Over the years, the technology has matured and been refined to the extent that is has been adopted by many major players in the social media and solutions space. The original idea was to move beyond a web of unrelated page links, to a situation where there was meaning behind those links, and thus the Semantic Web movement was formed. This became more commonly known as Linked Data and was envisaged to allow data to be the linking framework of the internet, not just web page URLs.

Today, graphs are an increasingly common part of the internet landscape. Companies can now host their graph in Amazon Neptune and pay to browse LinkedIn's graph of their people network. Graphs are a ubiquitous, flexibe and natural way to model data compared to a traditional relational database.

Triki stores all your website metadata in a graph database. What is metadata? Metadata is data that describes data, so it triki's case it is data that describes your web pages. The majority of the actual content in your pages is normally stored in Markdown (or HTML) files. Triki blends together the content with the metadata to build every page. And where a page naturally links to another page, for example a Gig links to a Band, or a Book links to an Author, all these data level links are stored in the graph.

For our graph database, we use Apache Jena which has evolved to a best-of-breed database which can support many different database formats. Importantly, it fully support SPARQL which is the query language for linked data and allows arbitrary nodes to be found. Triki uses and makes it possible for users to leverage to return simple data subsets, for example get the me last three more recently added Blogs.