May 13, 2020

For background on the game Workers and Resources: Soviet Republic (WR:SR) (my new all time favourite) please check out the following links. Some background is assumed on the part of the reader for this blog post, as it was originally a reddit post. Because it is a reddit post, the members of that community role play as comrades ;).

• Steam Description
• Game Trailer
• Game Review (Rock, Paper, Shotgun)

 

Post title:

Administrating planning has a cost: information processing as a new (?) mechanic towards “Red Plenty”

or

Administrating Planning has a cost: gameplay reflection & a new (?) mechanic towards OGAS & Cybersyn

 

Proposal: a new resource to process and workers to manage – INFORMATION and the BUREAUCRAT! We already have plenty of energy and matter 😉

 

TL;DR: musings on how  “information” might be implemented as a cost and resource to be processed in WR:SR as a byproduct of your complexifying state economy, in the early game by paper pushing bureaucrats and in the late game by your very own mainframe computers and a Soviet Internet mega project!

 

Introduction

Greetings comrades, finally getting into this game and loving it. This is a game which perfectly captures the benefits and issues of economic planning. On the one hand, you, the player as the state – Gosplan and Gossnab etc, have the ability to plan an economic engine with rational thought – no internal markets or private enterprise allowed. On the other end, the theme and execution of the game (very meta) highlights the big issues with planning. This is all in the context of WR:SR being, of course, a simplified version of an economy.

WR:SR is an excellent systems-thinking game and is an interesting treatment of how economic planning works. I also love the narrative of failed roads to nowhere, or inefficient coal mines, and blaming it on calculation errors of the state whenever I make a mistake!

One thing demonstrated by the Soviet Union is that one planner cannot be all knowing! Or at least it takes a great deal of time and growing expertise in order to make a well functioning system. I’ve found it essential to be able to research tips and tricks online in order to improve my planning skills. This research could be done by a planner, and imperfectly executed as I do, but for example takes place (imho) more efficiently through a market based system where distributed planners (ie entrepreneurs) partake in trial and error, figure out local optima of say… designing a steel mill factory system. There’s a lot of economic theory about this (Hayek etc) that I’d love to discuss…This is collective intelligence in action. It is simulated (ie automated) more in capitalist tycoon sims like Cities: Skylines and Transport Fever 2. The planning is automated by simulated firms. You just zone and provide transport (basically). 

However, in contrast, the intricacies of central planning are made meticulously explicit in WR:SR, but it is missing one thing. I believe the cost of information must be simulated to make this game reach levels of perfection it is so close to.

Planning has a cost!

As you may begin to guess my current area of research involves cybernetic economic planning. This is basically what this game simulates. It is a systems thinking game of a planned economy. Obvious right? But what it doesn’t take into account, as far as I can tell in my measly 20 hours of playtime, is the administrative burden of all this planning. 

WR:SR simply outsources all (most, I guess) planning and “economic computation” to the giant Gosplan or Gossnab in the sky – the player! But I want a domestic planning industry represented in-game! Information costs and processing! The biggest benefit of implementing this would create the need to create administrative centres (large cities that are mostly bureaucrats beyond universities), communication networks, and a big end-game information processing system project with contributing and supporting industries.

To illustrate the need, I will provide an excerpt from Benjamin Peters’ great book from 2016: How not to network a nation: the uneasy history of the Soviet internet:

“All in all, the coordination problem [of the Soviet Economy] was simple to state but bewildering to solve: how could the nation best manage, harmonize, and organize all the information variables, planned and otherwise, that were flowing through its economy? How, if at all, could the Soviet knowledge base – including economic cyberneticists, a group known for a taste for circular problems – hope to account for the deficiencies of accounting in the system? 

In 1962, the State Committee for Automation and the Institute of statistics estimated that roughly 3 million citizens (about 1.3 percent of the 220 million total) were engaged in public accountancy, data registration, statistical and planning calculations, and other supporting information services for the planned economy and that the number was rising fast. And yet no one, outside of strong-armed national commanders under extreme wartime conditions, could manage and execute all the operations necessary to sustain the administrative creep of bureaucrats that was necessary to oversee the businesses, factories, and industries that were driving a national economy. 

In 1962, Viktor Glushkov, the prominent cyberneticists and architect of the OGAS Project, formulated the problem that his network project proposed a cybernetic solution for: he estimated that if the current paper driven methods continued unchanged, the planning bureaucracy would grow by by almost fortyfold by 1980, requiring the entire adult population of the Soviet Union to be employed in managing its own bureaucracy.”

