Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[GoblinCookie]

@GoblinCookie

Treasury Workshop, Trade Depot, Shop, Dispensary are part of the logistics in economy like caravans or shipping.

Stockpiling overproduction and unpredictable caravans and unpredictable shipping is what makes this economy feel realistic to players. This way there is a created supply for visiting adventures to purchase weapons, armor, food or goods. There is opening for player to send carpenters to cut wood instead or idle miners to smooth floors and walls in caverns, so less vermin is created there. There is no need for creating additional trade imbalance between sites. Economy should work without famines and catastrophes as well. Imbalance already is ensured by environmental economic inabilities of Elves, militarism of Goblins and Human greed. The guaranteed shortage is of adamantine or various metals, as well as some species dying out from over-hunting. I wish it applied to vermin too lolz. However justification of encrusting industry is still missing imho by badly designed Dwarven need of items acquirement.

Imho economy does not need to be applied in one release, as ready top-down design. More naturally economy is, when it is released patch by patch, one  functional element after another functional element in bottom-up design, which is easier for developer to balance and to add further economical elements.

Dispensaries or whatever has nothing to do with the issue, I am not talking about the internal economy of the fortress here, but the global economy in general. 

I am not sure we are even having the same discussion.  The issue here is not the economy not working, it is also the economy working too well.  We need some economic demands to go unmet in order for the player as a seller to fit in.  But we also need the player as a buyer to be reliably able to obtain the things he needs at the moment.  This is why the economy needs natural distasters to exist first, if we introduce an economy in a world with no disasters then the AI will in the centuries before our arrival solve all it's own problems, which means that every demand in the wider world has been met by an AI already.

Add distasters in however and we end with controllable shortages that can be solved by the AI *or* the player alike.  This allows the player to fit into the economy, without making the economy permanently broken. 

For DF to have an economy running world wide so to speak, each individual site needs to have its own internal supply and demand... as things are unless the player makes restrictions on themselves any fort they make is capable of self sufficiency.
If we have less variety and need to look at other sites for things like flux stone or food or even tools, this starts a supply and demand in game, if the dwarves in your fort work for their own benefit rather than the good of the fort this creates an internal supply and demand, which can leade to the need to trade with other sites for things your dwarves need,  eg a site near a river would have game , fish and wood, to sell, a mountain start would focus on mining and would need food and wood to make tools for its self, a site on more of a plain will focus on farm land, sites on sand would make glass to send out and so on... focused sites as i said before rather than self sufficient sites will allow an economy to work,

It is not about who the dwarves work for, that is irrelevant here. 

The problem is that geography in the world is pretty uniform.  Any plain is pretty much the same as every other plain, any mountain is pretty much the same as any other mountain.  The world is also rather small, with travel times being rather short.  Why would I trade for something from an exotic locale, when I could just send one of my own dwarves off-site on a camping trip to get it myself, since the exotic locale is only a few days travel away. 

[Anandar]

So like i said if each site had a very limited range of any and all resources we would have to reach out to other site for things we need to to get variety, if the uniformity of the map was lessened camping a day away would become less viable, creating economic trade through supply and demand rather than self sufficient locations... this could further other game elements like raiding - trade caravans via ambushes, diplomacy and trade negotiations, shops  etc, disasters might create demand but when the populace has nothing but starving dieing people what can they offer to an economy?

[GoblinCookie]

So like i said if each site had a very limited range of any and all resources we would have to reach out to other site for things we need to to get variety, if the uniformity of the map was lessened camping a day away would become less viable, creating economic trade through supply and demand rather than self sufficient locations... this could further other game elements like raiding - trade caravans via ambushes, diplomacy and trade negotiations, shops  etc, disasters might create demand but when the populace has nothing but starving dieing people what can they offer to an economy?

Or they could just send dwarves out beyond the boundaries *of* the site to get whatever the site needs?  Problem is that even if we make the resources available in a given place, if the travel times remain as low as is currently they are then there is no reason to trade for the goods from other sites when you can just walk to where they are available yourself.  There are two problems, firstly the biomes are too uniform and secondly the travel times are too short, you have to solve both the problems, not that they are really 'problems' from the POV of the actual inhabitants of the universe  , but from the POV of us having trade+caravans. 

Starving dying people do not offer anything to the economy, that is essentially the point.  We need something to break bits of the economy so that player *as* a trader can then 'fit in' but by making it disasters rather than a flaw in the AI economic logic itself we ensure that the AI can fix the problem itself if the player does not turn up to fix it themselves.  If there are no disasters over the centuries the economy will be 'completed' so that every demand that can afford to be met is met already by the AI economy, leaving no room for the player. 

The disasters make an opening.  That opening can take the form of an increase in demand (a site that needs to rebuild after an earthquake needs bricks) or it can take the form of a collapse in supply (the mine collapsed in the earthquake) or both.  In any case, it forces the economy to *do something*, which allows the player to take the role of what would otherwise be an AI trader in solving the problem. 

Starting scenarios allow the player fortress to be initially 'slotted in' to an existing niche, the economic situation dictates the available starting scenarios and starting scenarios are 'generalised to every site foundation'.  You could for instance be tasking with rebuilding the above mentioned mine, that would be your starting scenario and it would serve an economic function to the game as well.

