various actors and disciplines, thus contributing to the freedom of choice of digital tools, avoiding a captive market. The connector's open-source design encourages community contributions, fostering continual improvement. On the other hand, Open Cascade Technology is an open-source geometric kernel. It is used in free software alternatives such as Freecad or Salome, and in open-source libraries such as IfcOpenShell, which allows the development of applications based on IFC, the open standard for Building Information Modeling. Also, the open-source nature of Open Cascade Technology enables many researchers and professionals to develop their own highly specialized digital tools. Our prototype would connect this ecosystem to all AEC industry software, via Speckle.

Established in 1975, the European Long-term Ecosystem Research (eLTER) facility, Stillberg, in the alpine ecosystem near Davos, Switzerland, has amassed extensive environmental and ecological data over almost five decades. These encompass treeline afforestation experiments, meteorological records, plant responses to carbon dioxide enrichment, soil warming effects, plant-snow and plant-soil interactions, and factors influencing tree seedling recruitment. Our project's aim is to contribute these valuable ecological datasets from Stillberg as open research data (ORD). By meticulously curating these datasets and uploading them to national and international ORD platforms like EnviDat and DEIMS-SDR, we enhance their visibility, quality, and accessibility. Sharing this long-term environmental data fosters research syntheses, meta-analyses, and understanding of long-term ecosystem processes in mountain regions, supporting adaptation strategies.

This project focuses on creating software containers crucial for the global seismological community. These tools, widely used but often complex to compile due to dependencies, will be encapsulated in virtual environments to ensure seamless interaction and ease of use. The portable containers will enhance software sharing and scientific reproducibility. The project involves a postdoctoral researcher and a doctoral student, and Continuous Integration and Continuous Development pipelines will maintain open access repositories. A dedicated workstation with GPU-accelerated hardware will support this. At the project's midpoint, a workshop with software developers will gather feedback and boost container awareness, which will guide final enhancements.

Switzerland's dominant tree species exhibit masting behavior, characterized by irregular, sporadic seed production patterns. These patterns significantly impact tree regeneration and ecosystem dynamics. Environmental factors and climate change influence these patterns. Despite their ecological importance, many phenological networks overlook seed masting. Our workshop aims to unite managers from various phenological networks to establish an ORD protocol for collecting seed mast data. This standardized method can be integrated into existing phenology networks. Researchers, including modelers, physiologists, and ecologists, will benefit from this comprehensive data collection. The curated data will be publicly available on MastWeb's ORD platform, fostering collaborations with global phenology networks.

Image data are essential in scientific research, from astronomy to microbiology. Advancements in technology have enabled the generation of vast and informative image datasets. To derive valuable insights from these datasets, open distribution is crucial. While centralized efforts exist for hosting open image data, many researchers struggle to prepare and share their data in a Findable, Accessible, Interoperable, and Reproducible (FAIR) manner due to data size, various formats, and complex analysis methods. This project's goal is to provide an open-source training resource that educates researchers on processing and preparing image data with best ORD practices. An online handbook will guide users on image ORD best practices and curate existing image ORD resources. Workshops based on the handbook will be conducted to train researchers and establish community consensus on image ORD best practices. These training materials aim to empower ETH domain researchers to effectively utilize ORD resources.

The raindrop size distribution (DSD) details raindrop concentration and size distributions in an air volume. It's vital for rainfall microstructural analysis, remote sensing interpretation, and accurate representation in atmospheric models. Disdrometers collect DSD observations globally, but data are dispersed, varied in format, and lack standardized tools for processing. As a result, large-scale DSD spatial and temporal variability exploration is challenging. DISDRODB addresses this by advocating common standards for data format, quality control, and processing. It establishes a database and processing framework to store shareable raw measurements, generate clean DSD data, and derive related products (e.g., rain rate, kinetic energy, mean size). This tool benefits scientific communities working with DSD for process understanding, remote sensing, and modeling.

This proposal aims to significantly expand the Solar Fuels Database, which introduced a machine-readable framework for solar-to-fuel devices and an online data entry and visualization interface. The project's objective is to broaden the database to include additional solar fuels and technological pathways not covered previously. It will encompass various fuels such as carbon monoxide, syngas, formic acid, methane, ethanol, and extend to thermochemical redox cycles. Collaboration with global research communities will ensure comprehensive metadata capture. Standardized data reporting will enhance findability and dissemination of results. The project aligns with open-science objectives and encourages community contributions, resulting in a continually updated and openly accessible resource. This resource will consolidate prior work and offer a comprehensive overview to drive future research trends in advancing solar fuels for widespread implementation.

Today, more and more digital data are generated by urban dynamics. Yet, the generated data is extensive and heterogenous. Datasets are large, multi-sourced, often noisy, and come in various formats and standards. In addition, a particularity of urban data is its mix in terms of level of restriction. While geolocation data is private and sensitive, public transit schedules are open data. In this context, there is a need (i) to provide a framework to re-unify the multiformity of urban dynamics data, (ii) to articulate open and restricted data, (iii) to cohere offer and demand data, and (iv) to keep track of a privacy metric. This project proposes to develop and release an open Python package to address these four needs and therefore contribute to Urban Mobility Open Research Data practices.