diff --git a/content/schedule.yaml b/content/schedule.yaml index 0b95b38..3b0f40e 100644 --- a/content/schedule.yaml +++ b/content/schedule.yaml @@ -26,6 +26,12 @@ templates: title: Break short: With coffee + - ®istration + title: Arrival and registration + time: 11:30 + location: main + short: Registration and lunch info + - &lunch title: Lunch time: 12:00 @@ -40,14 +46,20 @@ templates: - &dinner title: Dinner - time: 19:00 + time: 18:00 location: other short: Self-organized - &session title: (session) location: main - #short: To be filled with submitted talks + short: Two talks + + + - &discussion + title: (discussion) + location: main + short: Discussion session schedule: @@ -63,6 +75,7 @@ schedule: # # # # newline + - <<: *registration - <<: *lunch - id: intro1 time: 13:00 @@ -71,12 +84,53 @@ schedule: contributors: Samantha Wittke, Richard Darst abstract: > Introductory words. - - <<: *session - time: 13:10 + - id: kamyar + time: 13:30 + location: main + title: "Sharing GIS Tools Across Disciplines: Hard Choices, opportunities, and Trade-offs" + contributors: Kamyar Hasanzadeh + abstract: > + As GIS methods spread into interdisciplinary research, researchers are increasingly expected to package their workflows as usable software for others. + In practice, this is far from straightforward. This talk reflects on several common ways of sharing GIS tools—commercial extensions (e.g. ArcGIS toolboxes), + open-source plugins (QGIS), standalone desktop applications, web apps, and simply releasing code—and the challenges that come with each. + These include technical maintenance, licensing constraints, usability for non-GIS experts, reproducibility, institutional dependencies, + and long-term sustainability. Rather than advocating a single solution, the talk examines both the advantages and the challenges of each approach, + using these trade-offs as a starting point for discussion on how researchers and research software engineers can make more realistic and + context-aware decisions when sharing GIS methods for interdisciplinary use. + - id: ina + time: 13:30 + location: main + title: "TargetCAT: When a Script Refuses to Stay Small" + contributors: Ina Pöhner, Rafael Lopes Almeida + abstract: > + Imagine a project where a handful of researchers all try to do the same thing - except everyone does it manually, + in their own way, and the one automated step crashes regularly due to poor error handling. Out of frustration with this fragile setup, + TargetCAT was born. What began as a small collection of personal scripts to manage and process data on potential drug targets refused to stay small. + It quietly turned into a research pipeline that enabled several publications and projects, while itself remaining far from ideal in many places. + + + In this talk, we use TargetCAT as a case study to explore how research software typically evolves in academic projects: + how it survives, grows, and gradually accumulates technical debt. We reflect on familiar patterns such as ad‑hoc workflows, + "ghost development" carried out outside funded time, and feature creep driven by scientific needs without corresponding resources. + Along the way, the story touches on identity challenges faced by researchers whose work is effectively research software engineering, + but is assessed through publication‑centred metrics. + + + The second half of the talk turns to a reboot of TargetCAT as an open‑source pipeline for an academic–industry collaboration. + In this part, we share how earlier missteps and constraints, together with a fresh developer perspective and a conscious commitment to + RSE practices from the outset, are shaping its second life. We conclude by teasing lessons learned and opening a discussion on + how academic projects might better plan, fund, and recognise software work - and how this could support more sustainable, RSE‑centric career paths. - <<: *break time: 14:30 - - <<: *session - time: 14:45 + - <<: *discussion + time: 15:00 + - id: poster + time: 16:00 + location: main + title: "poster session" + contributors: Luca Ferranti + abstract: > + TBD - <<: *end - <<: *dinner @@ -93,19 +147,60 @@ schedule: # # # # newline - - <<: *session - time: 9:00 - - <<: *break - time: 10:30 - - <<: *session - time: 10:45 - id: rse-talk title: RSE talk - time: 11:15 + time: 9:30 + location: main + contributors: Jeremy Cohen + abstract: > + Jeremy Cohen is an Advanced Research Fellow in the Department of Computing and Director of Research Software Engineering Strategy at Imperial College London. + He has been involved in the Research Software Engineering community since the early days and held a research software development role in a research group + prior to the existence of the “RSE" term. He has a PhD in Computing from Imperial and held one of the 5-year Research Software Engineering Fellowships (from 2018) + that were funded by the UK Engineering and Physical Sciences Research Council (EPSRC). Jeremy is currently involved in a set of different grants relating to RSE and + the wider “digital Research Technical Professionals” (dRTP) space that’s developing in the UK to recognise not just RSEs but also research data and + research computing infrastructure professionals. He is the PI of STEP-UP (https://step-up.ac.uk), an EPSRC-funded Strategic Technical Platform with a regional focus + on developing skills, community and career pathways for dRTPs. + + + In his talk, Jeremy will discuss how Research Software Engineering has developed within the UK. He will highlight various challenges and opportunities around + developing skills and career pathways for RSEs. He will then look at how the RSE community is expanding to represent a wider group of dRTPs in a range of + technical roles who provide vital contributions to support and undertake modern digital research. + - <<: *break + time: 10:30 + - id: julia + time: 11:00 location: main - #contributors: Name, Name - description: > - TBA. + title: "Towards FAIR file formats: a case example with Origin & Python" + contributors: Julia Niskanen + abstract: > + Open science and FAIR principles have become increasingly important in the last decade. + However, the implementation of Interoperability is sometimes challenged by established software and analytical practices. + One example of such established software is Origin (OriginLab Corporation, Northampton, Massachusetts, USA), + an analysis software with various features involving graphing, statistical operations, and data processing and transformation. + Origin output files are comprehensive and can contain entire analysis pipelines; however, the file format (.opju) is proprietary, + the software is restricted to Windows and requires a paid license to access all features, and there is no easy option to export + all the contents of the file to other formats. Together, these factors impede the implementation of Interoperability. + To overcome these obstacles, I have developed a lightweight Python tool, convert-opju, that can be run within Origin to quickly and systematically export graphs, + images, workbooks, matrices and notes to open file formats. While not all objects of the .opju file are currently included, + converting the major objects to open formats is a notable improvement to Interoperability. Convert-opju is freely available (MIT License) on Github and Zenodo. + - id: frankie + time: 11:00 + location: main + title: "Realization: it's SymPy" + contributors: Frankie Robertson + abstract: > + SymPy looks nice, but it's not a real CAS... is it? + In this presentation I hope to show that as well as SymPy scaling down, SymPy is quite a capable CAS for helping to tackle real world problems we might encounter as RSEs. + The main content of the presentation is a case study of how SymPy has been a useful tool during my first project as an RSE, working on a simulation of a mass spectrometer, + both as a tool for ad-hoc tool enabling DRY and --- using more of its power --- to help design more efficient numerical sampling code. + So next time you have some maths to wrangle, I say: "Go on: treat yourself!" (to SymPy) + - id: outro1 + time: 11:45 + location: main + title: RSE meetup wrapup + contributors: Samantha Wittke, Richard Darst + abstract: > + Closing words. - <<: *lunch @@ -243,6 +338,7 @@ schedule: If there is no explicit requests, it is safe to assume this won't happen. + How can universities get infrastructure funding for local Research Software Engineer support. How can they work together? This is a discussion, taking into account everything we have