1.34.3 Hands Off

[doktarr]

With the submitted events for province based spawning of Thrace CoT even a late OE takeover will work well. It will also be created always if taken on time by OE, since the b option of their event is removed. I think that this is how far we should go.

I agree that the CoT situation is well in hand now. But remember that the AI won't realize that capturing Thrace will spawn a CoT, so it will undervalue the province. A human player would never let such a prize slip by.

I don't like fixating the outcomes of history. Forcing another nation that take Constantinople to cede it to the turks if they manage to take it, however briefly, is not the way I like the game to work. A strong Venice, for example, wouldn't let go of it either, and it wasn't the sole goal of Ottoman expansion, as proved by how many other conquests they did while in the vicinity of Constantinople, or even allied to the city.

Yes but we have a number of special events happen based on ownership of Thrace - events that go beyond just the CoT. If we really think that owning Constantinople would not become a crucial objective for the Ottomans, then they should probably be able to achieve these bonuses (CoMD, et cetera) without taking Thrace. Otherwise, we need to force the Turks to go after a province that is much more valuable than the AI files allow them to realize.

There are many other places that nations yearned to take or hold, without any special events. I think the game should have it's natural flow and these rare circumstances should be let to happen. OE is currently not a problematic nation.

It is not a problematic nation in terms of military strength or overall European advance, but it underperforms in the Levant and Africa fairly consistently.

We could give them more minor leaders spread out during the period (I'm sure there were plenty of them ), that would increase the likelyhood that they take it themselves, and reducing the already small chances that an ally take it.

This does not preclude an ally or friendly nation capturing Thrace - it would at best only slightly reduce the odds. There is a risk that this would overpower the Ottomans unless you're suggesting lots of 4/2/2 leaders. That would actually be fine, but again I'm not sure this addresses the problem.

Besides their allies will have great trouble holding a greek orthodox city completely surrounded by Ottoman land.

In my experience, by the time it revolts away, it's usually the 16th century. The barracks are in the wrong place, the capital never moves, the free conversions never come.

 

[unmerged(31425)]

For the Thrace problem you could make an event that is province specific for Thrace, contingent on Byzantium not existing, AI only, and contingent on being an ally to the Ottomans. It could hand over Thrace to the Ottomans. That could solve alot of problems right there. Except of course that this event would be the same kind as the vassalization event that was rejected.

 

[Norrefeldt]

I agree that the CoT situation is well in hand now. But remember that the AI won't realize that capturing Thrace will spawn a CoT, so it will undervalue the province. A human player would never let such a prize slip by.

This is undisputably true. We can never make the AI realise the value of Thrace. We could help the early Ottoman AI's go for rich provinces first though by raising the "enemy" value from the default 1.0 it got now. With a tax value of 17, much higher than all other available provinces at the same time, that ought to help. Needs testing of course.

Yes but we have a number of special events happen based on ownership of Thrace - events that go beyond just the CoT. If we really think that owning Constantinople would not become a crucial objective for the Ottomans, then they should probably be able to achieve these bonuses (CoMD, et cetera) without taking Thrace.

At least the event CoMD ought to be able to trigger much later then it does right now (1491).

It is not a problematic nation in terms of military strength or overall European advance, but it underperforms in the Levant and Africa fairly consistently.

I haven't seen this connected to taking Thrace or not, since they often take Thrace and still don't get into the right wars and show enough commitment to get what it should here. Of course economic resorces will matter but even with the proposed events this problem will remain to a large degree. AKK often sits very big in the Levant and will stop OE, with or without Thrace. Until we drain their resources by forming Persia we'll have this problem.

This does not preclude an ally or friendly nation capturing Thrace - it would at best only slightly reduce the odds. There is a risk that this would overpower the Ottomans unless you're suggesting lots of 4/2/2 leaders. That would actually be fine, but again I'm not sure this addresses the problem.

I was thinking of pretty bad leaders, even if I think we could find a few decent ones up to values of 3/3/3, higher than that would probably unbalance things. It would help OE steal sieges from their allies. Their allies have very few leaders 15th c. Any leaders they have ought to be checked to see if their rank is correct and cannot be lowered.

In my experience, by the time it revolts away, it's usually the 16th century. The barracks are in the wrong place, the capital never moves, the free conversions never come.

That can be fixed by prolonging the dates for CoMD.

What I have proposed doesn't solve the problem 100%, the only way to do that is writing events for OE to get it. People seem more and more eager to just set things straight by events, without first trying to do normal modding. I think this development is a sad one and that it will eventually lead to a very constrained mod with reduced replay value.

