Digital Science
1 like767 views
This document discusses the importance of reproducibility in astronomy research. It notes that the astronomy research lifecycle is digital but reproducibility is still challenging. Next generation astronomical archives will contain vast amounts of data in new formats, requiring services for knowledge discovery and data mining. The document advocates for archives to provide virtual data and software tasks as web services to enable interconnected and interoperable archives. The Wf4Ever project aims to develop workflows and research objects in astronomy to improve efficiency, reuse, discovery and credit for research. Overall, the document emphasizes the need for astronomy to embrace open science and reproducible research practices in the digital era.
Digital Science
- 1. The Reality of Reproducibility of in silico Science Prof Carole Goble FREng FBCS CITP JCDL Washington DC, June 2012
- 2. X REUNIÓN CIENTÍFICA DE LA SOCIEDAD ESPAÑOLA DE ASTRONOMÍA VALENCIA 9/13 JULIO Digital Science Reproducibility and Visibility in Astronomy José Enrique Ruiz, Lourdes Verdes-Montenegro, Susana Sánchez, Julian Garrido, Juan de Dios Santander and the Wf4Ever Team SESIÓN INSTRUMENTACIÓN Y COMPUTACIÓN VALENCIA, VIERNES 13 JULIO 2012 2
- 3. Digital Science - Reproducibility and Visibility in Astronomy Astronomy Research Lifecycle Astronomy research lifecycle is entirely digital » Observation proposals » Data reduction pipelines » Analysis of science ready data » Catalogs of objects and data » Publish process › Final data results › Experiment in DL ADS/arXiv Reproducible research is still not A normalized preservation of possible in a digital world methodology is needed A rich infrastructure of data (VO) Tools is not efficiently used 3
- 4. Digital Science - Reproducibility and Visibility in Astronomy The next generation of archives Much wider FoV and spectral coverage » Large volumes for an observed datacube » Subproducts are Virtual Data generated on-the-fly ASKAP Cubes Prof. Kevin Vinsen 4
- 5. Digital Science - Reproducibility and Visibility in Astronomy The next generation of archives Automated surveys » Huge amounts of tabular data » Services for KDD Extraction of scientifically relevant information from a multidimensional parameter space » Exploration services » Anomaly detection » Cross-matching data » Dimensionality reduction 5
- 6. Digital Science - Reproducibility and Visibility in Astronomy The next generation of archives » A cloud of Web Services Archives should evolve from data providers into » Virtual data providers » Software tasks providers » Archives speaking Web Services Astronomy of multi archives/facilities/wavelength Interconnected and interoperable archives » Software Tasks Preservation » Data Process should benefit of the same privileges acquired by data Preserving the method ensures replication of final results at any moment 6
- 7. Digital Science - Reproducibility and Visibility in Astronomy Efficiency and Reuse Optimize return on investments made on big facilities » Avoid duplication of efforts and reinvention » How to discover and not duplicate ? » How to re-use and not duplicate ? » How to make use of best practices ? » How to use the rich infrastructure of data ? » Intellectual contributions are encoded in softw More data in archives does not imply more knowledge » Time has come to go beyond the PDF » Expose complete scientific record, not the story » Allow easy discovery of methods and tools 7
- 8. Digital Science - Reproducibility and Visibility in Astronomy Reproducibility and The Scientific Method Benefits » Publishing knowledge, not advertising » The author, the referee and the re-user » Reputation, prestige and respect » Higher quality of publications › Authors will be more careful › Many eyes to check results Challenges » Hard and time consuming » Need incentives – not rewarded now Initiatives » Elsevier Executable Papers Challenge http://xkcd.com/242/ » Open Data / Open Science 8
- 9. Digital Science - Reproducibility and Visibility in Astronomy Reproducibility and The Scientific Method I don’t know how 9
- 10. Digital Science - Reproducibility and Visibility in Astronomy Discovery, Visibility and Credit 10
- 11. Digital Science - Reproducibility and Visibility in Astronomy Discovery, Visibility and Credit 11
- 12. Digital Science - Reproducibility and Visibility in Astronomy Discovery, Visibility and Credit 12
- 13. Digital Science - Reproducibility and Visibility in Astronomy Discovery, Visibility and Credit Exploring and understanding scientific metrics in citation 2010 Krapivin et al. 13
- 14. Digital Science - Reproducibility and Visibility in Astronomy Discovery, Visibility and Credit Paper discovery: the social dimension #SEA2012 14
- 15. Digital Science - Reproducibility and Visibility in Astronomy The Wf4Ever Project EU funded FP7 STREP Project December 2010 – December 2013 1. Intelligent Software Components (ISOCO, Spain) 2. University of Manchester (UNIMAN, UK) 2 3. Universidad Politécnica de Madrid (UPM, Spain) 7 5 4 4. Poznan Supercomputing and Networking Centre (PSNC, Poland) 5. University of Oxford (OXF, UK) 6. Instituto de Astrofísica de Andalucía (IAA, Spain) 31 7. Leiden University Medical Centre (LUMC, NL) 6 15
- 16. Digital Science - Reproducibility and Visibility in Astronomy Scientific Workflows Living Tutorials Templates for Re-use Expedites Training Reduce time to insight Avoids reinvention Digital Libraries of workflows may boost the use of the existing infrastructure of data (VO) 16
