This section presents a TEI/XML-encoded excerpt of The Book of Tea by Okakura Kakuzō. The encoding highlights key structural and semantic components such as chapters, paragraphs, names of places and people, and significant thematic references. Following TEI P5 Guidelines, the markup emphasizes the cultural, philosophical, and aesthetic ideas associated with the Japanese tea ceremony, including Zen, simplicity, and transience. Line breaks and metadata-rich elements were also used to ensure both textual fidelity and interoperability with semantic tools.

This encoded text has been transformed with XSLT using FreeFormatter into an HTML code file in a .xsl file format. The html code has been turned into and actual functioning and consultable .html page containing the annotated text. This .xml file has been also turned into a RDF file in the .ttl format using a .py script. The rdf file has been validated using RDF Grapher which produced a RDF Graph (the image right below) and a .rdf file format that contains all the relationships throughout the elements inside the encoded text only and that are visualized in the graph. All the files related to the book - including the .pdf of the entire book - can be found in a dedicated folder in our GitHub repository.

The RDF Graph

Click on the image to view a zoomable version in full size.

×

Custom URIs

To further express relations developed in the conceptual model in a concrete way, people, activities, concepts, and places particularly with connections to the theme and items were assigned to custom URIs.
Our base URI defined is https://w3id.org/a-lod-of-tea

CSV Production

We produced CSV files containing the full descriptions of each item, using either the metadata standard adopted in our project or the one provided by the respective institution. Based on the theoretical and conceptual models we developed, we selected and incorporated relevant connections and relationships among the items. They are available to view here

Triple Structure Overview

We then mapped the item descriptions to formal ontology schemas used in our conceptual model, representing them as RDF triples with each item serving as the subject, as shown by clicking on each entry below. In addition to custom URIs created for items and related entities, broader real-world instances were linked using external resources such as Wikidata and GeoNames. These instances, along with non-item entities defined by custom URIs, were assigned an rdf:type and a human-readable label. The "non-items" section contains triples in which these entities serve as subjects, enriching the overall ontology.

Python Script for Transformation

The transformation into RDF was done using Python, into the Turtle serialization format. Below is the Python script of one of the items ("The Portrait of Rikyu"). The Python files of all the items can be found here.

RDF Dataset

This is full data set of the generated RDF, transformed into turtle format.

The individual .ttl files can be found here.

After we transformed all of the triples into a ttl file, we then created a visualization of the whole dataset, using RDF Grapher.

Click on the image to view a zoomable version in full size.

×