Entities

The Organizations, Communities, Individuals, Activities, Artifacts, and Repositories that make-up this project.

Trying to understand Data in this project requires a fundamental understanding of Data Structures and Dimensionality, but since you may not want to jump straight into Learning Overview or Concepts we'll just summarize Entities as follows:

Entities and Connections are the two foundational building blocks of the Data.

We go into a bit more detail about Entities in the following two sections:

• Types of Entities - variations in what we use to ground Connections

• Filterable Attributes - attributes that we use with visualizations

• Non-Filterable Attributes - attributes that help us learn more

Types of Entities

When reviewing the existing work, we realized that there were a few different types of building blocks that characterized the national security ecosystem as it pertains to innovation.

Below you will find each of them listed and defined with an example or two for context.

Organization

An Organization is a formally established body of individuals with documented responsibilities.

We recognize that this definition may be obvious, but want to set everyone up with a common understanding of our terms.

To simplify the framework across different Categories we did not choose any Entities that might be used more specificially in a given context (e.g. Staff to indicate what type of Entity a Government Organization is, or Startup to indicate a more specific Commercial Entity).

We make up for this choice by using Tags, which are more modular and allow for simple Entity resolution with more nuanced contextual references (e.g. you can filter the Graph by all of the Organizations with Staff or Startup as one of the Tags).

Note: One of the beneficial aspects about starting with Data and being very specific about what we mean by different terms is that we can more easily make changes and updates based on Feedback. We will continue to evaluate this Type over time.

Example: The Defense Innovation Unit is a Government office, Stanford University is an Academic institution, Saildrone is a Commercial company, and the RAND Corporation is a Non-Profit. Each is an Organization distinguished by Categories.

Community

Communities are similar to Organizations, but more informal. They are a collection of Individuals.

This type is valuable because some meaningful contributors to innovation in national security do so within such informal networks and relationships. Sometimes these evolve into more formal structures or Organizations, sometimes they originate in such Organizations or emerge from Activities - regardless, they will typically be more amorphous and less defined, yet worth characterizing.

Additionally, a project might be understood in some ways as a Community that probably includes Activities and produces some sort of Artifact.

Example: The Federal Innovators Salon is gathering of government employees focused on innovation, but not a formal Non-Profit Organization. It also holds Events, which may or may not generate Artifacts.

Individual

An Individual is a person involved in an Organization, Community, or Activity that is deemed important as a Data element and either has publicly accessible information available or consents to be included.

Social network analysis (SNA) is a very useful tool in understanding multiple facets of a movement or organization (in fact, Kumu has a dedicated feature for just this purpose). While the Graph will support baseline network analysis across organizations, some of which is representative of underlying social networks, we know from our personal experience that the web of personal ties is much different than the lines draw by organizaitonal charts.

We see a couple use cases for this Entity type, both of which have pros and cons:

• Public individuals, such as government leaders or those in key organizational positions, whose tenure and activities can help enrich an understanding of why certain activities did or did not take place. The down-side of this is potentially giving too much weight to said individuals.

• Private individuals, typically those of lower power or who do not have a public-facing role, whose actions and connections often drive much of the ground-up drive behind innovative efforts. A potential con of this is not being able to represent enough of such individuals to accurately capture a network due to privacy concerns.

NOTE: While we recognize that Individuals are important to understanding certain aspects of innovation in national security, we want to be thoughtful about how they are integrated (for example, public vs. private figures and contact information).

Example: Hondo Guerts is an Individual whose tenure as the Assistant Secretary of the Navy for Research, Development, and Acquisition included the establishment of NavalX, a Government Organization, may benefit those studying innovation in national security.

Activity

An Activity is executed by an Organization, Community, or Individual to achieve a goal or end state.

There are many different activities tied to national security as well as innovation. Some are linear processes, some are cycles, others are more like projects or events, but each must be initiated to achieve an outcome and comprise actions, not just individuals (which is what differentiates an Activity from a Community, since they are otherwise very similar).

Example: The Defense Entrepreneurs Forum annual conference is an Activity geared toward the Organization mission of inspiring, connecting, and empowering people in the national security community.

Artifact

An Artifact is produced as part of an Activity by an Organization, Community, or Individual and contributes to an understanding of innovation in national security.