 

[doktarr]

For the Thrace problem you could make an event that is province specific for Thrace, contingent on Byzantium not existing, AI only, and contingent on being an ally to the Ottomans. It could hand over Thrace to the Ottomans. That could solve alot of problems right there. Except of course that this event would be the same kind as the vassalization event that was rejected.

Check post 133 in the submission thread. It is basically just that. And as you know, I support the vassalization event as well.

 

[doktarr]

This is undisputably true. We can never make the AI realise the value of Thrace. We could help the early Ottoman AI's go for rich provinces first though by raising the "enemy" value from the default 1.0 it got now. With a tax value of 17, much higher than all other available provinces at the same time, that ought to help. Needs testing of course.

This would be a good thing to put in an "anti-Byzantine" AI. Writing such an AI would be a good step towards solving this problem in an less invasive fashion. But such an AI would interfere with other Ottoman expansion, so its use would need to be restricted to a fairly narrow time range IMO.

At least the event CoMD ought to be able to trigger much later then it does right now (1491).

Probably. Should the barracks move as well?

I haven't seen this connected to taking Thrace or not, since they often take Thrace and still don't get into the right wars and show enough commitment to get what it should here.

True, I wasn't implying it was necessarily a Thrace problem; I was just disputing that the Ottomans are performing adequately currently. I don't think they are.

Of course economic resorces will matter but even with the proposed events this problem will remain to a large degree. AKK often sits very big in the Levant and will stop OE, with or without Thrace. Until we drain their resources by forming Persia we'll have this problem.

Certainly true. It's also a function of the insufficient alliance structure at game start. With expanded diplomatic relations we could ensure that the sheep go after each other and the Timurids rather than the Mamelukes.

I was thinking of pretty bad leaders, even if I think we could find a few decent ones up to values of 3/3/3, higher than that would probably unbalance things. It would help OE steal sieges from their allies. Their allies have very few leaders 15th c. Any leaders they have ought to be checked to see if their rank is correct and cannot be lowered.

I understood the point and don't have a problem with it - I just don't think it will make a significant difference. When the Ottomans decide to join the siege they are usually the first ones there. The problem is when they are not part of the war or don't bother to send troops there.

What I have proposed doesn't solve the problem 100%, the only way to do that is writing events for OE to get it. People seem more and more eager to just set things straight by events, without first trying to do normal modding. I think this development is a sad one and that it will eventually lead to a very constrained mod with reduced replay value.

I agree that we have not explored every other option. The biggest change I want to make to our current approach to the large nations is to have more AI switching. But rather than trigger switching by date (as Daywalker does) or monarch (as PAI does) I would suggest switching based on situations. This makes the AI think more like a person.

So the Ottomans would start by going after the Turkish, Greek, and Balkan minors. Then provided they have achieved a modicum of success, they go after Constantinople. Then they expand in Europe, then they expand in the Levant and Egypt, then they try to grab the whole Eastern Mediterranean.

That said, much of what I'm suggesting is not the "sledgehammer" style event. My proposal can be considered in three seperate parts:

1) I make allies of an AI OE either give Thrace to the Ottomans, or break their alliance and defend it. I think this is the most realistic/important of the three. I should probably add some reduced badboy and improved relations to it, but it is this situation (an ally or friend holding Thrace) that is currently the MOST unrealistic and the MOST likely to cause problems.

2) I force enemies of an AI OE to go to war with the OE if they hold Thrace past a certain date. This is a bit of a hack, I'll admit. It would perhaps be better to just have the event cut relations between OE and the Thraceowner, and/or trigger the antibyzantine AI. It needs to be tested. It would be nice if a "softer" approach could work here.

3) I force an immediate secession of Thrace if the OE controls it. Of the three, this is the most "sledgehammeresqe". And yet, I think it is pretty realistic.

Imagine a war between a Thrace-holding Venice and the OE. The Ottomans capture Thrace. The Venitians win some sea battles, and maybe get Bulgaria. The most likely outcome here is a white peace or an indemnity. This is absurd! Once the OE decided to capture Constantinople and succeeded, there wasn't any going back. This was the big moment; the end of the middle ages. It's likely that, even if Constantinople was part of a larger whole, the OE still would have moved their capital right away to secure the conquest.

Now, this situation is less common, and less of a problem, than the first issue I noted. I could accept this event being left out. But I do maintain that it is pretty realistic to include it, and I don't see a non-event-based solution that addresses the problem in a more elegant way.