[Sarmatian123]

Not every country has the same resources at their disposal. Taxation and border barriers happen on borders of countries. Combining it all with waste of limited resources, which are anti-economically mismanaged by current economies, it leads to international unhinged free trade. Free trade as fallacy of course leads to many more issues like shrinking economies, mass unemployment and other economic unpleasantness.  It can be helped as always from the time of old grandpa mercantilism with international trade balance checks, which are backed with man made trade barriers regulations. Complicated and unpleasantly administrative governmental work, but if you want all wheels of national economy to work at their top notch efficiency, you need to put lots of tender care into this management. Singapore example here shines and the death of its top manager I hope is not beginning of their national economic tragedy. Anyhow, Singapore can afford little mismanagement. Being the richest country of the world gives them lots of dough to smooth the decay unless they keep their management post staffed with people who know what they are doing. Not many countries can. Managers and economists are not educated in correct mercantile and mercantile-like procedures these days either. It all goes down into liberal ruin these days.

Back to topic. I think I know why we are having 2 discussions. You're talking about MULTIPLAYER DWARF FORTRESS. Maybe you're a developer yourself and not just a mere programmer. Multiplayer DF could be huge. Biggest thing yet even. Allowing players to raid players instead just Goblins and such. However I talk about pretty flattened perspective of single player game, which I do not want to burden with taking too much cpu taking care of artificially introduced into game instabilities. Therefore I am for AI to manage their embark economy on base of full settlement size and all embarks available resources, while you would like to limit this embark to some "enabled" production factors, so an inter-embark unhinged free trade is cause to exist. Players do what is fun for them, usually some stupid things, and they need caravans to sell their focus materials to buy what they failed to make. It speeds the embark process, like mass-immigrant waves. Now I would rather something fancy happened with immigration due economics, then with content of caravans. Also you missing the whole adventure experience. I rather my Dwarven adventurer visiting an Elven Retreat had something more to buy in a way of food, then uneatable nuts of a tree.

This is why our discussion goes apart. I talk about missing fundamental elements, which economy can be created in the game on bottom-up design even, like treasuries, depots, shops, dispensaries, while you are concerned with economic plans and how the economic plans imbalance of production can lead to "feel like real LIBERAL" economy in DF. Something typical of top-down design. For you, those elements I was suggesting about introducing, do not matter even.

Thank God we had in our history something more then liberal economies. Personally I favor industrious American Dream over British unhinged free trade of liberal economy. Even Great Britain became biggest economic power of world not on fantasies of Adam Smith, but on solid trade policies of British mercantilism like the Navigation Acts. However this is the part I didn't want to have even in this topic, when I started this thread on forum. Everyone has his/hers favorite economies and almost everyone can for hours, like me here for neomercantilism, argue for their economies. It is like PC, Mac and Linux discussions these days. Is it wise to have a quarrel on those? It supposed to be just suggestion of few improvements to the game. Not 1000 entries long quarrel about whose car shines best. Do trust me I had such prolonged argumentative battles with people from Mises Institute. It leads to nowhere, to an agreement that there is a disagreement.

[GoblinCookie]

@Sarmatian123
 
Quote when you respond to something, because context is important.  I don't know whether you are responding to myself, to Anander or to my response to Anander.

Back to topic. I think I know why we are having 2 discussions. You're talking about MULTIPLAYER DWARF FORTRESS. Maybe you're a developer yourself and not just a mere programmer. Multiplayer DF could be huge. Biggest thing yet even. Allowing players to raid players instead just Goblins and such. However I talk about pretty flattened perspective of single player game, which I do not want to burden with taking too much cpu taking care of artificially introduced into game instabilities. Therefore I am for AI to manage their embark economy on base of full settlement size and all embarks available resources, while you would like to limit this embark to some "enabled" production factors, so an inter-embark unhinged free trade is cause to exist. Players do what is fun for them, usually some stupid things, and they need caravans to sell their focus materials to buy what they failed to make. It speeds the embark process, like mass-immigrant waves. Now I would rather something fancy happened with immigration due economics, then with content of caravans. Also you missing the whole adventure experience. I rather my Dwarven adventurer visiting an Elven Retreat had something more to buy in a way of food, then uneatable nuts of a tree.

In effect the economy *is* multiplayer DF.  The idea of the economy is to have sites trade with eachother and with individual adventurers wandering out, plus nomadic groups one they are introduced.  This is to replace the current setup where goods are created magically in a pre-scripted fashion so the player in adventurer has something to buy, steal or help themselves to in either mode.  The current goods are conjured out of nowhere to meet player demand, the idea of the economy is to replace this situation with a situation that sites and individuals other than the player's produce and trade things. 

At the moment we can add whatever we like into *either* mode.  This means that proposals for new buildings or systems or money or wages or property or whatever are not the economy.  Anything internal to the fortress mode embark, or to the onloaded area of adventure mode can be added without in any way adding to the economy.  In fact, the more complex the situation is within the onloaded area, the harder it is to make the economy exist.

So yes, the outcome of successful introducing the economy can be described as making multiplayer DF a possibility, if the devs are that way inclined.

This is why our discussion goes apart. I talk about missing fundamental elements, which economy can be created in the game on bottom-up design even, like treasuries, depots, shops, dispensaries, while you are concerned with economic plans and how the economic plans imbalance of production can lead to "feel like real LIBERAL" economy in DF. Something typical of top-down design. For you, those elements I was suggesting about introducing, do not matter even.

You cannot build the economy in such a fashion, for a very simple reason.  What we have at the moment is an economy in fortress mode (the adventure mode economy is so rudimentary it can be ignored).  If you propose, as you are proposing adding more content for the existing fortress mode economy then this content does not automatically translate into the Economy when you add enough of it in. 

