User Guide

A gentle introduction to Wikibase

Wikibase is open-source software designed as an extension to MediaWiki, and it offers a collaborative space to collect, edit and curate information about things in a structured way, make links between those things and create a knowledge base. Wikibase is particularly good at opening up access to data, not only for computers but for human audience in their own language, and in a manner that the data can be shared with the rest of the world as part of the Linked Open Data (LOD) web. Originally developed for and still is part of the Wikidata project, Wikibase is used by a wide range of linked open data projects and in many areas of knowledge management, including academic research, life sciences, computer science and cultural heritage.

Key concepts

Wikibase does not provide a specific data model, and rather, it is up to users to 'model the world' according to their needs. However, it does have a base data structure, consisting of two core types of entities, items and properties, and items are described using statements which are in item-property-value format (like the subject-predicate-object triples in RDF).

Item

Represents a concept or thing that is the subject of a Wikibase page, or put another way, every Wikibase page is an item.

Each item has a label and description, and further details are given as statements.

Statement

Describes an item using property and value pairs.

Property

Describes a relationship between an item and value of the property.

Each value of a property can be further described or refined using qualifiers (which are themselves properties).

For example, the item Maya Angelou in Wikidata includes the following two statements:

• Maya Angelou (item) is an instance of (property) human (value)
• Maya Angelou (item) was spouse (property) of Paul du Feu (value) and this was from 1974 (qualifier) and until 1983 (qualifier)

In the second statement, the value of the spouse property is an item describing Paul du Feu, and this creates a link between the two items (i.e. Maya Angelou and Paul de Feu) and describes the relationship between the items (i.e. spouse).

Data structure for the Congruence Engine Data Register

The main purpose of the Congruence Engine Data Register is to record and describe datasets that exist elsewhere, and to facilitate comprehensive search, discovery and linking of the datasets. (See Project Documentation for the rationale)

In the Congruence Engine Data Register, an item is a dataset, and each dataset can have a number of statements to describe its core characteristics, such as subjects and topics of a dataset, where the dataset can be accessed, the organisation/person holding the dataset, copyright status, license status, etc. The statements use a small set of properties, a full list of which is available on the All Properties page.

The following table describes the data structure and guidelines for the Congruence Engine Data Register.

Property	Property ID	Data type	Requirement	Description	Guidelines
Label	-	String	Required	Name of the dataset.	Make sure that abbreviations are expanded and that the name does not include versions or creation dates or any project-specific details that might be difficult to understand/decipher for general users.
Description	-	String	Required	Description of the dataset i.e. what the dataset is of.	Be as descriptive as you can, including (where available) which region or historical period the dataset covers, what type of data it includes e.g. text, images, videos, etc.
Alias	-	String	Optional	Other names by which the dataset is known.
instance of	P1	Item	Required	Always Dataset (Q1).
keywords	P3	String	Preferred	Words or phrases that relate to or describes the dataset.	Add each keyword as individual/separate value.
date created	P4	EDTF Date/Time1	Optional	Date on which the dataset was created/published/released.	The date can be entered as full date (YYYY-MM-DD), year and month (YYYY-MM), year only (YYYY) or as a range e.g. YYYY/YYYY or YYYY-MM/YYYY.
held by	P5	String	Preferred	Dataset holder's name, qualified with described at URL property.	For the value of described at URL property, try finding/creating a Wikidata item first, and fallback to a equivalent webpage elsewhere e.g. about page on the dataset holder's website.
available at URL	P6	URL	Preferred	URL at or from which the dataset can be accessed or downloaded.
copyright status	P7	String	Optional	Copyright status e.g. public domain, copyrighted, qualified with described at URL property.	For the value of described at URL property, find a Wikidata item that defines the status e.g. https://www.wikidata.org/wiki/Q19652
copyright license	P8	String	Optional	License terms e.g. CC0, qualified with described at URL property.	For the value of described at URL property, find a Wikidata item that defines the license e.g. https://www.wikidata.org/wiki/Q6938433
Digital and Cultural Heritage Datasheet URL	P9	URL	Preferred	URL at or from which the datasheet can be accessed.
used by	P10	String	Preferred	Name of project(s) using or known to have used the dataset, qualified with described at URL property.	Add each project as individual/separate value. For the value of described at URL property, use the URL where further information about the project(s) can be found.

