“Open by Default”

Developing reproducible, computational research


Dani Arribas-Bel
@darribas
Martin Fleischmann
@martinfleis

Reproducible research

  1. Transparent
  2. Accessible
  3. Transferrable

“Open by default”

  • Use (and nurture) the existing OSS infrastructure
  • Treat research artifacts as OSS from Day 1
  • Only “stay closed” when there is a good reason

For example…

Urban Grammar

A characterisation of space based on form and function designed to understand urban environments

+ Satellite imagery & AI

Needs

  • Data, data, data
  • (New) methods, (new) code
  • Infrastructure
  • (Academic) dissemmination
  • Open Data Products
  • Journaling/reporting
    		•

    The kitchen…

    Data

    • Standard (formats) better than niche
    • Open better than proprietary
    • Static better than DB

    E.g., parquet, COGs

    New methods

    New code

    Infrastructure

    … and the “sausage”

    Dissemmination (I): Papers

    Dissemmination (II): Talks

    Dissemmination (III): Other

    Open Data Products (I): Data

    Open Data Products (II): Exploratory tools

    The takeaway

    • “Release early, release often”
    • Make open a feature, not a bug
    • Share your knowledge, it is a way to achieve immortality (Dalai Lama)

    A bit more to read…

    Rey, S., Arribas-Bel, D., & Wolf, L. (in press). Computational Tools for Geographic Data Science. Chapter 2 in Geographic Data Science with Python. Available at: https://geographicdata.science/book/notebooks/02_geospatial_computational_environment.html

    Boeing, G., & Arribas-Bel, D. (2021). GIS and Computational Notebooks. Geographic Information Science & Technology Body of Knowledge, 2021(Q1).

    Arribas-Bel, D., Green, M., Rowe, F., & Singleton, A. (2021). Open data products-A framework for creating valuable analysis ready data. Journal of Geographical Systems, 23(4), 497-514.

    https://urbangrammarai.xyz

    “Open by Default”

    Developing reproducible, computational research


    Dani Arribas-Bel
    @darribas
    Martin Fleischmann
    @martinfleis