In order to make the Economy, the devs need to be able to fit the outcome of fortress mode economy (not the mechanics) INTO an abstracted model since the folks in the other sites don't 'actually exist' when we are not there.  The more content you add into fortress mode, the more complex it becomes to make an abstracted model that roughly corresponds to the outcomes of the fortress mode economy. 

If the outcomes do not correspond, then we end with a situation where retiring the fortress in itself causing total economic meltdown in the wider world. 

Thank God we had in our history something more then liberal economies. Personally I favor industrious American Dream over British unhinged free trade of liberal economy. Even Great Britain became biggest economic power of world not on fantasies of Adam Smith, but on solid trade policies of British mercantilism like the Navigation Acts. However this is the part I didn't want to have even in this topic, when I started this thread on forum. Everyone has his/hers favorite economies and almost everyone can for hours, like me here for neomercantilism, argue for their economies. It is like PC, Mac and Linux discussions these days. Is it wise to have a quarrel on those? It supposed to be just suggestion of few improvements to the game. Not 1000 entries long quarrel about whose car shines best. Do trust me I had such prolonged argumentative battles with people from Mises Institute. It leads to nowhere, to an agreement that there is a disagreement.

That is another thing that makes adding in the Economy hard.  If we have only economic and trade system for everyone, then you can produce an extremely abstracted model that responds to what the player does, it does not matter that if the trade+production going on is vastly abstracted since it will correctly depict the consequences of the player's actions on the economy.

Having multiple economy systems however means you have to model the whole economy in far more detail since you have to describe the differences explicitly so that the game can accurately produce the consequences of the co-existence of the two systems.  With only a single system you can simply model the consequences of the player's own actions on that system, which makes things a lot simpler and requires far less detail.

[Sarmatian123]

@Sarmatian123
 
Quote when you respond to something, because context is important.  I don't know whether you are responding to myself, to Anander or to my response to Anander.

It was just too much of copy paste in there.

So yes, the outcome of successful introducing the economy can be described as making multiplayer DF a possibility, if the devs are that way inclined.

Introducing economy will not solve programming issues of multiplayer introduction. Toady's Dwarf Fortress has no server-client architecture. However, I think I saw some indie DF-like game on YouTube with pleasant 3d graphics. Maybe it already has?

You cannot build the economy in such a fashion, for a very simple reason.  What we have at the moment is an economy in fortress mode (the adventure mode economy is so rudimentary it can be ignored).  If you propose, as you are proposing adding more content for the existing fortress mode economy then this content does not automatically translate into the Economy when you add enough of it in. 

In order to make the Economy, the devs need to be able to fit the outcome of fortress mode economy (not the mechanics) INTO an abstracted model since the folks in the other sites don't 'actually exist' when we are not there.  The more content you add into fortress mode, the more complex it becomes to make an abstracted model that roughly corresponds to the outcomes of the fortress mode economy. 

If the outcomes do not correspond, then we end with a situation where retiring the fortress in itself causing total economic meltdown in the wider world. 

Economy down the basics is just flow of goods, no matter how they are generated.
Financial flow is ideally just reversed flow of goods, no matter what financial controls are applied.
Those are economy type independent elements.

Abstracted economy model is generating goods out of thin air upon demand. Right?
Abstracted economy model does not generate stockpile of goods at an embark for the game to play around with it, when trade is involved. Goods appear from thin air on demand. Right?
Retiring fortress creates huge issues with tracking huge quantity of items. Right?
Mechanics of abstracted economy applied on retired fortress is creating strange anomalies, which can be described simply as bugs. Players tend to build fortresses in peculiar ways, which pushes abstracted economy model into spinning out. Right?

To add to the pile of programming issues with abstracted economy model, which is missing out entirely on any logistic system. I suggested functional, generalized and simplified logistic system's solutions, which will survive banishment of Quantum Stockpiling and maybe could even improve FPS. Solutions like warehouses and shops, which will fit any possible chosen economy system, even as dumb as planned real communism, but will not fit an abstracted economy model where goods drop from thin air.

So, how we can marry abstracted economy model with the simplest elements of logistics like Trade Depots, Trading Caravans and numerous stockpiles without causing mayhem at fortress retirement? Particularly, if this part of DF still after years of development is causing unexpected weird bugs. Answer is pretty simple, a new abstracted economy model!

That means my suggestions, if to be implemented before big economy release, need at least some conceptual work on flow of great river of goods. Not to mention the economic system flavors (tribal, feudal, corporate, liberal) engaged with creating goods. As I can see it, the devs have a clear path here.

Unified logistics (could exist without rewrite of abstracted economy model and actually Trade Depots and caravans already do) ->
Economic systems (need rewrite of abstracted economy model for goods generation and needs to factor in unified logistics like at least Trade Depots and caravans) ->
Specialized Logistics (ship-caravans, player owned ships and player owned caravans in fortress mode and player owned wagons and ships in adventure mode)

One unexpected result of a mix-market economy model there, could be mixed race fortresses, which can produce not only large cloths, but also foreign items given expertise arrived with a guest looking for citizenship. However I doubt we will see it rather in DF in its current form. It could be for devs a viable possibility though to program, if they choose so.

Problem with large quantity of items is that this thing leads to FPS death in DF. This is why abstract model is favored over logistic+abstract model.

That is another thing that makes adding in the Economy hard.  If we have only economic and trade system for everyone, then you can produce an extremely abstracted model that responds to what the player does, it does not matter that if the trade+production going on is vastly abstracted since it will correctly depict the consequences of the player's actions on the economy.

