How the topic taxonomy works
The Topic Taxonomy is a classification scheme for organising and finding content on GOV.UK, based on its subject area.
Not to be confused with the Topics published through Collections Publisher.
Editing the Topic Taxonomy
The taxonomy is managed in content-tagger. Users must have the “GDS Editor” permission in content-tagger in order to see the relevant pages.
The topics in the taxonomy (we call them “taxons” in code) are persisted in the publishing-api as content items. For an example see the “Education” taxon.
This means that taxons inherit the publishing-api workflow, and can be in either draft state or published.
The link type
parent_taxons is used to store the relationship
between taxons. A reverse link called
child_taxons is setup through the publishing-api.
There is no technical limit in what can be tagged to the taxonomy, but not every type of content in the publishing-api is suitable for tagging to the taxonomy.
Content Tagger has a generic interface for tagging content to the taxonomy.
The relationship between a page and a taxon is persisted in the publishing-api “links hash”. For example, see the taxons link in the content item for this guidance document.
Accessing the taxonomy
The level one taxons are associated with the GOV.UK home page through
root_taxon link type. The GOV.UK home page in turn has a
corresponding reverse link of link type
level one taxons.
This is the content item for the GOV.UK home page with all level one taxons listed.
An example of a level one taxon is the “Education” taxon:
You can use this to find the structure of the taxonomy by following
Accessing tagged content
You can fetch content tagged to a particular taxon from the Search API (search-api).
This is used in some GOV.UK search pages. For example https://www.gov.uk/search/news-and-communications has a topic/subtopic facet that allows filtering. Advanced search pages like https://www.gov.uk/search/advanced?group=news_and_communications&topic=%2Feducation (which are accessed from topic pages) are prefiltered by topic.
The filter works with a
content_id rather than URL. To find all content
tagged to the above mentioned “Education taxon”:
By default search-api returns a handful of fields in a search result item. You are able to override the default fields by naming which fields you want returned. If a content item does not have one of the named fields provided, it will be left out of the returned item. See full documentation here.
You can filter on multiple different field names if you wish to narrow
down what you are searching for. You are able to filter by one or many
field names and multiple values can be given for each field name.
To find all
speech content tagged to “Education taxon”:
You can also access all content tagged to a taxon and the part of the taxonomy below it. The following will give you everything tagged to topics in the “Education” taxonomy:
You can see the number of documents in each topic by using
Editors can use Whitehall to tag content to the taxonomy.
Individual branches can be hidden from Editors by clearing the
visible_to_departmental_editors flag on level one taxons in
Additionally taxons have a ‘phase’, which can be 'alpha’, 'beta’ or 'live’. The phase of a taxon controls its visibility on the front end apps. This allows us to publish the entire taxonomy while making those parts which are not considered mature enough for production harder to find.
High level metrics regarding the taxonomy are recorded in Graphite, and can be looked at through a Grafana dashboard.