¹ What is EDTF Date/Time? (video)

Contribute to the Congruence Engine Data Register

Creating an account

In order to add and edit dataset records to the Congruence Engine Data Register, you will need to be a registered user. Please create a user account if you don't have one already.

Registering a dataset ​​

1. Using the Search, check and make sure your dataset is not already registered in the Data Register.

2. Log in to Congruence Engine Data Register.

3. Select New Item from the Wikibase menu in the left side bar, and you will be directed to the 'Create a New Item' page.

4. In the ‘Create a New Item’ page, enter the basic information for the item you wish to create as follows:

Create a New Item in Wikibase - Example

Language

By default, a language will be automatically selected based on your computer’s language settings. You can leave it as is or manually select another language that is appropriate to your entry.

Label

This is the name of your dataset. Please make sure that abbreviations included in the name are expanded (e.g. West Yorkshire Queer Stories Bradford oral histories ‘ instead of WYQS Bradford oral histories) and that the name does not include versions or creation dates or any project-specific details that might be difficult to understand/decipher for general users.

Description

This is the description about your dataset i.e. what the dataset is of. Please be as descriptive as you can, including (where available) which region or historical period the dataset covers - basically, the details that you think would help others discover your dataset via search.

Aliases (optional)

These are other names your dataset may be known by. If there are multiple names, enter as pipe-separated string e.g. WYQS|Queer Stories

5. Once you have entered the basic information, click the ‘Create’ button to register your dataset. You will then be redirected to the page where you can add further details about your dataset, known as statements.

Wikibase Statements - Example

Adding statements

Statements are used for recording data about an item, consisting of property and value pairs. In the context of the Congruence Engine Data Register, a statement would be information about subjects and topics of a dataset, where the dataset can be accessed, the organisation/person holding the dataset, copyright status, etc.

1. Click the ‘+ add statement’ link.

2. Start typing into the property input field and select a property you wish to add from the suggested list of properties.

3. Enter a value for the property and click ‘save’, or continue below to add qualifiers.

4. If you need to add more values to a property, such as with keywords, click the ‘+ add value’ button.

Adding qualifiers

Some statements in the Congruence Engine Data Register will require you to add ​​qualifiers. Qualifiers, added as property and value pairs, are used to further describe or refine the value of a property in a statement. For instance, what is meant by 'West Yorkshire Archives' can be elaborated on by using the property 'described at URL' as a qualifier to link to an entry in Wikidata.org or other web page(s).

Wherever possible, we would like further descriptions be given using the described at URL property with a link to a relevant entry in Wikidata.org. This typically applies to the held by (dataset holder), used by, copyright status and copyright license properties.

1. If you are not editing already, click the ‘edit’ button for the statement you wish to add a qualifier to.

2. Click the ‘+ add qualifier’ button.

3. Add property and qualifier value, and click ‘save’.

Guidance: Please make sure to check for an entry that describes the property you are qualifying at Wikidata.org, and add the link to the entry as a value. Where there is no pre-existing Wikidata entry, please create a new Wikidata entry, or alternatively, find an equivalent webpage elsewhere e.g. about page on the dataset holder's websit e.

Search the Congruence Engine Data Register

Datasets in the Congruence Engine Data Register can be searched in a number of ways:

• Search which is powered by CirrusSearch (enhanced MediaWiki Search). For advanced search, see the help pages for MediaWiki Search and WikibaseCirrusSearch.
• Query Service and SPARQL (a type of query language). We have some examples to get you started.
• Data Register App which provides an alternative search tool.