Skip to content

Linking data

Linking data is what turns 4☆ data into 5☆ linked data. This guide will help users create their first linked data where your data reuses standard definitions.

A transcribed audio screencast covering the contents of this page


This page assumes that you have a well structured CSV with a corresponding JSON qube-config.json as built in describing your CSV.

Column Mappings

We saw in the describing your CSV section that we can provide a qube-config.json configuration file to describe the data set. We can also use this file to describe columns.

    "$schema": "",
    "title": "Sweden at Eurovision",
    "summary": "List of Swedish entries to the Eurovision Song Contest since 1958.",
    "description": "Sweden has been competing in Eurovision since 1958, with an enviable track record of wins. This dataset covers all contests since 1958, their artists, the song names, language (if mono-lingual), and some observations covering points in final, rank in final, and number of artists on stage. Data originally sourced from and",
    "license": "",
    "publisher": "",
    "dataset_issued": "2022-04-08",
    "keywords": [
        "Song Contest",
        "European Broadcasting Union"
    "columns": {
        "Column Name": { ... column definitions ...}

Note the newly added columns property which takes a key-value mapping for each CSV column that you wish to describe.

For example, in the sweden_at_eurovision_no_missing.csv data set we have been working with up to now, we could specify that the Year column should make use of the year template:

"columns": {
    "Year": {
            "from_template": "year"

Column Templates

Column templates are pre-configured column definitions which speed up creating linked data. csvcubed has several templates for columns converting calendar and UK Government time periods, as well as observation-status.

For more information see the templates guide.

Next steps

The next step is to inspect your CSV-W.