Problem with economy type in DF belongs rather to racial ethics, then to known to us economy systems. Specialized workshop types available only to given races for example. So in DF universe our chat liberal vs anti-liberal economy, tribal vs feudal, have really no meaning. So basically there can be one abstracted economy model, just with limitations in production for given races or immoral allowances for another race and that's basically it.

Having multiple economy systems however means you have to model the whole economy in far more detail since you have to describe the differences explicitly so that the game can accurately produce the consequences of the co-existence of the two systems.  With only a single system you can simply model the consequences of the player's own actions on that system, which makes things a lot simpler and requires far less detail.

You are over-complicating pretty simple things. Abstracted economy models like in real life need unified interface to coexists with each other.  The international trade, you know? For this part you already have caravans and Trading Depots. To redesign those logistics, is not the same with redesigning abstract economy models. We are not talking here different economic systems, but the unified logistic interface between them. Something that can bring more fun and flavor for players already playing DF and maybe something to attract new ones. Even before entirely new economic systems redesign is created. However as currently existing abstract economy model is creating things from thin air on demand, it maybe is playing loose ball with already generated items during fortress retirement, so it maybe needs to get bug-fixed first on logistics. This bug fixing of currently existing abstracted economy model, could lead to simplified programming effort later, when economy is introduced. Such bug fix will have to come with new economy. Don't you think so?

[Shonai_Dweller]

The game will have been gutted three times over by the time the economy turns up, so no, fixing a bug now will probably make absolutely no difference. Besides leaving a placeholder which will have to be removed cleanly.

[GoblinCookie]

Introducing economy will not solve programming issues of multiplayer introduction. Toady's Dwarf Fortress has no server-client architecture. However, I think I saw some indie DF-like game on YouTube with pleasant 3d graphics. Maybe it already has?

Those are not programming issues, those are just things not programmed into the game!  Point I was making is that the economy is rather like having a multiplayer DF, except of course all the other players are AI.  That in turn makes actually adding multiplayer possible, so in a sense you were right about me wanting a multiplayer game. 

Economy down the basics is just flow of goods, no matter how they are generated.
Financial flow is ideally just reversed flow of goods, no matter what financial controls are applied.
Those are economy type independent elements.

Abstracted economy model is generating goods out of thin air upon demand. Right?
Abstracted economy model does not generate stockpile of goods at an embark for the game to play around with it, when trade is involved. Goods appear from thin air on demand. Right?
Retiring fortress creates huge issues with tracking huge quantity of items. Right?
Mechanics of abstracted economy applied on retired fortress is creating strange anomalies, which can be described simply as bugs. Players tend to build fortresses in peculiar ways, which pushes abstracted economy model into spinning out. Right?

On retiring the fortress, the game would have to convert the items into tracked 'bundles' which can be used by the AI economic logic when the player leaves the scene.  This part is easy, it is basically just like the population groups that we already have, you have an abstract number of a something and certain historical items are tied to that number. 

The hard part is that we have to also model the production *and* consumption of the player's fortress.  This is a problem because the player production and consumption will not correspond to an AI site, if upon retirement it suddenly produces more or less than before or consumes more or less we will create an economic storm, whole sites might well starve to death simply because the player retired!

This means that everything about the player's production and consumption has to be turned into numbers that can be replicated in an abstract way.  The more complex the economy in fortress mode is, the harder it becomes to model it once the fortress has been retired.  That means that really you cannot build the Economy out of increasing development of the fortress mode economy, since fitting the fortress mode economy *into* the Economy is the main challenge the devs face, as opposed to simply making the Economy work in the abstract with AI players. 

The more complex the fortress mode economy is, the harder it is to fit into the Economy.

To add to the pile of programming issues with abstracted economy model, which is missing out entirely on any logistic system. I suggested functional, generalized and simplified logistic system's solutions, which will survive banishment of Quantum Stockpiling and maybe could even improve FPS. Solutions like warehouses and shops, which will fit any possible chosen economy system, even as dumb as planned real communism, but will not fit an abstracted economy model where goods drop from thin air.

So, how we can marry abstracted economy model with the simplest elements of logistics like Trade Depots, Trading Caravans and numerous stockpiles without causing mayhem at fortress retirement? Particularly, if this part of DF still after years of development is causing unexpected weird bugs. Answer is pretty simple, a new abstracted economy model!

That means my suggestions, if to be implemented before big economy release, need at least some conceptual work on flow of great river of goods. Not to mention the economic system flavors (tribal, feudal, corporate, liberal) engaged with creating goods. As I can see it, the devs have a clear path here.

Unified logistics (could exist without rewrite of abstracted economy model and actually Trade Depots and caravans already do) ->
Economic systems (need rewrite of abstracted economy model for goods generation and needs to factor in unified logistics like at least Trade Depots and caravans) ->
Specialized Logistics (ship-caravans, player owned ships and player owned caravans in fortress mode and player owned wagons and ships in adventure mode)

One unexpected result of a mix-market economy model there, could be mixed race fortresses, which can produce not only large cloths, but also foreign items given expertise arrived with a guest looking for citizenship. However I doubt we will see it rather in DF in its current form. It could be for devs a viable possibility though to program, if they choose so.

Problem with large quantity of items is that this thing leads to FPS death in DF. This is why abstract model is favored over logistic+abstract model.

