The CNRL for short. The Computational Network Reference Layer design and concept is inspired by librarians. For whole countries and beyond it is possible to walk into a library and find a book. But we live in a digital networked age where we need to connect to the Internet and be able to access the right information wherever it resides. This is only possible, if like a library we know where all the knowledge is and how it is described.
Describing data, information or knowledge is an on going pursuit. Ontology to conceptual model thinking are applied but if we all apply different thinking using these tools then society creates disjointed knowledge. Some peers strive to create closed or patented knowledge while others even with open standards will reach few participants. There is also many levels on which community standards need set. We have set out already the Ontology/conceptual model level, but we also need to sink deeper to the technical level where content address designs are / have been put forward IPLD or SOLID to name two.
It feels like a much deeper and bigger goal is required to align all those creating and describing knowledge. The summary identifiers of knowledge should actually contain the knowledge. For example a hash taken of a Wikipedia web page e.g. swimming page would resolve to #hash of strings and numbers using cryptography. This would give us a start and the Peer to Peer data protocols use this technique to produce their technical content addresses. But we want more. We want a descriptor of the knowledge to be the best science can describe nature. An example helps explain again. Take the chemical bonding structure of H20 i.e. water. As you can see if you click on the link to the wikipedia page, on the right hand side many identifiers exist for this one molecule. The CAS number is the first listed. What does this number represent? It is just allocated by the American Chemical Association. Further down the list we have an InChI number where they use cryptography to produce a hash from a file format that describes the bonding structure of the H2O molecule. The later seems a far better description of nature and given the use of cryptography gives prosperity that help universally to describe knowledge on a network. I.e. as far as we know H20 will be the same all across not just Earth but the whole universe. Of course, chemistry is a lot more complicated and complex, even just for H20, we have heavy water for example and then we have very long complex molecules to describe and these keep on being added to too, even a change in the laws of physics may require an upgrade to the bonding structure files. These changes can be addressed if we think of these descriptors as a Data Type Smart Contract (usually secured on a blockchain). The programmability of such contracts give us flexibility to evolve while still guaranteeing backward compatibility.
Flexibility can also be applied via democracy over the standard setting progress. DAO decentralised autonomous organisations are one way to apply liquid democracy or voting on standards. Now, each standard setting body especially in open science will assert this stance too. However, the practical reality is that a small groups of peers drive forward standards for debate and resolution. This may well continue but if set via a DAO then the any one can participate at any time and history shows bringing in a diversity of ideas often is when new conceptual or practical break thoughts occur.
The CNRL is at the heart of the DIY HealthScience Toolkit and is a demonstration of using data type smart contracts and DAO voting to describe data. Much has been learnt and more software needs to written to complete the build out of the concept. It is an open question whether this new approach just adds to the pile of attempts to describe knowledge or with a take up of practical daily use, starts to take off.