quote: Originally posted by Dienstag Thanks everyone for the praise and criticism of my spy-screen proposal. It seems to have brought out many good ideas by others, such as Mad Viking's confidence factor, lago's effect of trade and government ideas, and many others. I'd be happy to try to compile these ideas into a comprehensive idea of espionage, but I've got Finals coming up (should be studying or sleeping now) and as MKL aluded to, this was a thread on "unconventional warfare."

There, all done now. I was waiting for someone post something in this thread before I was done, to make me look silly, but it looks like I got through

Obviously, spies in Civ II gave completely accurate information because programming the shades of grey must be a nightmare!

So the two broad categories are:
1. (Opponents civ/Your civ) has (greater/lesser) (military/tech/economic/reputation) strength than actual. This could be reflected in modifications to the powergraph. Eg. Each player sees the true powergraph through modifiers. The modifiers are affected by the amount of production/gold that supports a player's trade and intelligence efforts.

However, misinformation given with the specific intent to manipulate the receiver into believing in a false construct as though it were true is called "disinformation." For instance, information leaked by the Allies in WWII to convince the Germans that the invasion would be in Norway was disinformation. It's a square can be a rectangle but not the reverse sorta deal.

Unless I'm mistaken, I think that's it.

Slingshot - Sounds good.

The intelligence screen is good because it adds more to the civ feel instead of "collections of city states" feel.

I suggest the following:

1. The intelligence budget is to be divided between intelligence (used against other civs) and counterintelligence (internal security) operations.

2. The success chance of your agents is modified by things such as tech and SE.

3. There is a small chance to recruit enemy spies to become your double agents. Then again, they maybe just pretending to work for you

4. Likewise, your agents might become double agents for an enemy.

5. Counterintelligence should normally be just stopping enemy agents from stealing your information. However, if an enemy agent is discovered, you should be given the option of: 1. Recruiting him 2. Giving him disinformation or 3. Making him disappear.

6. You must train your initial spies, but then they can be sent into enemy territory to recruit more.

7. There is a chance for your spies to recruit more in enemy cities. The success chance is a function of the happiness of that city: the greater the happiness, the less likely you will find somebody who is willing to betray her own country. Specialists have the least chance of turning into an agent, followed by happy citizen, content citizen, and drones. Drones rendered content by martial law have double the chance of turning into enemy agents.

Having said that...

Harel - Thanks for another thorough examination of the possibilities. There's lots of good ideas in there. If possible however, it might be best if we nailed things down to what we considered the most important functions. Firaxis aren't going to implement anything that's too complicated.

Whilst a lot of these are good ideas, I was thinking more along the lines of spies 'acquiring' this knowledge based on their location and funding, rather than having 29 different 'actions'. Of course, some of them would be actions, such as bribing a city, or assasinating a leader.

Perhaps we should make a list of what a spy would be able to 'learn', and what things in that list would require an 'action'. Remembering of course that these actions would be initiated in the Espionage Screen rather than the normal map (in the spirit of the idea).

That's what I'm thinking anyway. What do you think?

Since I am all for the creation of a Special Forces unit in Civ3, I feel that part of intelligence operations in your list really are "war-time operations" that can be only performed by the Special Forces unit.

MKL,

I was thinking about converting cities to your side. It occurred to me that this should be a lengthy operation with a very low chance of succeeding. A player should never be able to bribe an enemy city outright. What should happen should be the player must conduct a series of operations designed to fan separatist flames and to cultivate loyalty to your country. It should be after a number of turns before a player is given the option of trying to "bribe" the city, which to me is just a military uprising. However, success is not automatic, as the people you have been funding could acutally be more sympathetic to yet another country, or in some cases, want independence.

I reckon adding garrisons to a city does not reduce the chance of success, as there well maybe separatists in the armed forces. However, counterintelligence will put a damper on such operations, even completely shutting them down.

I know, you summarized it and yes it was a good one (except you reduced my suggestion to replace spy units with spy missions to one freaking sentence! ) It's just the way it comes across that looks so...arrogant.