 

[Norrefeldt]

This discussion essentially breaks down to what a lot of discussions have been about. What kind of subtractive events (in Twoflowers words) should we use. It's the same discussion as Independence events, Portugal/Spanish events regarding Spanish ownership of Portuguese provinces etc. We are likely to get into this over and over again. It ought to be solved once and for all IMO.
I hardly see a principal difference in just ceding Thrace straight away 1453. It's all about creating ahistorical events, possibly AI only, to get a historical outcome.

 

[doktarr]

This discussion essentially breaks down to what a lot of discussions have been about. What kind of subtractive events (in Twoflowers words) should we use. It's the same discussion as Independence events, Portugal/Spanish events regarding Spanish ownership of Portuguese provinces etc. We are likely to get into this over and over again. It ought to be solved once and for all IMO.

I don't think that's possible - or, more precisely, I don't support a policy that can be boiled down to "yes" or "no" on these issues.

A good idea would be to come up with five or six "talking points" about these sorts of events. This would help give us some traction in the discussion about the usefulness or ahistoricity or realism of these events.

I hardly see a principal difference in just ceding Thrace straight away 1453. It's all about creating ahistorical events, possibly AI only, to get a historical outcome.

If you see no distinction there, then I completely understand why you oppose the event. To me, there is an enormous difference.

------------------------------

Thought experiment. Consider the following approaches:

1. No events for OE/Byzantium; just a core and middling relations.
2. Something like the current sequence that just breaks vassalage at a certain time and leaves the rest to the engine.
3. Add an event "Mehmed's ambition". "A" choice activates an AI that focusses on Thrace.
4. Same as above, but now it is the only choice.
5. Same as one of the two above, but adds a significant hit to relations with the owner of Thrace.
6. Same as above, but gives an allied/friendly Thrace owner a "b" option to simply cede the province to the Ottomans in exchange for BB reduction and improved relations.
7. Same as above, but with ceding the province becoming the "a" choice for a friendly nation.
8. Same as one of the two above, but the non-ceding option now sends you to war.
9. Same as one of the three above, but only a second, later version forces war or secession.
10. Same as one of the four above, but any nation that loses Thrace to the Ottomans gets an event where the "b" option simply cedes the province immediately.
11. Same as above, but now this is the "a" option.
12. Same as above, but now this is the only option.
13. Give as one of the above, but give even enemy nations the "a" option to cede Thrace to the Ottomans without conflict.
14. Simply cede Thrace to the Ottomans by event if they are strong but have not taken it by a certain date.
15. Cede Thrace to the Ottomans in 1453.

Now, honestly, is there really no difference between those options? In my opinion, you can create some historical backing for everything through number 12. Some folks might only support approaches through 11. Some folks might draw the line at 6 or 7. Others might only support 2 or 3. Others may be fine with 15! The point is, I think there's a true continuum here. Different "talking points" might support drawing the line between "acceptable" and "forced" at different spots.

 

[unmerged(31425)]

Check post 133 in the submission thread. It is basically just that. And as you know, I support the vassalization event as well.

That's an AI cheat like the vassalization one alright. Where was Twoflower when you proposed that? I look forward to seeing him be consistent and hooting and hollering and forcing it to a HC vote. Youshould expect a vicious fight. He gave me one. Good luck! For your sake I hope it passes.

By the way I think the event I proposed would be more efficient and less awkward. But I probably won't post my scripting so you may want to think it over a bit.

 

[Castellon]

That's an AI cheat like the vassalization one alright. Where was Twoflower when you proposed that? I look forward to seeing him be consistent and hooting and hollering and forcing it to a HC vote. Youshould expect a vicious fight. He gave me one. Good luck! For your sake I hope it passes.

By the way I think the event I proposed would be more efficient and less awkward. But I probably won't post my scripting so you may want to think it over a bit.

I suggest you read my comments in the Iberian thread before posting again!

 

[unmerged(31425)]

This would be a good thing to put in an "anti-Byzantine" AI. Writing such an AI would be a good step towards solving this problem in an less invasive fashion. But such an AI would interfere with other Ottoman expansion, so its use would need to be restricted to a fairly narrow time range IMO.

I think you guys are placing too much hope in AI files. they can make difference in alot of odds and ends areas, but there are alot of problems that they can only diminish slighlty. This is why Daywalker uses AI cheats. I'm not saying this to be hostile or quarrelsome. Besides if you guys really commit yourselves to experimenting and playtesing with AI files alot, I think what I'm saying here will become more obvious.

I understood the point and don't have a problem with it - I just don't think it will make a significant difference. When the Ottomans decide to join the siege they are usually the first ones there. The problem is when they are not part of the war or don't bother to send troops there.

Exactly. Granted Norrefldt ideas here will probably make a difference, but really it's just reducing the problem slighly not removing it or reducing it to a rarity.

