Universal Knowledge Base
Another consequence of the “semantic problem” is that it is very difficult to build reusable, universal digital knowledge bases. In order for information to be reusable and universal it needs to be semantic, it needs to have meaning for computers.
We can give digital information very precise meaning if we store it in structured databases. However, we will reduce the type of information the knowledge base can handle dramatically. Structured databases and its client-applications can generally only handle information of specific, predefined types.
Traditional knowledge bases, CMS, Wikis, etc. store information as plain text and then link and cross-reference topics, tag articles, create word indices and allow full-text search. However, computers still do not understand the meaning of the information, certainly not the way a human can. Even if they use advanced OWL ontologies on top of document stores or graph databases, they are still just describing the content, giving hints, the content itself is not self-descriptive to a computer.
A universal knowledge base (UKB) is a semantic, application-independent, universal vessel that can absorb arbitrary information in random order. It can handle information irrespective of type and complexity and it does not need to be prepared in any way for specific information types. Related information merge automatically. It is capable of creating dynamic structures from chaotic data.
Most importantly, it is a semantic database. It understands – to a certain degree at least – the information that is stored in it, it can handle and process information depending on information type. It is an intelligent memory that can recollect information semantically; a search can be semantically fuzzy and still produce meaningful results.
A UKB is a very powerful tool. Users can store arbitrary information application-independently much as they could write that information inside a textbook. But unlike using a textbook, in a universal knowledge base the order in which information is stored is irrelevant. It behaves more like a human brain that can create coherent pictures from seemingly random information bits.
A UKB can be a universal database for applications. Applications can query and alter data similarly as they can when using traditional databases - with the difference that the data they use is not proprietary but universally shared between all applications. But a UKB can be much more than just a database for applications, a UKB can become the application. As it is universal, it can store information and the instructions needed to process them. Information (data) and functionality (application) are stored using the same language. All it needs is an environment that executes the processing instructions when required.
A universal repository for Havel is such a universal knowledge base.
Havel, or rather the semanto-relational data model it is based on, has the fantastic property of chaotic information storage. A Havel repository does not need any structure whatsoever and any new information you add into a Havel repository automatically merges with existing information. Information do not just merge within the repository itself but also with any other information expressed in Havel, be it with a single expression a user just created or with a massive information hub in the cloud.
Universal repositories allow us to rethink the basic concepts of how we handle digital information. They allow information-centric computing and are the base for semantic social networks. Imagine if we one day have a semantic version of a general knowledge base (like Wikipedia) joined with a semantic functionality sharing platform (like an app-store but for functionality associated with specific information types). Anything we author can semantically merge with a significant part of the entire knowledge available to mankind. We can automatically join, extend, verify and interpret our information and ideas using the united wisdom of humanity. We would have instant access to any functionality and tools available for our tasks, regardless of features installed on our devices.
This post is part of a comprehensive introduction into SRDM, graph languages and Havel.
Continue reading: Universal Computer Language
Comments