Just use a little forethought next time. To others: don't let this list discourage you! More ideas are always welcome!
--------------------

although I would like to have access to all of those things, but it's just too easy (or is it cause I was playing citizen level?).
Rules should be:

-You may have only some part of the info. You can choose what area or randomly let the spy decide.

-you lose the spy unit. it has to stay in the city and continue infiltrating

-you have access to info for some time, until your spy is discovered. the longer your spy stays the bigger the chance of discovery. or maybe after say 15 years the spy is so well nested, that his chance of discovery will get lower each year.

-spys will die: meaning after 40 - 70 years you lose the unit.

-if enemy has his psys in infiltrated city, chances of discovery are higher

-on dicovery of enemy spy (with your spy unit) you can choose what do do with him:
*a. assasination - will get rid of him for good, but may cause attitude of his nation becoming hostile. and if spy was high ranked it may cause incident (high level "diplomat" or "ambassador" killed)
b. deportation - sent him back to his civ. causes origin nation ingrity down.

c. double crossing - this thing is way cool. you pay him, so he gives disinformation to his nation and gives you information. but if his nation will increase how much they pay him, they might buy his loyalty back. and then you might do the same again.

*d. disinformation - for this turn, the spy gets wrong info and sends it to his nation.

e. trade for your captured spies. (idea needs work)

* marked things can only be done by another spy unit attacking the enemy spy.

Also I didn't like that in SMAC you couldn't sabotage units. only bribe them. I want to sabotage without any diplomatic penalty (unless discovered. discovery has low chance since in open field).

Bribing cities has always struck me as purest nonsense. Inciting revolt is perfectly legitimate, but you should then wait to see if the breakaway element appeals to you for military intervention. Agreeing would effectively be a declaration of war - and the breakaway state may be less than pleased if you go about absorbing it rather than withdrawing them once it has established its independance.

To keep the number of operations to a minimum, all similar attacks should be labelled under one heading. For example low intensity low risk incitement to revolt would effectively be low profile funding of existing dissidents in the nation. Low budget high risk operations would equate to assassinations. High budget high risk would equate to mass arming of pseudo terrorist organisations. High budget low risk could be interpreted as flooding drugs into the country.

"Pardon?"

"I want to buy it - all the people, all the buildings, you know, fully loaded."

"Oh that! Let's see, we're on special this week, that'll be 2,371 gold."

"Ooh, what a deal. That includes Women's Suffrage, right?"

I'd like to strongly support all those suggesting that bribing a city is far too easy. As suggested, it should be a multiple step process, with different options and strategies, that could end up costing you a lot of money and some reputation and still fail.

There are really two distinct functions - intel / counterintel and "special ops".

Assassination should be somehow limited or difficult, IMHO. Poisoning water, sabotaging military units, planting nukes etc. (The 007 stuff) involve a different sort of personnel than intel gathering. This could be represented on the screen as a separate budget. Reputation hits for getting caught should cause severe trade penalties but only for a couple of turns, depending on the severity. And briefcase nukes don't blow up the city and the surrounding squares! Even one huge nuke couldn't do that - what is it a briefcase MRV?

1. General espionage
Rationale: When you start setting up a spy network you can't specify the job they will manage to obtain or which members of the opposition they will manage to convert, blackmail, seduce e.t.c. All you can do is aspire to improve your intelligence network and choose whether to send in more agents or attempt more high profile targets. Low risk = civilian intel and outdated tech, medium risk = civilian tech and more detailed city info, high risk = military dispositions, extreme risk = latest military tech. The bigger the budget the more items of info you get each turn (most would just be map updates, city info and civilian unit locations etc which need not be announced unless you dialled the reporting level up to max.) A sudden drop in financing could see your foreign agents casting around for a new employer...