Stockpiles are not very difficult to model.  They are simply a number that controls the total maximum amount of an item that a site is able to have.  Upon retire the devs can very easily simply count up the number of stockpile spaces in our fortress, while for AI sites the size of the stockpiles that are generated when the site is loaded could be controlled by an abstract number.  The tricky part is adding new stockpiles to retired sites when the demand requires that they do so, but that applies to a lot of things. 

One of the problems the economy creates is the need for unfinished player sites to be 'finished' in some cases, if they are not yet economically viable but are retired. 

Problem with economy type in DF belongs rather to racial ethics, then to known to us economy systems. Specialized workshop types available only to given races for example. So in DF universe our chat liberal vs anti-liberal economy, tribal vs feudal, have really no meaning. So basically there can be one abstracted economy model, just with limitations in production for given races or immoral allowances for another race and that's basically it.

Well no.  The systems are relevant in so far as they effect the production, consumption and flow of goods between sites and individuals.  We can abstract away the actual mechanics by which this is done (whether it is buying-selling, giving-taking, taxes-grants etc), provided the player is not around  , but the systems still have to be taken into account when they affect the production, consumption and flow of goods.  If the player is around, we have to depict how the system actually works, but that can be ornamental rather than essential.  In other words, what happens is decided by the Economy but how it is depicted *as* happening depends upon the economic system.

You are over-complicating pretty simple things. Abstracted economy models like in real life need unified interface to coexists with each other.  The international trade, you know? For this part you already have caravans and Trading Depots. To redesign those logistics, is not the same with redesigning abstract economy models. We are not talking here different economic systems, but the unified logistic interface between them. Something that can bring more fun and flavor for players already playing DF and maybe something to attract new ones. Even before entirely new economic systems redesign is created. However as currently existing abstract economy model is creating things from thin air on demand, it maybe is playing loose ball with already generated items during fortress retirement, so it maybe needs to get bug-fixed first on logistics. This bug fixing of currently existing abstracted economy model, could lead to simplified programming effort later, when economy is introduced. Such bug fix will have to come with new economy. Don't you think so?

The current abstract economy model is a placeholder.  The caravans are just conjured out of nowhere and the specific demands they make are randomly generated.  If there was only one economic system, working in the same way everywhere then we could build on the placeholder, because we could replace the random demands with specific demands that reflect simply the effect of the past actions of the player in either mode on the wider economy.  Or to put it another way, we can bundle everything together so the player is interacting with a general category (the player effects carpentry in general) rather than having to model the effect in a generalised way.

But instead we need to be able to break down all the economic systems into their relations of production, consumption and transfer of goods.  Only then can we model interactions *between* different economic systems be modelled which is real problem the Economy has.  This sounds fine, except it is not fine because it reduces the level of abstraction that the game can use in order to model the Economy. 

[Anandar]

I see where your coming from GoblinCookie,  and i agree particularly around caravan demands working off of previous trade experiences and player actions witin each mode, based on previous trades in fort mode caravans will put lower values due to oversupply of particular items or item types from former visits and offer more for items/types based on the rarity of and "NEED".
If the game ran say a year or 2 of time between abandonment of 1 and creation of another fortress (similar to world gen) could this be enough time to allow for any unfinished sites to become economically viable/stable enough given that the player wasn't a total bastard or assuming the fort was not abandoned due to being slaughtered by demons or fb etc.... and in doing this is there a potential problem for other sites to change too much which could cause problems also? Curious if this has potential given values are frozen at world gen completion...

[Sarmatian123]

[...]so in a sense you were right about me wanting a multiplayer game.

Multiplayer Dwarf Fortress would be epic. I wish that was possible. We live in times of "Star Citizen" and "Life if Feudal" indie multiplayer games and lone Toady is doomed into limited "Stardew Valley" game format. I have no idea even how extensive rewrite of DF would have to be to adapt DF to client-server architecture, or if this even is possible at all.

Though an idea of running Dwarf Fortress's both modes in a form of "Eve Online" sandbox could create a much larger, active and engaged community of players, who don't mind ASCII graphics instead of 3d. There could be 3d clients though working with the server too. However game doesn't support multiple CPU cores, not to mention multiple processors. Players complain about FPS death and strong server could carry the game despite some deficiencies, as work around for that.

Due entire mechanics of creating multiple worlds rather then plays, maybe distributed client-server architecture like "0 A.D." has on Linux would be more proper for DF. It wouldn't much differ from what DF is today, just it would be using the client-server architecture and allowing some 2-8 players games, if someone got computer strong enough to run server for them all.

If on Kickstarter would be launched a project for rewriting DF into client-server architecture, then there is no guarantee people would provide enough funds to realize this. Worth a gamble though, if Toady would like to learn how to hire additional programmers to help on project basis. Kickstarter could help also with hardware like funding computers for server for DF and maybe some funds to keep it running for some time too. There are no limits what could be done, so long people would fund it, right? Just devs would need balance the economic environment wisely and don't spend funds like crazy.

On retiring the fortress, the game would have to convert the items into tracked 'bundles' which can be used by the AI economic logic when the player leaves the scene.  This part is easy, it is basically just like the population groups that we already have, you have an abstract number of a something and certain historical items are tied to that number. 

The hard part is that we have to also model the production *and* consumption of the player's fortress.  This is a problem because the player production and consumption will not correspond to an AI site, if upon retirement it suddenly produces more or less than before or consumes more or less we will create an economic storm, whole sites might well starve to death simply because the player retired!

