# [Open (data) science](#/1/) # [(free and) open-source software](#/2/) # and # [doing GIS in code](#/3/)

# Open (data) science ## “A large proportion of practical quantitative work in geography relies on the analysis of data or on the running of simulation models. That analysis, and the results it generates, are the outcome of a process involving data verification, re-formatting, computer programming, modelling, data analysis and visualisation.” (page 607) (from Harris R, D. O’Sullivan, M Gahegan, M Charlton, L Comber, P Longley, C Brunsdon, N Malleson, A Heppenstall, A Singleton, D Arribas-Bel and A Evans. 2017. [More bark than bytes? Reflections on 21+ years of geocomputation](https://journals.sagepub.com/doi/abs/10.1177/2399808317710132). *Environment and Planning B: Urban Analytics and City Science* **44**(4):598–617.)

## Data verification ## Reformatting ## Computer programming ## Modelling ## Data analysis ## Visualisation

# Reproducible research ## Requires that *all* of these stages be reproducible by others

## Why not also in ‘real-world’ settings? ## Perhaps even *especially* in public policy contexts?

# What is required? ## Open data ## Open publication ## *Open software*

# (Free and)
open-source software ## (F)OSS

Free as in speech vs free as in beer

source gnu.org

image source commons.wikimedia.org

Isn’t proprietary better?

Cost? NO

Support? MAYBE

Reproducibility? NO

Recent examples I

Images provided by Rubianca Benavidez from the LUCI project

Changes are due to something different between Arc 10.3 and 10.5

Recent examples II

An example of a rapidly resolved issue with open source software

Yayo Kusama infinity mirrored room
image source boredpanda.com

Reproducibility

Can look at the code

Know exactly what it does

Can also potentially contribute over time

QGIS as an alternative to ArcGIS

Essentially can do everything ArcGIS can do

Doing GIS in code

Python

`geopandas`

`pysal`

R

`sp` and `maptools`

More recently `sf` and `tmap`

# Advantages ## Completely flexible and open-ended ## Repeatable ## Often quicker

# Resources ## [Spatial Data Science with R](https://rspatial.org/) website ## [R-spatial](https://www.r-spatial.org/projects/) list of resources ## Lovelace R, J Nowosad and J Münchow. 2019. [*Geocomputation with R*](https://geocompr.robinlovelace.net/). Boca Raton: Taylor & Francis. ## Brunsdon C and L Comber. 2015. [*An introduction to R for spatial analysis & mapping*](https://au.sagepub.com/en-gb/oce/geographical-data-science-and-spatial-data-analysis/book260671). Los Angeles: SAGE.