Dev Diary #54 - Intelligence 📼
What's happening / TLDR: Developer diaries introduce details of Espiocracy - Cold War strategy game in which you play as an intelligence agency. You can catch up with the most important dev diary (The Vision) and find out more on Steam page.
---
A game featuring true espionage has to feature true information - existing, propagating, interacting in the game world. How can we achieve this?
In a topical and historical parallel (many of which permeate development of Espiocracy), Claude Shannon offerred an answer in 1948:
Here, he brilliantly focused on the objective flow of information, detaching information from the meaning or the form. Whether you are sending Shakespeare or insults, whether you are telegraphing them or sending them on Discord, this information can be always quantified into bits (nowadays usually represented as 0s and 1s) and then approached from the universal perspective of encoding trasmitters, noise-introducing channels, and decoding receivers.
The game implements information and therefore intelligence in similar way:
Let's explore this system, with plethora of examples from the perspective of Cuba in the game.
[h2]Information in the Game World[/h2]
At the most basic level, many entities in the game have secret states (such as a terminal disease) and execute secret actions (such as preparing a terrorist attack). These, when significant enough, become information in the intelligence mechanics, which is then conveyed (encoded) in various materials.
A few cases for a good start:
As examples above suggest, materials have different forms, which directly affect how materials can be collected (collection is professional term for acquisition of intelligence).
Physical materials are intuitive: documents, photos, recordings, devices, weapons, fingerprints, presence on the ground, and so on. They can be observed and inspected, some of them also copied, or just stolen.
Mental materials rely on human memory about information or about other materials. They have much higher noise and are forgotten over time but they can reach anywhere, including most critical information. Collection usually relies on conversations, briefings, and interrogations.
Ephemeral materials require immediate collection - eavesdropping on a conversation, intercepting chatter in military communication, observing an action, and so on. People who were a part of such event are also secondary collectors, usually by remembering the event.
Two first categories remain in the world (physically or in minds) and can further propagate through documents, conversations, phone calls, and even rumors (commonly overheard by operatives from intelligence stations). What an intelligence agency does with rumors?
[h2]Deriving Intelligence[/h2]
Once materials are collected, the player is aware of the kind and to what extent a material may potentially contribute. Potentially - because it is extracted through analysis that depends on:
In especially pressing cases, a player may cooperate with more advanced player to extract more information, at the unusual cost of potentially revealing source of the material and details of own operational methods.
Once extracted, information may still remain at potential stage - usually, single material provides only fractional intelligence. To arrive at actual intelligence, operatives usually have to acquire and analyze more materials. For instance, Cuban intelligence agency may first get a hint of incoming guerrilla invasion through rumors overheard by a station in Washington. Then, operatives in investigative way can explore this lead further by wiretapping potentially relevant actors, recruiting spies among Cuban dissidents, conducting risky overflights over places with potential training camps, and so on. This allows the player to further define the details of the invasion, such as date and place, that allow military to successfully and quickly repel it.
[h2]Intelligence[/h2]
Every intelligence agency around the world has its own definition of "intelligence", often complicated by local language. Espiocracy avoids this futile task and instead implements Wittgensteinian approach: intelligence is defined by examples, context, and actual use.
Derived intelligence is used either at national or at operational level. National intelligence contributes to country's international position, available actions, and sometimes even survival. It ranges anywhere from ordinary industrial espionage (eg. trade secrets) all the way to grave revelations (eg. real nuclear position of an adversary).
Operational intelligence, does not interest the government but it is very useful for the player, directly expanding number of available actions. Primarily, it clusters around vulnerabilities and secrets which can be found near all actors in the world.
Derived and even used intelligence remains in the world and has life after life:
[h2]Behind The Scenes[/h2]
â–º UX for intelligence is not satisfactory yet. Quick sneak peek at work in progress:
â–º What about false intelligence? Good question! There are a few ongoing experiments, credibility assigned to materials or ability to manufacture false materials (eg. spreading false rumors, sending a walk-in to adversary's embassy with forged documents, producing misleading military chatter...) but they require special care with more iterations to extract the best possible gameplay.