This means that everything about the player's production and consumption has to be turned into numbers that can be replicated in an abstract way.  The more complex the economy in fortress mode is, the harder it becomes to model it once the fortress has been retired.  That means that really you cannot build the Economy out of increasing development of the fortress mode economy, since fitting the fortress mode economy *into* the Economy is the main challenge the devs face, as opposed to simply making the Economy work in the abstract with AI players. 

The more complex the fortress mode economy is, the harder it is to fit into the Economy.

Economy is mere pushing around almost infinite number of transactions, which can not be comprehend or abstracted easily. You can write Perpetum Mobile, which works on paper like Adam Smith fantasy, but in reality economic system is a mere set of governmental regulations restricting transactions in some way. So, you see the "tracked bundles" of items and any transactions that will AI take based on their availability is the Economy.

When you have fortress retired, you list existing resources and add stockpiles to them. That's it. You list workshops with matching workforce and you have site potential productivity. AI economic logic should handle that.

Stockpiles are not very difficult to model.  They are simply a number that controls the total maximum amount of an item that a site is able to have.  Upon retire the devs can very easily simply count up the number of stockpile spaces in our fortress, while for AI sites the size of the stockpiles that are generated when the site is loaded could be controlled by an abstract number.  The tricky part is adding new stockpiles to retired sites when the demand requires that they do so, but that applies to a lot of things. 

One of the problems the economy creates is the need for unfinished player sites to be 'finished' in some cases, if they are not yet economically viable but are retired. 

So, you say factoring existing quantum stockpiles in is tricky by AI, as there are more goods then storage?

Well no.  The systems are relevant in so far as they effect the production, consumption and flow of goods between sites and individuals.  We can abstract away the actual mechanics by which this is done (whether it is buying-selling, giving-taking, taxes-grants etc), provided the player is not around  , but the systems still have to be taken into account when they affect the production, consumption and flow of goods.  If the player is around, we have to depict how the system actually works, but that can be ornamental rather than essential.  In other words, what happens is decided by the Economy but how it is depicted *as* happening depends upon the economic system.

People tried to invent such economic abstract system for centuries and failed. Even Adam Smith model is a loose approximation, detached from real economic mechanisms. All known economic systems are just governmental regulations, which are in some way having impact on economic transactions. You can't have economy without transactions and you can't have economy system without regulations.

The current abstract economy model is a placeholder.  The caravans are just conjured out of nowhere and the specific demands they make are randomly generated.  If there was only one economic system, working in the same way everywhere then we could build on the placeholder, because we could replace the random demands with specific demands that reflect simply the effect of the past actions of the player in either mode on the wider economy.  Or to put it another way, we can bundle everything together so the player is interacting with a general category (the player effects carpentry in general) rather than having to model the effect in a generalised way.

But instead we need to be able to break down all the economic systems into their relations of production, consumption and transfer of goods.  Only then can we model interactions *between* different economic systems be modelled which is real problem the Economy has.  This sounds fine, except it is not fine because it reduces the level of abstraction that the game can use in order to model the Economy.

Something like abstract economy model has to be an empty placeholder. For every economy systems in game, you need separate rule-set. For the economy itself, you need fill this placeholder with listing of all goods and resources available in the game. Then caravan does not need to be conjured out of nowhere and carry items, which are listed in this placeholder. To make sure no error happens we can apply an old method of double book keeping. Book keeping always works. Not to keep tab on checks and balances, but even when you start to conjure something from thin air... Be it magic or theft. I am afraid there is no work around for data base need in DF, if there is economy to be had. Any "abstract economy model" will be just another "conjuring stuff from thin air", no matter if you name it feudalism, tribalism or liberalism.

Anyhow adding few more QS workshops in game, working alike Trade Depot or ships working like wagons just on water, should not break any economic shenanigans you could possibly invent through programming.

[GoblinCookie]

Economy is mere pushing around almost infinite number of transactions, which can not be comprehend or abstracted easily. You can write Perpetum Mobile, which works on paper like Adam Smith fantasy, but in reality economic system is a mere set of governmental regulations restricting transactions in some way. So, you see the "tracked bundles" of items and any transactions that will AI take based on their availability is the Economy.

When you have fortress retired, you list existing resources and add stockpiles to them. That's it. You list workshops with matching workforce and you have site potential productivity. AI economic logic should handle that.

The problem that I referred to but you seem to have missed is that the player's economic activity is not necessarily in any way similar to how the AI would behave with the same population and resources.  Let us say that that one particular player takes 500 chairs and throws them into the volcano every year.  In order to get enough chairs the player buys chairs from the world market, which modifies the economy.  Then a whole world of production springs up that hinges upon the player throwing 500 chairs into the volcano every year.

Then the player retires his fortress.  Since the whole world economy has become dependent upon the throwing of 500 chairs into a volcano, the act of the player retiring now has all the destructive power of a natural disaster.  Since the AI did not destroy 500 chairs the next years, a whole load of sites are suddenly ruined, causing a knock-on effect that cascades through the whole world to major horrific effect.  This is not necessarily a problem, but not really something that does not really exist in-universe, if there are to be disasters should they not be actual disasters rather than the player stopping playing one part of the game?

What has to be done here is to keep track of what the player does numerically.  If 500 chairs get melted in a volcano every year, then when the game is retired 500 chairs will continue to be melted in a volcano every year, so that we end up with a continuity when the player retires the fortress. 

So, you say factoring existing quantum stockpiles in is tricky by AI, as there are more goods then storage?

I am saying the opposite.  That getting rid of quantum stockpiling means we need to take into account storage-space. 

