• Women in Statistics and Data Science group, Monday February 10th, 12-1pm, 327 Thomas
• Institute for Computational and Data Science (ICDS) events
• “The Data Deluge”, ICDS Symposium, March 16-17, 2020, NLI.

• General comments on homework
• Version control
• Version control with git and GitHub
• RStudio projects

• Where are my data stored?
  • Can I create a reproducible workflow for importing them?
• Automate as much as possible
• Use names that are transparent to others

• What are your document’s dependencies, e.g. packages, external scripts or functions
• Make your code as readable as possible.
  • Consider <COMMAND/CTRL>I to reindent lines
  • <SHIFT><COMMAND/CTRL>A to reformat code
• Plot first, analyze later

• “management of changes to documents, computer programs, large web sites, and other collections of information.” https://en.wikipedia.org/wiki/Version_control
• Track changes an example
• Wikipedia another

• Keep record of who made what changes when
• Revert to previous version
• Transparency

• My typical scheme for collaborative writing
• Requires consistent file-naming (and incrementing) conventions
  • gilmore-etal-nature-2017-01-28-1319.docx
  • gilmore-etal-nature-2017-01-28-1400.docx
  • or gilmore-etal-nature-2020-02-04v01.docx

• No special software needed
• Works with every file type
• Good for any file type

• Not everyone uses the same file naming conventions
• Dealing with conflicts
• File management

• Gilmore, R.O., Diaz, M.T., Wyble, B.A., & Yarkoni, T. (2017). Progress toward openness, transparency, and reproducibility in cognitive neuroscience. Annals of the New York Academy of Sciences, 1396, 5–18. doi: 10.1111/nyas.13325.

• Works with docs and sheets
• Unlimited, automatic track changes
• Stored in cloud
• No special naming conventions
• Writers can work in parallel

• NYAS Editor wanted Word doc with track changes!
• Not all types of files
• Not everyone likes/knows how to use Google docs

• Pros
  • Keep the same file name
  • Let Box or OneDrive do automatic versioning
  • Shared file space
  • Any kind of file supported

• Cons
  • limit on # of versions (was 100 for Box)
  • Penn State enterprise license

• Integrates with Box, Dropbox, GitHub
• Pros
  • Free, open source, devoted to open science
  • Many different file types
• Cons
  • Version control depends on storage sources

• git
  • free, open source version control system
  • created by Linus Torvalds, creator of the Linux operating system, to manage that project’s software development

• GitHub
• GitLab
• Bitbucket

• Designed by and for software developers
• Text files
• Wonky
• Longevity of web-based repos?
• No automatic synching
• Users have to specifically push (upload), pull (download) files, merge/manage conflicts

• Users have to specifically push (upload), pull (download) files, merge/manage conflicts
• Great for R code, data
• Great for Jupyter notebooks (later in course)
• Exploits the power of the web

• Easy-to-use “Pages” feature
• Supports Markdown, Jupyter notebooks

• Everyone works on their own local copies
• Pull before they start work
• Save and push when they’re finished

• Use RStudio projects (next)
• Have your local directory structure mirror GitHub
• Pull before you start working

• Work in a branch (I usually work in dev)
• Commit often
• Push to GitHub
• Create pull request to merge your changes into master

• Complete Hello World exercise on GitHub

• Install: http://happygitwithr.com/install-git.html
• Tell git who you are: http://happygitwithr.com/hello-git.html

• Instructions: http://happygitwithr.com/push-pull-github.html
• (if necessary; might not need to do this if you installed recently) Avoid authenticating each time by caching credentials or using ssh keys

• Separate “home” directory for each piece of work
• Switch to using Open Project... command in File menu
• Saved metadata in *.Rproj (not saved on GitHub)

• Clone the Hello World repo, following these instructions: http://happygitwithr.com/rstudio-git-github.html#clone-the-new-github-repository-to-your-computer-via-rstudio

• Instructions: http://happygitwithr.com/new-github-first.html#new-rstudio-project-via-git-clone
• GitHub repo first
• New RStudio project by cloning repo

• Create repo on GitHub
• Copy link
• Open RStudio
• Create new project in RStudio

• Add *.Rproj to .gitignore
• Edit README.md
• Save and commit as ‘first commit’
• Push to GitHub to test connection

• Add sensitive files, directories to .gitignore

• Use git locally, but not push to GitHub
• Use RStudio projects, but not use git or GitHub
• Use private repository (limit of 3 collaborators/repo)

• packrat
• Trying it again for this course. Progress report soon.

