The ODISSEI code library is a collection of code and scripts used to execute projects using the ODISSEI infrastructure. ODISSEI (Open Data Infrastructure for Social Science and Economic Innovations) is the national research infrastructure for the social sciences in the Netherlands. ODISSEI brings together researchers with the necessary data, expertise and resources to conduct ground-breaking research and embrace the computational turn in social enquiry. Through ODISSEI, researchers have access to large-scale, longitudinal data collections as well as innovative and diverse new forms of data. These can be linked to administrative data at Statistics Netherlands (CBS). Combining data from a wide range of sources enables researchers to answer new, exciting, interdisciplinary research questions and to investigate existing questions in novel, new ways.
The data_design column in _data/cbs.csv is populated automatically from the ODISSEI Knowledge Graph, matched via each entry's CBS_project_nr. This happens at build time — a script (enrich_cbs.py) runs as a step in the GitHub Actions workflow before Jekyll builds the site, querying the KG's production SPARQL endpoint for every valid project number found in the CSV and writing back one entry per distinct CBS data table used (formatted as ShortCode|DOI|DutchTitle, separated by ; for projects using multiple tables), which the site then renders as clickable, filterable table codes with the Dutch title as a hover tooltip. This means the column is fully KG-managed: you never need to fill it in by hand when adding a new entry to cbs.csv — as long as the row has a plain numeric CBS_project_nr, the next build (triggered by any push, or automatically every Monday) will populate it — and any manual edits to that column will be overwritten on the next build. If the KG is temporarily unreachable during a build, the step fails safely (continue-on-error) and the site simply deploys with whichever data_design values were already committed from the last successful run. Note: Claude (Sonnet 5) was used to implement this upgrade
Do you want to submit your own project and code to be added to the library? Please submit an issue using the Submission code issue template (or send me an email).
Edit the source files _data/cbs.csv, _data/liss.csv or _data/port.csv.
A few formatting rules to keep the site's search and filter dropdowns working correctly:
- For any field that can hold multiple values (e.g.
programming language,data used,platform,method), separate multiple entries with a semicolon (;), not a comma — e.g.Python; R. Commas are only for prose inside a single value (e.g. a dataset description). - For CBS entries,
CBS_project_nrshould be digits only (e.g.8674), with no prefixes or extra text. This is the key used to automatically look up that project's CBS data tables from the ODISSEI Knowledge Graph. - Do not edit the
data_designcolumn in_data/cbs.csvby hand — it's generated automatically at build time from the Knowledge Graph (see "CBS data design enrichment" above) and any manual edits will be overwritten on the next build.
Commit your changes to a branch of the GitHub repository.
Once the pull request is accepted, and the checks are all successful, the result will be visible on the ODISSEI code library within a few minutes.
If you would like to test locally, see Testing your GitHub Pages site locally with Jekyll. Once you have installed jekyll, you can simply run a local copy on your own machine by starting a server on the command line with:
$ bundle exec jekyll serve
If you've added new CBS entries and want to see their data_design column filled in locally rather than waiting for the next deploy, run python3 enrich_cbs.py (requires pip install requests) before starting Jekyll — it rewrites _data/cbs.csv in place from the live Knowledge Graph.