People tried to invent such economic abstract system for centuries and failed. Even Adam Smith model is a loose approximation, detached from real economic mechanisms. All known economic systems are just governmental regulations, which are in some way having impact on economic transactions. You can't have economy without transactions and you can't have economy system without regulations.

We don't have to worry about the regulations.  The game mechanics are regulations, the hard part is making sure that the mechanics represent the functioning of the different economic systems, even when they coexist in the same society.  For instance stealing, giving, taxes and commerce can all exist in the same society, but are mechanically different economic systems. 

Something like abstract economy model has to be an empty placeholder. For every economy systems in game, you need separate rule-set. For the economy itself, you need fill this placeholder with listing of all goods and resources available in the game. Then caravan does not need to be conjured out of nowhere and carry items, which are listed in this placeholder. To make sure no error happens we can apply an old method of double book keeping. Book keeping always works. Not to keep tab on checks and balances, but even when you start to conjure something from thin air... Be it magic or theft. I am afraid there is no work around for data base need in DF, if there is economy to be had. Any "abstract economy model" will be just another "conjuring stuff from thin air", no matter if you name it feudalism, tribalism or liberalism.

Anyhow adding few more QS workshops in game, working alike Trade Depot or ships working like wagons just on water, should not break any economic shenanigans you could possibly invent through programming.

If there is nobody around the witness it, then there is no problem in a computer game with conjuring things out of thin air.  The game will never be able to handle actually modelling the whole world economy as if it were fortress mode, so yes we are always going to be conjuring things out of thin air and nothing wrong with that as long there is nobody around to witness it.

What makes the present caravan a placeholder rather than an incomplete mechanic is that given the existence of different economic systems in the game, we cannot just model the effect of the player on the game economy *in general*.  If the goods you sell to the caravan will never exit the economic system of buying and selling, then we can model the whole world by modifying abstract values at a site/civilization level in response to the player's behavior.  What requires that we track the actual production, consumption and transfer of the goods is that there are other things like giving, taxes, stealing and so on that mean that we have to figure out where the goods end up. 

If bandits waylay your caravan once it leaves the site, this modifies the whole economic situation since the goods are now part of the criminal economy rather than the legit one. 

[GoblinCookie]

I see where your coming from GoblinCookie,  and i agree particularly around caravan demands working off of previous trade experiences and player actions witin each mode, based on previous trades in fort mode caravans will put lower values due to oversupply of particular items or item types from former visits and offer more for items/types based on the rarity of and "NEED".
If the game ran say a year or 2 of time between abandonment of 1 and creation of another fortress (similar to world gen) could this be enough time to allow for any unfinished sites to become economically viable/stable enough given that the player wasn't a total bastard or assuming the fort was not abandoned due to being slaughtered by demons or fb etc.... and in doing this is there a potential problem for other sites to change too much which could cause problems also? Curious if this has potential given values are frozen at world gen completion...

The problem here is the whole thing about the AI having to physically complete the infrastructure of an unfinished player site.  If I don't finish my site enough that it become economically viable, how can the AI know how to complete it for me?

[Bumber]

Storage space isn't really an issue. Stockpiles are just zoned areas, so any free accessible tile can be one.

[Bortness]

I see where your coming from GoblinCookie,  and i agree particularly around caravan demands working off of previous trade experiences and player actions witin each mode, based on previous trades in fort mode caravans will put lower values due to oversupply of particular items or item types from former visits and offer more for items/types based on the rarity of and "NEED".
If the game ran say a year or 2 of time between abandonment of 1 and creation of another fortress (similar to world gen) could this be enough time to allow for any unfinished sites to become economically viable/stable enough given that the player wasn't a total bastard or assuming the fort was not abandoned due to being slaughtered by demons or fb etc.... and in doing this is there a potential problem for other sites to change too much which could cause problems also? Curious if this has potential given values are frozen at world gen completion...

For one, the AI is WAY smarter than you are.

The problem here is the whole thing about the AI having to physically complete the infrastructure of an unfinished player site.  If I don't finish my site enough that it become economically viable, how can the AI know how to complete it for me?

[Sarmatian123]

The problem that I referred to but you seem to have missed is that the player's economic activity is not necessarily in any way similar to how the AI would behave with the same population and resources.

Well, as I wrote earlier, you need to factor in sites resources, workshops and workforce. For realistic economy you need to keep database, which is tracing all transactions. For unrealistic economy you conjure from thin air with help of rule-set, but one based on dev's common sense. AI is regulated by ethics (an economy of the race) like players, so same economy/regulations apply to both. If player couldn't build high boots, then AI won't either. Even if player traded cups and mugs with caravan to get weapons and armor and build some sort of farm plot, brewery, kitchen and build tavern. AI is not obligated just to maintain those production lines, given there is overflow of migrant population. AI should go for balancing act, so site will not have massive emigration to new player's embark. DF is not a trade game. Only few embarks missing some resources.

Also, if you don't want always to play with one rule-set for goods, when conjuring from thin air, you can let AI to shuffle some ideas of products. Player maybe was producing 500 chairs to be thrown to lava, but AI can stop all production chains in when surplus of 10 items. Factor in rest of available untapped resources on site, create new workshops as needed and workforce available and conclude that there could be 10 further items of each sort to be in stockpile for the rule-set.