Artifacts are the sibling Entities to Activities and allow for a more nuanced understanding of an Organization, Community, or Individual in the form of physical or digital remnants.

Example: General Charles Brown is an Individual whose tenure as Chief of Staff of the Air Force and associated Artifact generation in the form of memos to "Accelerate Change or Lose" may benefit those studying innovation in national security. Research is also useful as Artifacts.

Repository

A Repository is an Artifact collection tied to a national security-relevant Organization, Community, Individual, or Activity, especially one where the content is non-public.

Repositories store information, some of which may be in the form of Artifacts, though most will be so large or non-public that specific Artifacts are unlikely to be linked. Repositories are typically a data set or database, though the exact structure may be non-public or unknown.

Example: AFWERX is a Government Organization that contracted with Mobilize, a Commercial Organization, to create Vision, a non-public web platform that includes a Repository of US Air Force innovation projects and activities, among other data holdings and features.

NOTE: There are likely to be Repositories whose existence or details would be contrary to this project's goal of supporting national security, so we want to approach any such inclusions holistically (for example, the inclusion of links or other potentially sensitive information that is not publicly available).

Filterable Attributes

Now that we have an understanding of what Entities are, we need to dig into a bit more what characterizes them - for this, we have a few different features or attributes, the most importance of which we use as filters to modify different visualizations in the Graph, Timeline, and Map.

Categories

A Category is essentially an attribute with a singular value used to help organize or filter different Entities.

More is covered on these in the dedicated ensuing Categories section, but for the purpose of this project, an Entity can only have one Category. Because this gets into serious questions about Dimensionality and even has potential implications for Ethics, we dive deeper into it in the next section.

The important thing to understand is that, like Organization and Tags, this attribute is malleable to meet the needs of the project. While we are defaulting to a more robust set of Tags and fewer Categories, this is largely driven by limitations of Kumu, which only allows one set of multi-select fields.

The option to sub-divide Categories exists (e.g. making Government - Staff an option instead of just Government), but we want to give some time to work with users before diving too far into the weeds, since we know whatever the project starts with will inevitably change (and the Categories page will be the one-stop shop for any such changes to be documented).

Example: The Defense Innovation Unit cannot be both a Government Entity and a Commercial Entity - it must be one or the other, in this case Government.

Tags

Entities can have multiple Tags, which allow for greater understanding of their different dimensions by creating multiple facets through which to view them.

Tags are both powerful and confusing, which are really two sides of the same coin. To understand them further, we highly recommend learning about Dimensionality, but here's a quick synopsis.

Because Entities can have multiple Tags, they allow for flexible viewing, reducing complexity by filtering out Entities that do not meet the relevant criteria when such criteria can easily apply to multiple Entities and is not a defining characteristic in the same way Categories are set-up. This makes them powerful.

They are confusing, though, because of this same feature: who picks the Tags and what do they mean? Should Tags be defined by the nomenclature of a given Organization or their dictionary defition? This gets confusing quickly, so read on in Tags for more.

Example: The Defense Innovation Unit has Defense, Military, Innovation, and Unit tags in order to help visualize it alongside other Entities with the same tags.

Range

The Range is a time period or series of years where an Entity existed, used most with the Timeline.

While this feature will likely change over time, we believe there is value in seeing not only the current state of innovation in national security, but also its historical evolution. For this reason we need to capture Data for the Timeline and this field is where we do so.

NOTE: This formatting is very specific to Kumu, which is our initial visualization tool, but as we move through various stages of the process the associated Data fields may change to accomodate other tools or approaches.

Example: The Defense Innovation Unit has a Range of 2015..2022 because it was established in 2015 and continues to this day.

Non-Filterable Attributes

In addition to the attributes listed above, which allow us to reduce the size or complexity of our Data by filtering, there are other attributes which are essentially flat and do not further complicate the Dimensionality of an Entity (though they do still add dimensions).

What this functionally means is that some of these attributes might change over time or only apply to certain Entities based on whatever helps us better understand the Data.

Because these are simple and pretty intuitive, we do not go into much more detail. As much as possible, we try to reduce complexity and be cognizant of Dimensionality by not creating unique fields for Organizations in comparison to Groups or Individuals.

All Entities

Any entity type can have the following non-filterable attributes:

• Website

• LinkedIn

• Twitter

Read on to learn more about Categories and Tags!