Interconnectivity

The “Scattered Data” problem

Developers constantly struggle with discovering, sharing and managing data coming from different systems in different formats. This requires understanding, structuring, integrating and verifying the data each time new features or applications are built based on that data.

In Web2, discoverability is enabled by search engines, which return a list of web links based on the query you search for. Sharing and managing data is governed by centralized services and protocols which do not share common data structures and interfaces, making it complicated to access and use this data.

Linked Data saves the day

The core idea behind linked data is to actually represent all things with relationships between them in a common graph. Linked data is built on primitives called "triples", which connect a subject entity with an object entity via a relationship. Having such a "semantic network" of data, we inherently add context and enable easy extensions. The semantic graph can be easily queried in many ways and enables growing a body of knowledge around things, rather than keeping "tables of strings".

Connecting Data the Google Way

Branimir Rakic, OriginTrail co-founder, maintains that the mechanisms within the OriginTrail protocol today closely resemble the way Google utilizes hyperlinks between web pages and manages to understand their data contents, as both technologies harness the power of their respective connection-first data structures, also known as knowledge graphs. However, important differences in the nature of the supply chain IT landscape and the World Wide Web require a more tailored approach in building the global supply chain knowledge graph than the one Google has taken, specifically related to data governance, decentralization, and employed standards. OriginTrail makes it possible to map virtually any data model, yet the core development team has been prioritizing the most relevant standards: GS1 EPCIS & CBV and those of W3C (Web of Things, Verifiable Credentials, PROV, etc.).