- 17. Digital Science - Reproducibility and Visibility in Astronomy Scientific Workflows ! Survey in the domain of astrophysical workflows ! Scientific › Personal script-based recipes Insight • Python, IDL, Software.. › Multi-archive VO recipes • Euro-VO, IVOA.. › Internal group developments Accessible • GRID, Clusters, Specific knowledge.. Shareable › Processing pipelines Reusable • Facilities provide data, computing infrastructure, tools.. Adaptable Understandable » Clarity (workflows) for re-use and re-porpuse vs. automation (pipelines) » A black box is not re-usable, cannot be broken into parts » Reproducibility vs. industrial paper publishing 17
- 18. Digital Science - Reproducibility and Visibility in Astronomy Research Objects Organization is more sexy than automation Assistive building Completeness evaluation 18
- 19. Digital Science - Reproducibility and Visibility in Astronomy Research Objects Expose experiment in a structured way in order to be understood Technical Objects Social Objects Distributed 19
- 20. Digital Science - Reproducibility and Visibility in Astronomy Research Objects ! Similar initiatives in Astronomy ! » Semantic curation of digital objects › CDS Centre Données Strasbourg › US Virtual Astronomical Observatory › SAO/NASA ADSLabs » Workflow users platforms › Cyber-SKA › IceCore › Montage › Astro-WISE › Helio-VO » Semantically auto descriptive WS › Workflows VO-France 20
- 21. Digital Science - Reproducibility and Visibility in Astronomy Research Objects ADSLabs Initiative ADO Linked Components » Authors » Publications » Journals » Objects SIMBAD » Tabular data behind the plots CDS » ASCL reference of used software » Observing time Proposals » Used facilities, surveys or missions Incentives http://labs.adsabs.harvard.edu/ 21
- 22. Digital Science - Reproducibility and Visibility in Astronomy Research Objects ! The Incentive ! Papers with data links are cited more than those without Effect of E-printing on Citation Rates in Astronomy and Physics 2006. Edwin A. Henneken et al. 22
- 23. Digital Science - Reproducibility and Visibility in Astronomy The Wf4Ever Project » Development of AstroTaverna plugins to access and manage VO data » Development Golden Exemplars of astrophysical Workflows and Research Objects that use the Wf4Ever technological support › Curation of physical quantities in 1D catalogues • Data retrieved from external repositories and stored locally • Only local processes for calculations › Environment and Modelling from 1D catalogues and 2D images • Data retrieved from external repositories (SDSS DR7) • Local software and external web services as processes › Modelling and Analysis of 3D formatted data • Only external data and processes 23
- 24. Digital Science - Reproducibility and Visibility in Astronomy Astronomical Research Objects in Action ! Curation by inspecting propagation of changes in quantities ! Credit: Zsolt Frei and James E. Gunn. The Galaxy Catalog 24
- 25. Digital Science - Reproducibility and Visibility in Astronomy Astronomical Research Objects in Action AMIGA Catalog Panchromatic properties for a sample of the most isolated nearby galaxies How is the User DB affected ? - Changes in External DB Evaluate variations with time - Modifications in Calculations - Modified External Quantities - Affected User DB Quantities Update External DB Calculations User DB 25
- 26. Digital Science - Reproducibility and Visibility in Astronomy Astronomical Research Objects in Action Curation by inspecting propagation of changes in quantities ! ! Multi-workflow Research Object 26
- 27. Digital Science - Reproducibility and Visibility in Astronomy Astronomical Research Objects in Action Create, annotate and run a workflow ! ! 27
- 28. Digital Science - Reproducibility and Visibility in Astronomy Astronomical Research Objects in Action Populate the Research Object and annotate Extract ! 28
- 29. Digital Science - Reproducibility and Visibility in Astronomy Astronomical Research Objects in Action Add documents and references 29
- 30. Digital Science - Reproducibility and Visibility in Astronomy Astronomical Research Objects in Action Create and explore relations among components ! ! 30
- 31. Digital Science - Reproducibility and Visibility in Astronomy Astronomical Research Objects in Action Add schema of the experiment 31
- 32. Digital Science - Reproducibility and Visibility in Astronomy Astronomical Research Objects in Action Publication for later discovery Import and re-use ! 32
- 33. Digital Science - Reproducibility and Visibility in Astronomy Astronomical Research Objects in Action ! Curation by inspecting propagation of changes in quantities ! » Taverna 2.3 » MyExperiment Pack › http://www.myexperiment.org/packs/231 Related Publication The AMIGA sample of isolated galaxies XI. A First Look at Isolated Galaxy Colors 2012 A&A 540, A.47 33
- 34. Digital Science - Reproducibility and Visibility in Astronomy Conclusions How NOT to be a good e-astronomer » Search the beautiful plot for high impact instead of real scientific results » Write a obscure paper, do not say clearly how to reproduce the results » Do things quickly and forget about them once you’ve submitted the paper » Be untidy, spread your code and data in a variety of formats, folders and disks » Practise the “data mine-ing” – input data are mine » Practise the “data flirting” – call me if you would like to have more » Do not provide data results, including the plots is just fine » Always cite the same authors and papers or those that cite you » Do not cite other resources than papers, neither provide their URL links » Do not search info on Internet with other tools than ADS or arXiv » Work alone and email/phone one friend if you have any doubt http://amiga.iaa.es/p/212-workflows.htm http://www.wf4ever-project.org jer@iaa.es bultako 34