Manifesto

Values & IdealsAll Data ConnectedStandardsData MarketsOpen World

1. Much higher levels of transparency, openness, sustainability, fairness and accountability are going to be required at all levels in any organization or ecosystem.
2. Human Capital needs to be known, valued, leveraged and optimized.
3. Data Capital needs to be known, valued, leveraged and optimized.
4. Increasing competitiveness depends more and more on having the highest quality and depth of data, information and knowledge.
5. Having one "censored", biased, Single version of the Truth (SVOT) is no longer good enough for many---if not most---use cases in most domains.
6. The world becomes more and more polarised due to "information bubbles" that many people are not escaping from, more depth, context and connectedness is needed.
7. The world becomes more and more complex and harder to understand, a holistic view around every given topic, showing all viewpoints, would provide people---and AI's---with deeper understanding enabling them to make better decisions.
8. Reusability of deliverables, components, artefacts, data, information, knowledge and logic needs to effectively be the highest priority since it positively affects all of the above.

1. All data that can be connected, will be connected.
   • Information, Knowledge, Meaning, Logic: it's all data.
   • Knowledge & Meaning will be captured as machine-readable executable models.
   • Data is considered explained when its usage has no misconceptions or ambiguities.
   • All your connected data is an EKG.
2. All data will be made available anywhere—secured and within entitlement limits—at any time to any device, node or edge.
   • We embrace the Open World and deal with the realities of Multiple versions of the Truth (MVOT).
     • PS: Open World does not mean that all data is open to everyone
   • Leveraging all processing power and storage capacity that is available at the edge
   • Reducing energy consumption
3. We combine the digital footprint of activities along with a digital representation of information and knowledge, from which an EKG emerges.
4. An EKG is connections of Knowledge Graphs across an Enterprise, an Ecosystem or beyond.
   See principle 3: Distributed.
5. We encourage the use and widespread proliferation of EKG identifiers (see principle 1: Identity)

1. EKGs are based on standards and therefore interoperable across boundaries.
2. There are many types of standards that an EKG needs to be able to deal with.
3. Standards are described as machine-readable models—i.e. ontologies—that EKG Platforms can execute, interpret or enforce.

See also principle 10: Standards.

1. Any data source will be turned into a data publisher---or supplier---of one or more s self-describing datasets.
   • Any data sink will be turned into a data consumer, consuming one or more datasets.
2. Data suppliers and consumers will find each other via a data market using a standard "lingua franca" for the data itself, its meaning, all its associated policies and metadata and especially also its use cases as executable models.
3. The data market manages the information supply chains between all the various suppliers and consumers.
4. The global data market will consist of many other more specific data markets e.g. per industry or per enterprise.
5. An EKG is the combination of one or more data markets and the deployment of its use cases.

1. For any given "Thing"---i.e. an object—there may be many representations in many datasets.
2. An object's representations may be different in shape, meaning, timeliness, relevance and quality---i.e. any given representation of information about a given object may represent a different version of the truth.
3. All representations of any given object shall be linked via shared identifiers.
   • Identifiers shall be meaningless, opaque, web-resolvable and universally unique.
     See principle 1.
   • An object can have multiple identifiers.
4. Any given object consists of 1 or more datapoints.
5. A datapoint represents a logical property of a given object.
   • The identifier for a datapoint is the identifier of the object it belongs to plus at least one identifier of the axiom that describes its meaning (which is also an object).
6. Datapoints---or datapoint-values for the same object---can exist in many datasets.
   • with potentially multiple versions of the truth in terms of meaning, timeliness, relevance and quality.
7. Any representation for any given datapoint of any data source shall be made available to any device, node or edge in the network within legal, policy and entitlement limits in real-time.
8. Every "object" or "thing" that is represented as data in whichever dataset anywhere, shall have an identifier that is universally unique, permanent, meaningless or opaque and therefore shareable, resolvable through the HTTP protocol.
   See principle 1.

See also principle 4: Open World.