Concept

Principles

• Advanced graph visualization
• Human brain associative thinking
• Proficient User eXperience

Graph

(Network / Multidimensional Array / Database)

The memory of the system is a graph.

There are only two main features to store content in Graphiy KMS.

Item and it’s relations are links.

This makes SIMPLICITY.

Item

(Element / Unit / Node)

Every piece of information is an Item just like our thoughts or memory. It may be anything: just a word, number, date, url, or a bigger chunk of text, separate file or even binary data: image, video.

This makes UNIVERSALITY.

Item can be assigned with own value. A user just puts it in by editing. It is the same like human perceptual piece of memory - a pattern obtained with perception we possess (visual, auditory). Computer will perfectly store the value unlike rough human memory.

Item may have no own value and be a complex derivative formed by linking to others.

Link

(Relation / Connection / Edge)

Item to item relationship.

Works like human brain association. It has neither direction nor metadata. It is just a notion of item to item relatedness.

Two items relatedness extent is defined by link weight. Same as human brain associations have weight storing the “how much” two things are related.

This makes POWER.

Itemtype

(Metadata / Predicate / Context)

Here comes the tricky part. Contemporary graph databases store Items along with metadata. Same goes for links. They all have a type. Making typed links between items is like putting verbs into a sentence: “John” -likes-> “Boston”. And having Item typed is the way to define the context of the specific information. Here “John” is a “Person” type and “Boston” is a “City” type. That’s definitely an abstraction of some kind. Human thoughts and memories don’t have types as well as associations.

Type is all about what this Item is linked to.

For details, see Implementation