I agree that we have not explored every other option.

Ya but Daywalker has playtested alot more than any of us and he apparently came to the conclusion sometime ago that AI cheats are unavoidably neccessary. That ought to mean something.

The biggest change I want to make to our current approach to the large nations is to have more AI switching. But rather than trigger switching by date (as Daywalker does) or monarch (as PAI does) I would suggest switching based on situations. This makes the AI think more like a person.

This might actually help in some situations but man you're opening up a can of worms here. One thing I've realized from playtesting is that there are just a ton of variables and you don't always think of them ahead of time even if you're going out of your way to try to. You got to figure just what could go wrong with this kind of a switch.

So the Ottomans would start by going after the Turkish, Greek, and Balkan minors. Then provided they have achieved a modicum of success, they go after Constantinople. Then they expand in Europe, then they expand in the Levant and Egypt, then they try to grab the whole Eastern Mediterranean.

Doesn't it already do something like this?

That said, much of what I'm suggesting is not the "sledgehammer" style event. My proposal can be considered in three seperate parts:

1) I make allies of an AI OE either give Thrace to the Ottomans, or break their alliance and defend it. I think this is the most realistic/important of the three. I should probably add some reduced badboy and improved relations to it, but it is this situation (an ally or friend holding Thrace) that is currently the MOST unrealistic and the MOST likely to cause problems.

Breaking alliances can cause other countries to DOW the Ottomans and this would be just after a war and with a weakened alliance. This would probably back-fire often and wind up screwing the Ottomans.

2) I force enemies of an AI OE to go to war with the OE if they hold Thrace past a certain date.

This could also cause the problem I mentioned above. AIs factor in whether or not a country is at war when deciding to DOW. That combined with the all of a sudden reduction in the economic strength of the Ottoman alliance could cause all hell to break loose. They got alot of neighbors who hate them. Lots of potential for things to go haywire here.

This is a bit of a hack, I'll admit. It would perhaps be better to just have the event cut relations between OE and the Thraceowner, and/or trigger the antibyzantine AI. It needs to be tested. It would be nice if a "softer" approach could work here.

Man one thing Ican tell you about forced DOWs for the AI. They can really really screw him up bad. That's why when Ido this I make other additions to the event, like extra transports, increase in stab(could be breaking a peace treaty or just plain have low stab), and negative RR. I also make a ton of specific triggers and believe me when running this over and over it's very easy to wind up finding yourself needing to make the triggers even more specific.

3) I force an immediate secession of Thrace if the OE controls it. Of the three, this is the most "sledgehammeresqe". And yet, I think it is pretty realistic.

Younkow where I stand on this. Why would some chump ally be allowed by the Ottomans to get Constantinople instead ofthem. What the engine does here is just plain unrealsitic. But eh alot of the time the ally won't invite the Ottomans. I don't see whythey wouldn't have just invited themeselves though.

 

[unmerged(31425)]

I suggest you read my comments in the Iberian thread before posting again!

OK Iread it. No problems here.

 

[doktarr]

The biggest change I want to make to our current approach to the large nations is to have more AI switching. But rather than trigger switching by date (as Daywalker does) or monarch (as PAI does) I would suggest switching based on situations. This makes the AI think more like a person.

This might actually help in some situations but man you're opening up a can of worms here. One thing I've realized from playtesting is that there are just a ton of variables and you don't always think of them ahead of time even if you're going out of your way to try to. You got to figure just what could go wrong with this kind of a switch.

I agree. The triggers would have to be quite complicated. And there would have to be a date beyond which the AI switches to the next goal, whether or not the previous one has been met. The OE needs to be able to say "oh well, Europe just ain't happening this century. Let's go get Egypt."

I'm not saying this would be easy, or even necessary in all cases. But I think it would be useful in certain cases, and the OE is probably the clearest example.

 

[Norrefeldt]

I don't think that's possible - or, more precisely, I don't support a policy that can be boiled down to "yes" or "no" on these issues.

A good idea would be to come up with five or six "talking points" about these sorts of events. This would help give us some traction in the discussion about the usefulness or ahistoricity or realism of these events.
If you see no distinction there, then I completely understand why you oppose the event. To me, there is an enormous difference.

A clear positive or negative stance is probably not possible as you said.