â–º While it may seem unusual (or even worrying) that a game with espionage in name still substantially changes espionage mechanics after 3 years of development, this is how innovation is made. For a telling case, see "Shadows of Doubt" - a game nominated in Steam Awards for most innovative gameplay in 2023, which was in development since 2015 and made major pivot in core gameplay in 2018. Returning to Espiocracy, the best parts of current gameplay were more or less not possible to be invented three years ago - instead, they required three years of implementing, playtesting, and iterating. You can trace trace this journey in dev diaries, with "Strategic Materials" (DD#11) in 2021 as the first solid stab at discrete intelligence materials, then "Secrets and Opportunities" (DD#23) in 2022 as a implementation of more materials that grew to contribute a lot of fun to gameplay, then "Espionage Gameplay" (DD#47) in 2023 as a wider attempt at unifying and expading these, and now we are here in 2024 with true information.
â–º Speaking of iterations, I'll drop a screenshot of current mechanics around spies without any explanation:
[h2]Final Remarks[/h2]
The next dev diary will be posted on July 5th!
---
If you're not already wishlisting Espiocracy, consider doing it
https://store.steampowered.com/app/1670650/Espiocracy/
There is also a small community around Espiocracy:
---
"I think that the public reaction, as I judge it, has not been one of shock or horror; it has been much more along the lines of - the intelligence services carry out intelligence work, good" - David Cameron about Edward Snowden
---
A game featuring true espionage has to feature true information - existing, propagating, interacting in the game world. How can we achieve this?
In a topical and historical parallel (many of which permeate development of Espiocracy), Claude Shannon offerred an answer in 1948:
Here, he brilliantly focused on the objective flow of information, detaching information from the meaning or the form. Whether you are sending Shakespeare or insults, whether you are telegraphing them or sending them on Discord, this information can be always quantified into bits (nowadays usually represented as 0s and 1s) and then approached from the universal perspective of encoding trasmitters, noise-introducing channels, and decoding receivers.
The game implements information and therefore intelligence in similar way:
Let's explore this system, with plethora of examples from the perspective of Cuba in the game.
[h2]Information in the Game World[/h2]
At the most basic level, many entities in the game have secret states (such as a terminal disease) and execute secret actions (such as preparing a terrorist attack). These, when significant enough, become information in the intelligence mechanics, which is then conveyed (encoded) in various materials.
A few cases for a good start:
- When Castro decides to deploy troops in Angola, this information is materialized in the game through military plans, military communication, and change of behavior of military forces in Cuba
- Construction of nuclear power plant in Juragua is visible from the ground, air, and space, and therefore can be materialized by witnesses or devices with cameras
- After arrival of Mig-21 jets in Cuba, every deployment and flight conveys this information (until it is well known that Cuba procured Mig-21s)
As examples above suggest, materials have different forms, which directly affect how materials can be collected (collection is professional term for acquisition of intelligence).
Physical materials are intuitive: documents, photos, recordings, devices, weapons, fingerprints, presence on the ground, and so on. They can be observed and inspected, some of them also copied, or just stolen.
Mental materials rely on human memory about information or about other materials. They have much higher noise and are forgotten over time but they can reach anywhere, including most critical information. Collection usually relies on conversations, briefings, and interrogations.
Ephemeral materials require immediate collection - eavesdropping on a conversation, intercepting chatter in military communication, observing an action, and so on. People who were a part of such event are also secondary collectors, usually by remembering the event.
Two first categories remain in the world (physically or in minds) and can further propagate through documents, conversations, phone calls, and even rumors (commonly overheard by operatives from intelligence stations). What an intelligence agency does with rumors?