• Simulation as a tool for reproducible and transparent science
• Visualization tools in R

This talk was produced on 2020-02-27 in RStudio using R Markdown. The code and materials used to generate the slides may be found at https://github.com/psu-psychology/psy-525-reproducible-research-2020. Information about the R Session that produced the code is as follows:

## R version 3.6.2 (2019-12-12)
## Platform: x86_64-apple-darwin15.6.0 (64-bit)
## Running under: macOS Mojave 10.14.6
## 
## Matrix products: default
## BLAS:   /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
## LAPACK: /Library/Frameworks/R.framework/Versions/3.6/Resources/lib/libRlapack.dylib
## 
## locale:
## [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
## 
## attached base packages:
## [1] stats     graphics  grDevices utils     datasets 
## [6] methods   base     
## 
## other attached packages:
##  [1] DiagrammeR_1.0.5 forcats_0.4.0    stringr_1.4.0   
##  [4] dplyr_0.8.3      purrr_0.3.3      readr_1.3.1     
##  [7] tidyr_1.0.0      tibble_2.1.3     ggplot2_3.2.1   
## [10] tidyverse_1.3.0  seriation_1.2-8 
## 
## loaded via a namespace (and not attached):
##  [1] viridis_0.5.1      httr_1.4.1        
##  [3] tufte_0.5          jsonlite_1.6      
##  [5] viridisLite_0.3.0  foreach_1.4.8     
##  [7] modelr_0.1.5       gtools_3.8.1      
##  [9] assertthat_0.2.1   highr_0.8         
## [11] cellranger_1.1.0   yaml_2.2.0        
## [13] pillar_1.4.3       backports_1.1.5   
## [15] lattice_0.20-38    glue_1.3.1        
## [17] digest_0.6.23      RColorBrewer_1.1-2
## [19] rvest_0.3.5        colorspace_1.4-1  
## [21] htmltools_0.4.0    pkgconfig_2.0.3   
## [23] broom_0.5.3        haven_2.2.0       
## [25] scales_1.1.0       gdata_2.18.0      
## [27] farver_2.0.3       generics_0.0.2    
## [29] withr_2.1.2        lazyeval_0.2.2    
## [31] cli_2.0.1          magrittr_1.5      
## [33] crayon_1.3.4       readxl_1.3.1      
## [35] evaluate_0.14      fs_1.3.1          
## [37] fansi_0.4.1        nlme_3.1-142      
## [39] MASS_7.3-51.5      gplots_3.0.1.2    
## [41] xml2_1.2.2         tools_3.6.2       
## [43] registry_0.5-1     hms_0.5.3         
## [45] lifecycle_0.1.0    munsell_0.5.0     
## [47] reprex_0.3.0       cluster_2.1.0     
## [49] packrat_0.5.0      compiler_3.6.2    
## [51] caTools_1.18.0     rlang_0.4.4       
## [53] grid_3.6.2         iterators_1.0.12  
## [55] rstudioapi_0.10    visNetwork_2.0.9  
## [57] htmlwidgets_1.5.1  labeling_0.3      
## [59] bitops_1.0-6       rmarkdown_2.1     
## [61] gtable_0.3.0       codetools_0.2-16  
## [63] DBI_1.1.0          TSP_1.1-8         
## [65] R6_2.4.1           gridExtra_2.3     
## [67] lubridate_1.7.4    knitr_1.27        
## [69] pwr_1.2-2          utf8_1.1.4        
## [71] KernSmooth_2.23-16 dendextend_1.13.3 
## [73] stringi_1.4.5      Rcpp_1.0.3        
## [75] vctrs_0.2.2        gclus_1.3.2       
## [77] dbplyr_1.4.2       tidyselect_1.0.0  
## [79] xfun_0.12