Adding to this, from Phillips & Rozworski’s 2019 book The People’s Republic of Walmart and old Soviet Joke:

“Mathematicians have calculated that in order to draft an accurate and fully integrated plan for the material supply just for the Ukraine for one year requires the labor of the entire world’s population for 10 million years.” 

Gameplay Mechanic Suggestion

That’s the preamble to my gameplay suggestion: implement a planning bureaucracy with the eventual end game that is mostly replaced by computers for your SSR to survive into the 21st century! Basically, in 1960 start – at least 2% of your pop (and rapidly growing, or the equivalent manpower/cost) should be dedicated to planning alone (processing the information of your economy), and this should only grow as your economy complexifies. If you don’t have that pop, you should be paying to import the calculations from the USSR’s Gosplan, comrade!

Initially in the game the planning could be outsourced for Roubles to the Soviet Union, but you could eventually set up your own bureaucracy to manage planning just as you set up any other resource in the game. Different administrative state planning offices for different separate supply chains could be set up, with a system of head vs satellite offices that depend on a range to give oversight to the actual industries. These offices would be required for efficient functioning, receiving directives and so on. These offices, when staffed, represent your information processing capabilities for economic planning.

Your state could start with a general pioneering office, and then it could become more diverse and specialized as information processing needs increase. This would eventually incentivize the construction of a large capital city that focuses mainly on secondary, but primarily tertiary and quaternary industries – mainly planning administrative, political, education, and advanced manufacturing. The workers would essentially be planning calculators, processing the implications of your “on high” decisions. It would be best placed near the center of the map, incentivizing growth there, as you would want access to the whole country easily!

As your economy gets more complex, there is a greater strain on the information processing (computational resources, whether brains and paper (less efficient) or computers (grades of efficiency, depending on how they are networked)). So you will need a larger chunk of your educated population dedicated to managing the complexity of your economy over time (as per the quote above). The 2% figure will grow quite quickly and dynamically with your industrial and economics developments, you see. Eventually, it will become utterly unsustainable… but we will get to that in a bit. First, a summary of the gameplay mechanic I am proposing.

Gameplay Suggestion Summary

• New Cost/currency: information
  • Basically, there is an accumulating “information/directive cost” to every decision you, the player and high level planner, makes based on the complexity of your economy. 
    • This is both accumulated discreetly – in terms of placing buildings; and continuously- in terms of the operations of your economy.
  • Everything previously automated has an information processing cost – plans, supplies, schedules, deliveries and rations don’t make themselves, comrade!
  • Factory efficiency depends on information flow of communications of directives ie telecommunication networks
  • Expertise is implied and simulated here – ie in workers building complex buildings like refineries have an additional information cost compared to flats
  • Example of how it might work: unprocessed information is the value that accumulates as your economy complexifies. The goal is to keep the value of unprocessed information as low as possible, or else there might be delivery lags and other penalties in a tiered penalty system – like Happiness in Civilization: Beyond Earth (TBD).
    •  Ie, with simply an oil refinery and your base village, your information processed requirement might increase at a rate of 10gb per day (for instance figure) on top of the information cost of placing and planning all the infrastructure.
    • If your rate of information processing planning bureaucracy is less than 10gb/day (ie, you don’t employ enough bureaucrats or have a sophisticated enough office set up, for instance) , you will accumulate a paperwork / economic calculation backlog that will have to be dealt with.
    • Once you have an economy with oil processing, chemical industry, agriculture, and clothing manufacturing, your planning information processing needs might increase per day by some multiple of complexity of your operations. 
    • The goal is to match the processing capabilities as close to the rate of increase as possible, with a surplus to minimize backlog. Certain tiers of backlog might cause certain inefficiency or delay creep on your planning directives – or some other repercussion.
    • This incentivizes developing a well functioning bureaucracy appropriate to the size of your state quickly, so that the costs of processing this information is not borne in Roubles (or in dollars, if you get NATO’s IBM to tabulate it ;)).

