Geographic Diversity in Public Code Contributions — Replication Package
Bibliographic record
Abstract
Geographic Diversity in Public Code Contributions - Replication Package This document describes how to replicate the findings of the paper: Davide Rossi and Stefano Zacchiroli, 2022, Geographic Diversity in Public Code Contributions - An Exploratory Large-Scale Study Over 50 Years. In 19th International Conference on Mining Software Repositories (MSR ’22), May 23-24, Pittsburgh, PA, USA. ACM, New York, NY, USA, 5 pages. https://doi.org/10.1145/3524842.3528471 This document comes with the software needed to mine and analyze the data presented in the paper. Prerequisites These instructions assume the use of the bash shell, the Python programming language, the PosgreSQL DBMS (version 11 or later), the zstd compression utility and various usual *nix shell utilities (cat, pv, …), all of which are available for multiple architectures and OSs. It is advisable to create a Python virtual environment and install the following PyPI packages: click==8.0.4 cycler==0.11.0 fonttools==4.31.2 kiwisolver==1.4.0 matplotlib==3.5.1 numpy==1.22.3 packaging==21.3 pandas==1.4.1 patsy==0.5.2 Pillow==9.0.1 pyparsing==3.0.7 python-dateutil==2.8.2 pytz==2022.1 scipy==1.8.0 six==1.16.0 statsmodels==0.13.2 Initial data swh-replica, a PostgreSQL database containing a copy of Software Heritage data. The schema for the database is available at https://forge.softwareheritage.org/source/swh-storage/browse/master/swh/storage/sql/. We retrieved these data from Software Heritage, in collaboration with the archive operators, taking an archive snapshot as of 2021-07-07. We cannot make these data available in full as part of the replication package due to both its volume and the presence in it of personal information such as user email addresses. However, equivalent data (stripped of email addresses) can be obtained from the Software Heritage archive dataset, as documented in the article: Antoine Pietri, Diomidis Spinellis, Stefano Zacchiroli, The Software Heritage Graph Dataset: Public software development under one roof. In proceedings of MSR 2019: The 16th International Conference on Mining Software Repositories, May 2019, Montreal, Canada. Pages 138-142, IEEE 2019. http://dx.doi.org/10.1109/MSR.2019.00030. Once retrieved, the data can be loaded in PostgreSQL to populate swh-replica. names.tab - forenames and surnames per country with their frequency zones.acc.tab - countries/territories, timezones, population and world zones c_c.tab - ccTDL entities - world zones matches Data preparation Export data from the swh-replica database to create commits.csv.zst and authors.csv.zst sh> ./export.sh Run the authors cleanup script to create authors--clean.csv.zst sh> ./cleanup.sh authors.csv.zst Filter out implausible names and create authors--plausible.csv.zst sh> pv authors--clean.csv.zst | unzstd | ./filter_names.py 2> authors--plausible.csv.log | zstdmt > authors--plausible.csv.zst Zone detection by email Run the email detection script to create author-country-by-email.tab.zst sh> pv authors--plausible.csv.zst | zstdcat | ./guess_country_by_email.py -f 3 2> author-country-by-email.csv.log | zstdmt > author-country-by-email.tab.zst Database creation and initial data ingestion Create the PostgreSQL DB sh> createdb zones-commit Notice that from now on when prepending the psql> prompt we assume the execution of psql on the zones-commit database. Import data into PostgreSQL DB sh> ./import_data.sh Zone detection by name Extract commits data from the DB and create commits.tab, that is used as input for the zone detection script sh> psql -f extract_commits.sql zones-commit Run the world zone detection script to create commit_zones.tab.zst sh> pv commits.tab | ./assign_world_zone.py -a -n names.tab -p zones.acc.tab -x -w 8 | zstdmt > commit_zones.tab.zst Use ./assign_world_zone.py --help if you are interested in changing the script parameters. Ingest zones assignment data into the DB psql> \copy commit_zone from program 'zstdcat commit_zones.tab.zst | cut -f1,6 | grep -Ev ''\s$''' Extraction and graphs Run the script to execute the queries to extract the data to plot from the DB. This creates commit_zones_7120.tab, author_zones_7120_t5.tab, commit_zones_7120.grid and author_zones_7120_t5.grid. Edit extract_data.sql if you whish to modify extraction parameters (start/end year, sampling, …). sh> ./extract_data.sh Run the script to create the graphs from all the previously extracted tabfiles. sh> ./create_stackedbar_chart.py -w 20 -s 1971 -f commit_zones_7120.grid -f author_zones_7120_t5.grid -o chart.pdf
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.009 | 0.069 |
| Meta-epidemiology (narrow) | 0.001 | 0.002 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.005 | 0.005 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.004 | 0.006 |
| Open science | 0.003 | 0.006 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.166 | 0.136 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".