2. General counter-espionage
Attempting to deprive the opposition of key information, locate, eliminate or subvert their agents. Funding level is all important, the risk level could be used to run high gambit 'stings' to expose enemy agents. Failure would weaken your overall security level as your agents become known to the opposition but success could wreak havoc on the enemies network and improve your own chances against them. Again, don't cut their paychecks to drastically or you could be in trouble. Some of these guys have had access to your own secret reports for years!

3. Industrial sabotage
An attempt to weaken industrial output (shields) in the target nation. The smaller the output the harder it will be to make a noticable difference. A target city around which efforts should be centered may be needed (stop that Wonder!) but mainly affects at a national level. A sudden outbreak of strikes will cause mass witch hunts for foreign donations to union officials.

4. Economic sabotage
An attempt to weaken the economy (trade arrows) in the target nation. Otherwise identical to 3.

5. Rabble rousing
This covers all forms of terror attacks on the civilian population. From low budget high risk missions like lone gunman assassinations to high budget low risk operations like covertly funding every lunatic organisation that moves. A city that riots or revolts near your border may give you the opportunity for annexation.

6. Military sabotage
Attempts to reduce the military effectiveness of the target nation. Pick a city to focus your effort upon. At low risk long stationed border units near your country may offer to switch sides. medium risk may see military garrisons become damaged or have their morale drop. With high risk a particularly enterprising spy may manage to sneak open the town gates or blow up the magazine just as you launch your assault. More about emulating a Troy than a Jherico.

7. Things for a diplomat unit to do
The espionage screen should not be used to establish embassies or provoke diplomatic negotiations. Save that stuff for the guy trudging round the map in a top hat. Since he's feeling a bit underpowered now, perhaps his role could be merged with that of the explorer.

The only thing I'm unsure about is the role of the diplomat, but that's getting into the territory of diplomacy models anyway.

I think the general idea of spies slowly getting information is great, and much more realistic. And counter-intelligence could be both fun and sometimes frustrating (if the AI knows how to use it properly). Imagine the fun you could have in a multi-player game handing out the wrong info to your enemies!

On the other hand, it's pretty tough to visualize, say, Spain paying the Mayor of Minsk a wheelbarrow full of euros and suddenly having a foothold in Belarus. The whole thing looks silly when viewed in a modern context.

Therefore, my suggestions would be as follows:
1: The bribery mechanism would remain intact until the modern age, although prices should go up. As it stands, it's way too easy to simply buyan enemy empire outright. Perhaps some modifications could be made to make the action less certain and more dangerous.

2: In the modern era, bribery of cities should take on a less concrete aspect. Perhaps you'd receive a portion of the city's economic and resource output, complete and continuously-updated info on improvements in the city's radius, increased unhappiness in the city, and continuous information on all units the city supports (or that the city built, if we do away with support from a single city) no matter where they are on the map. Since that is less useful than simply grabbing the whole city outright, the price for that should drop.

We should be careful that we do not make Civ3 something like Spy3... Espionage should be important, but must not require our constant attention.. We should also have to take care of military and economy...

I agree that spying should not require our constant attention. I think this is why a spy stationed in an enemy city gradually giving you small pieces of information is a good idea. You wouldn't have to pay him too much attention, but they'd still be quite useful.

It was also suggested that the longer a spy has been working for you, and the longer he's in a city, the more effective he becomes. So it's not like you'd be moving them around all the time.

Also, keep in mind that, in most times and places, people identified themselves by what city or (at most) region they belonged to. A medieval Italian might refer to himself as Milanese or from the Piedmont, but the idea of calling himself Italian simply because he spoke the language (and, in Civ terms, probably belonged to the "Italian Republic") would have seemed a bizarre and alien thought. That's why, for so much of history (especially European history, but this holds true for the pretty much every civilization for which we have records), it was seen as perfectly respectable for a city's leaders to offer their allegiance to the strongest power that happened to be in the neighborhood at the time. So the idea of "suddenly becoming part of the enemy nation" is an imposition of modern thought upon a situation that doesn't support it. The aforementioned Milanese would probably have thought of it more in terms of getting on the winning team so as to better share in the spoils.