[h2]Deriving Intelligence[/h2]
Once materials are collected, the player is aware of the kind and to what extent a material may potentially contribute. Potentially - because it is extracted through analysis that depends on:
- Parameters of the material (such as signal-to-noise ratio, eg. very high in a single photo vs very low in thousands of hours of phone call recordings),
- Skills of analysts from responsible section,
- Access to specific facilities (eg. DNA forensics),
- Agency-wide levels of specialization (in politics, military, digital devices, and so on)
In especially pressing cases, a player may cooperate with more advanced player to extract more information, at the unusual cost of potentially revealing source of the material and details of own operational methods.
Once extracted, information may still remain at potential stage - usually, single material provides only fractional intelligence. To arrive at actual intelligence, operatives usually have to acquire and analyze more materials. For instance, Cuban intelligence agency may first get a hint of incoming guerrilla invasion through rumors overheard by a station in Washington. Then, operatives in investigative way can explore this lead further by wiretapping potentially relevant actors, recruiting spies among Cuban dissidents, conducting risky overflights over places with potential training camps, and so on. This allows the player to further define the details of the invasion, such as date and place, that allow military to successfully and quickly repel it.
[h2]Intelligence[/h2]
Every intelligence agency around the world has its own definition of "intelligence", often complicated by local language. Espiocracy avoids this futile task and instead implements Wittgensteinian approach: intelligence is defined by examples, context, and actual use.
Derived intelligence is used either at national or at operational level. National intelligence contributes to country's international position, available actions, and sometimes even survival. It ranges anywhere from ordinary industrial espionage (eg. trade secrets) all the way to grave revelations (eg. real nuclear position of an adversary).
Operational intelligence, does not interest the government but it is very useful for the player, directly expanding number of available actions. Primarily, it clusters around vulnerabilities and secrets which can be found near all actors in the world.
Derived and even used intelligence remains in the world and has life after life:
- Many pieces of existing intelligence with enough tradecraft may lead to new intelligence in the process of inference,
- Operatives remember intelligence that they derived - which then can be extracted from them in interrogation... or they may reveal it after defection,
- Actors remember intelligence that was distributed to them - so when we steal nuclear blueprints and pass to inventor-actor who has a butler-spy, retelling or even copies of said blueprints may reach the agency that handles that spy!
- Intelligence may be sold to other players (again, with inherent risk of revealing own methods, much lower than in materials but still present),
- Leaked to press, revealed after years in a book,
- And of course it will be also uncovered to the player in the game over screen
[h2]Behind The Scenes[/h2]
â–º UX for intelligence is not satisfactory yet. Quick sneak peek at work in progress:
â–º What about false intelligence? Good question! There are a few ongoing experiments, credibility assigned to materials or ability to manufacture false materials (eg. spreading false rumors, sending a walk-in to adversary's embassy with forged documents, producing misleading military chatter...) but they require special care with more iterations to extract the best possible gameplay.
â–º While it may seem unusual (or even worrying) that a game with espionage in name still substantially changes espionage mechanics after 3 years of development, this is how innovation is made. For a telling case, see "Shadows of Doubt" - a game nominated in Steam Awards for most innovative gameplay in 2023, which was in development since 2015 and made major pivot in core gameplay in 2018. Returning to Espiocracy, the best parts of current gameplay were more or less not possible to be invented three years ago - instead, they required three years of implementing, playtesting, and iterating. You can trace trace this journey in dev diaries, with "Strategic Materials" (DD#11) in 2021 as the first solid stab at discrete intelligence materials, then "Secrets and Opportunities" (DD#23) in 2022 as a implementation of more materials that grew to contribute a lot of fun to gameplay, then "Espionage Gameplay" (DD#47) in 2023 as a wider attempt at unifying and expading these, and now we are here in 2024 with true information.
â–º Speaking of iterations, I'll drop a screenshot of current mechanics around spies without any explanation:
[h2]Final Remarks[/h2]
The next dev diary will be posted on July 5th!
---
If you're not already wishlisting Espiocracy, consider doing it
https://store.steampowered.com/app/1670650/Espiocracy/
There is also a small community around Espiocracy:
---
"I think that the public reaction, as I judge it, has not been one of shock or horror; it has been much more along the lines of - the intelligence services carry out intelligence work, good" - David Cameron about Edward Snowden