• New Workers and Resources:  As the Soviet system did: bureaucracy is required to allocate resources! Use computational (cognitive resources) of a bureaucracy as a requirement to create and process information. These would be the knowledge workers of a planned economy. In terms of workers: first these would be educated people, then specialized educated people, then computer mainframes. What would the benefits be, beside reducing the information overload deficit of a planned economy?
  • Make the buses and trains coordinate and run on schedule
  • Automate: Order resources from abroad – there is a cost to this beyond deliveries!
  • Coordinate deliveries
  • Managing internal resource distribution (flows that you direct create an information cost that must be processed) and set internal “prices” (all automated, as it already is)
  • Collecting and computing Statistics (somewhat implemented already)
  • Essentially be a new cost/resource for anything that was priorly automated – automation comes at a cost, comrade, even if it is in the processing of information!
  • Build better buildings in a better network for processing the information

• Telecommunication networks
  • Administrative offices are linked to the work sites, factories, and towns via a new piece of infrastructure: telephone lines with telex
  • Necessary for maintaining coordination and keeping efficiency up, distributing your orders
  • If we wanted to complicate the 1960 early game even more, this communication network could be done by mail. It could be implied that it is done in person too, before you set up telephone lines (this will make sense later)
  • Improved communication systems improve the flow of information, and provide a basis of infrastructure for the end game.

• Supporting the Apparatchiks
  • New chains for Luxury goods and service production chains (movies, operas, consumer products) to keep the bustling centre city going!

 

Towards a parallel, meta end game

Now here’s where it gets interesting. Eventually, as with the humorous quote above , your information processing needs will become completely unsustainable by the mid-1980s, leading to collapse. How will you keep the socialist revolutionary dream alive! Well look no further.

Are you aware of the incredible socialist/ communist plans to automate entire economic planning bureaucracies with computer mainframes, advanced mathematics and telecommunication networks!? Check out Cybersyn and OGAS if not. The TL;DR of these is that they were attempts to create a kind of internet and automated economic planning computer system in Chile and the Soviet Union in the 1970s respectively. As illustrated above, this was seen by leading thinkers at the time as the only way of maintaining a command economy as it complexifies – or even making any type of economy perform better. Check out Francis Spufford’s amazing book Red Plenty if you want to learn more and have a great read.

While the early to mid game would be about building up your planning bureaucracy to gain independence from GOSPLAN and reduce costs, the mid to end game would be computerizing and networking the whole bureaucracy you have created which has now become too intensive to be managed by people alone. The goal would be to do this entirely with your own resources, including building the main frame and other industrial computers (these might be implemented in existing admin buildings like slots for vehicles in a distribution centre ie modular), as well as researching the technology (for instance, the advanced operations research mathematics and economic cybernetics), educating the workers, building resources that satisfy a different type of worker class (opera houses, limousines, etc). Don’t forget to build your very own cybersyn command room! Or, if you prefer a more command driven style, an OGAS Gosplan centre. Taking a Chilean vs Soviet approach will have different costs and benefits (TBD).

 

Conclusion

What exactly would be the gameplay implications of fully automated space communism?

• Information flows and processing would be simulated as a cost and resource in game
  • Being the gamer – the planner in the sky- is no longer completely free, and anything that was previously automated comes at a cost of info processing (bureaucracy)
• Creates a new meta-chain to work with
• Creates more reasons to have giant capital cities and educated workers
• Creates more reasons to have communication networks
• Directives are no longer instantaneous, there is a lag time leading to more interesting game play choices (for instance, about efficiencies of factories)
• Could perhaps be a resource that leads to automation or coordination of other flows, like bus schedules ( I mean spacing out busses and trains for more consistent frequencies… I haven’t figured this out in my own play through, so forgive me if this has already been figured out)
• If you implemented an AI in the game, perhaps it would provide optimization recommendations, comrade, or allow you to plop down optimal prebuilt set-ups to rapidly industrialize the rest of the map. I understand that this is already planned, so maybe this is a complementary feature?
• Further, there could be a political element that directs five year plans and makes all production in a chain cheaper and more efficient for the 5 years. There you go with the communist party headquarters or soviet council.

Drawbacks

• What is the first stage of communication networks – is it mail? Does this slow down the whole game by giving a lag to your orders?
  • Could this even be playable on easy difficulties? Or is it fully hard mode?
• Would this just complicate the game too much? I say no, the more complex the better 😀
• How exactly would this rationalize train schedules, deliveries etc?
• Are small lag times of command really a fun gameplay mechanic?
• That all said, this could be all turned off or modified in the set-up…

 

Thanks Comrades, I just wanted to bring my area of interest to the group and see how you could build on it. I think that this feature would really tie together the whole theme of the game, and would be really fun, especially for an end game paralleling the existing end game. I think this would be a most interesting way of implementing administration rather than just having a token capital building. This will only get more interesting as more resources and supply chains are added to the game. As I play the game more, I will think more about how this could be implemented. But as I said, I would love to hear your thoughts. Obviously I haven’t worked out the exact details of how this would be implemented. I just wanted to get the idea down as a sketch.

 

Onward!