Ultimately player from another embark can add changes to the rule-set as well with existing mechanics of tender offer, by requesting some items (or groups of items, that would be an improvement to the existing system though) in larger amounts then those 10. Right now request goes in 1-5 stages, so how sounds +3,+6,+12,+25,+50? Though it would be conjuring from thin air, without database, specially in case of mining. Database searches and transactions could be speed up and be very fast. It just takes to learn few SQL phrases, which you inject into your programming. AI could take advantage of database to set up its own choices (clustering of examples you know?), besides the ones based on rules and react more intelligently. However you would need to put entire world in the database first to play with it from start too.

The thing is to let lose the site-2-site trades instead of civilization trading caravans. Here my suggestions about dedicated Trade Depots factor in, as solution. You see? Even if you conjure from air items, which are on a list.

So, you say factoring existing quantum stockpiles in is tricky by AI, as there are more goods then storage?

I am saying the opposite.  That getting rid of quantum stockpiling means we need to take into account storage-space. 

I see. This is not implemented yet in DF. This is true. Warehouses and Treasuries, which I propose, instead of stockpiles could magically take care about this QS future issue, solving it permanently. AI would need just have some free space and some blocks or logs to build some more of those magic QS-Depots. Stockpiles could be still useful for linking production chains. However QS-Depots would be required now for some refined items like prepared food barrels. Preventing vermin from spawning inside of them or entering into them for example. Alike Trade Depot does right now.

We don't have to worry about the regulations.  The game mechanics are regulations, the hard part is making sure that the mechanics represent the functioning of the different economic systems, even when they coexist in the same society.  For instance stealing, giving, taxes and commerce can all exist in the same society, but are mechanically different economic systems. 

Currently only economy differences in DF, between races, are those based on moral ethics. Those do restrict which civ can build what workshops. Nothing else. So each set of regulations = an economy.

If there is nobody around the witness it, then there is no problem in a computer game with conjuring things out of thin air.  The game will never be able to handle actually modelling the whole world economy as if it were fortress mode, so yes we are always going to be conjuring things out of thin air and nothing wrong with that as long there is nobody around to witness it.

What makes the present caravan a placeholder rather than an incomplete mechanic is that given the existence of different economic systems in the game, we cannot just model the effect of the player on the game economy *in general*.  If the goods you sell to the caravan will never exit the economic system of buying and selling, then we can model the whole world by modifying abstract values at a site/civilization level in response to the player's behavior.  What requires that we track the actual production, consumption and transfer of the goods is that there are other things like giving, taxes, stealing and so on that mean that we have to figure out where the goods end up. 

If bandits waylay your caravan once it leaves the site, this modifies the whole economic situation since the goods are now part of the criminal economy rather than the legit one.

Why would you want to model player's impact on economy? Economy is economy no matter by whom or what it is resolved. In worst case scenario you have site bankruptcy and you just factor that case in.

Abstract model of economy is non-functional. Conjuring from thin air patches its gaping holes and makes it workable. So you are best with making the economy by setting up those regulations for each civilizations. At least then abstract economy model will be less fake. You will still need to conjure item from air to make it work, but it will feel to players more realistic, then right now with what goods Trade Caravans are offering.

To kick start world economy, you need just send in tax collectors. Let mountain-home collect yearly tributes from local sites. That's it. You don't need anything else really in DF. Just taxation. Trade Caravan already has existing donation system in it. There, whole economy was born like in real life.

About the rest of economics, which you may want in economy DF release:
-Dwarves using dispensaries instead of stealing items from stockpiles
-Dwarves earning money as wages/shares
-Embark creating currency and storing it in vaults
-Embark manufacturing requested and desired by Dwarves items at highest quality to get some of that coinage back.
You just use double book keeping here for check and balances. Store unsold goods in QS-Depots, stockpiles while with sold goods in chests, bags, coffers, cabinets and such. You keep on Dwarves info how much fortress owns them in wages and if it goes over the top (lets say 12 months worth?), then Dwarves will revolt or go to strike, unless receiving from dispensary owned them coins or desired items. You keep unspent coins in vault instead of dispensary btw. Keep in mind, that some goods like cloths, food and alcohol are workplace privileges for workers. They don't need pay extra for them. Rooms either. Though own personal chests/coffers/bags would make here some sense.

Interesting side effect of Dwarves owning coins. They can buy goods from Trading Caravans as well, so here having local shops for each embark, even your own, could be also an option. Metal can be rare, so financial flow should be bound to metal coins imho.

Caravans instead of traveling shop, could become just another safe goods transportation method. Then you can allow wagons and even ships in adventure mode to be bought by players. That would rise some fun.

Unless you strive to create ultimate multiplayer DF with player-driven economic system, then you can get away with conjuration from thin air using sets from lists. Else you need seriously consider using database. Actually you could create economy. Then create multiplayer. Then release some items from being conjured and allow them to be manufactured and consumed by players themselves. This is how Eve-Online step by step liberating items from their conjuration, created rich and thriving economic system. Both those steps of development would move DF to entirely new level of gaming. DF could start compete with big mmorpgs out there.

Though I would suggest always allowing for simplest graphics, which could allow DF multiplayer to be played even on mobile or tablet without much fuzz. Big AAA titles go for flashy graphics and limit their player base.

PS. Unless you want just freeze with the rest of the world player's embark, after retirement. With all its oddities. I advise against that. Let AI think through and sort local economy first. Patch it and balance it as best it can. Then freeze the result. Though you would need to go through rules for AI, so no crazy stuff results from AI actions like 1000 Dwarven merchants, wagons scattered everywhere and such mayhem. Though, you always could try to consider imitating economy and even keep open gates for future development like multiplayer and real player-driven economy. So long you keep modular architecture and keep track of their interfaces, then there should be no programming chaos either.