Also, we have to remember that this whole system is necessarily abstracted, as is every other aspect of Civ. It's very tempting to see a spy on the screen and think of it as a lone individual doing specific actions, but in reality it represents an allocation of resources by the owning civilization that is sufficient to achieve a specific goal -- the Natasha Fatale-looking figure is just an abstraction of the focus of the effort, and may represent anything from dozens of actual agents ferreting out information to an extensive and long-lasting propaganda campaign that makes a sizeable portion of population of the target city prefer you to your enemy. Thinking about it as a sort of Jane Bond can only mislead you about what it should and should not be able to do.

Sorry if the foregoing was a bit pedantic. I have a tendency to get that way at times. I'm working on it.

---

1. The way spies are used in the Espionage Screen is rather different to the way they were used in previous installments. A major part of this is the idea that you will not be able to trust any spies you are financing. These spies could be working for someone else. Either to provide your enemies with information, or to give you misinformation.

Keeping this in mind, is it still best that we have to "build" our spy units? Is there a better system for doing this? After all, who's going to want to disrupt their production queue to build a unit (although not in the traditional sense) that may not even be loyal to you? (On another side note, production is a misrepresentitive way to 'build' a spy. Money is far better.)

Should you have to scout for potential spies amongst your population? Should spies 'offer' information to you? What is a better system?

---

2. I've suggested (and have not heard any arguments against this) that spies stationed in a city should slowly acquire information rather than just respond to unit actions such as stealing maps, or viewing a city screen. (Another side note - some 'actions' such as 'plant nuke' could perhaps still be included as actions.) Anhyway, there will be a number of different types of info a spy may 'acquire'. (I'll try to draw up a list of these soon).

With this in mind, should we able to prioritise which type of information we would like the spy to concentrate on trying to get? This could be as simple as selecting one type. (eg. choosing 'steal tech' which gives you twice as much probability as you would otherwise have of stealing a tech). Or it could be as complex as giving you percentage ratings to help you prioritise. (50% to steal tech, 20% to steal maps, 20% to locate enemy units, 10% to find what the city is building).

These 'priorities' should not guarantee that you will find any information, nor should failing to give a priority disclude you from finding certain information. It would just let you focus more on what you wanted to find out. It's still up to how skillfull and loyal your spy is, and how good the enemy spies are at keeping the information from you.

Priorities could be set on a per spy basis, or set at a global level. Perhaps a better compromise would be to have all the spies who are spying on the same civ to have the same priorities. That way you wouldn't have to set too many, but you'd still have control. It's a matter of micro-management vs. control.

---

3. Should you have to keep track yourself of which spies you believe are staying loyal and truthful, or should the computer come up with a figure to help you decide? I believe that some sort of indication would be helpful to the player so that they don't have to focus all their attention on what their spies are doing. On the other hand, being given a figure (perhaps loyalty as a percentage) takes away some of the uncertainty that I found interesting about this idea. Perhaps you could be told how much of the information the spy has given you has been proven to be true. You should have some way to track your spies (and know which are your best ones), but not be totally certain that your spy is 77% (for example) loyal. Any ideas?

---

I'm done now. What do you think? I still think an Espionage Screen could be really cool, especially in multi-player, but I think we need to flesh out the way it would work a bit better. All comments are appreciated.

Thinking as I am typing, so well anyway...

I think that there should be a improvement that is the CIA/KGB/Scotland Yard/etc. Headquarters that would allow you to train spies that would be more effective. They would have a higher loyality rating and would return greater information.

You should have a recruiting office that hires the number of spies that you pay for out of college (your best) and then trains them. This way you could have 2% of your population involved in spying.

Definetly in need of tracking spies. But I think it would be best simply if a screen tells you what city you are receiving the information from and the % of what that city's spy has revealed to be true. In that way it would be a less of a hastle then thinking about how many spies you have, where, and their individual percentage.

I'm just wondering about where to go from here...

Back to the wall of frustration. It would be nice for a programmer to give some feedback.