Thought experiment. Consider the following approaches:

1. No events for OE/Byzantium; just a core and middling relations.
2. Something like the current sequence that just breaks vassalage at a certain time and leaves the rest to the engine.
3. Add an event "Mehmed's ambition". "A" choice activates an AI that focusses on Thrace.
4. Same as above, but now it is the only choice.
5. Same as one of the two above, but adds a significant hit to relations with the owner of Thrace.
6. Same as above, but gives an allied/friendly Thrace owner a "b" option to simply cede the province to the Ottomans in exchange for BB reduction and improved relations.
7. Same as above, but with ceding the province becoming the "a" choice for a friendly nation.
8. Same as one of the two above, but the non-ceding option now sends you to war.
9. Same as one of the three above, but only a second, later version forces war or secession.
10. Same as one of the four above, but any nation that loses Thrace to the Ottomans gets an event where the "b" option simply cedes the province immediately.
11. Same as above, but now this is the "a" option.
12. Same as above, but now this is the only option.
13. Give as one of the above, but give even enemy nations the "a" option to cede Thrace to the Ottomans without conflict.
14. Simply cede Thrace to the Ottomans by event if they are strong but have not taken it by a certain date.
15. Cede Thrace to the Ottomans in 1453.

Now, honestly, is there really no difference between those options? In my opinion, you can create some historical backing for everything through number 12. Some folks might only support approaches through 11. Some folks might draw the line at 6 or 7. Others might only support 2 or 3. Others may be fine with 15! The point is, I think there's a true continuum here. Different "talking points" might support drawing the line between "acceptable" and "forced" at different spots.

OK, I'll give my opinion on those. 1 is fine of course, 2 is also OK, provided there is a historical justification for it. I think AI files can be just anything, so both 3 and 4 are good ideas. However it would be hard to do if we don't know the owner of Thrace. 5 is more questionable, unless there is historical background. This is more grey to me, events for AGCEEP should be historical. That's a hard rule but I think in the majority of the cases we can do much better scripting and testing, or just set everything straight through events, which I don't like. In this particular case the relation hit would be justified. After this I think we are only making history up, don't let the game evolve in a "natural" way and arbitrarily trying to correct things that happen in the game. Admittedly there is a difference in how we do it, and if one think this kind of solutions is OK I can image that some are better than others. We can tell this invented straightjacketed story in different ways, but the difference isn't that big and we aim for exactly the same outcome.

This kind of events will be very common if we want to make the majors succeed in 99% of the cases. I wouldn't like to play, or mod, anyhting where there are tens of "setting straight" events around, that set invisible boundaries for me and for what can happen.
I think we should aim for the highest possible outcome of some important historical events, without setting the history that "went wrong" right afterwards, as I think these events are doing.

 

[unmerged(31425)]

I agree. The triggers would have to be quite complicated. And there would have to be a date beyond which the AI switches to the next goal, whether or not the previous one has been met. The OE needs to be able to say "oh well, Europe just ain't happening this century. Let's go get Egypt."

I'm not saying this would be easy, or even necessary in all cases. But I think it would be useful in certain cases, and the OE is probably the clearest example.

Well Doktarr, I technically don't disagree with this. I'm certainly one who likes lots of AI switching. I have additional AIs for Portugal and Spain, which are the two countries I'm working on so far with my version.

My point is that based on my experience with modding AIs is that it is extremely difficult to factor in all the variables in AI switching and despite the fact the you may be able to grasp alot of them ahead of time, being somone who's played and modded alot, I think you'll find it's just too easy to miss things. So if you're gona go this route I strongly reccommend that this is one of those areas that will really need alot of playtesting.

Sorry if I'm the barer of bad news. I'm actually just trying to be kind.

 

[unmerged(31425)]

This kind of events will be very common if we want to make the majors succeed in 99% of the cases. I wouldn't like to play, or mod, anyhting where there are tens of "setting straight" events around, that set invisible boundaries for me and for what can happen.

But this kind of event if made into a 2 or 3 event, event set, doesn't need to have any invisible boundaries for human players. They can be made to trigger for 2 AIs only. This is how I do it in my version. I certainly don't like invisible boundaries for me either. But I don't mind them for AIs if those AIs are having difficulty reaching historical objectives with a relelvant degree of consistency. Anotherwards things like the Ottmans being severely stunted from growing big like they are supposed to too often.

If I'm not mistaken this is Doktarr's view. Correct me if I'm wrong Dokatarr.

So saying there would be an "invisible boundary" for you isn't what is really at stake here. Where the other problem comes in with what Doktarr is trying to get in the mod, is that these kinds of events are technically "AI cheats." They help an AI accomplish something and we don't get the same special priveledges. And if this is allowed than just think of what other kinds of AI cheats could be allowed too. Making AIs do things that we can accomplish pretty regularly.

I think we should aim for the highest possible outcome of some important historical events, without setting the history that "went wrong" right afterwards, as I think these events are doing.

Well if this is your final view I think it's safe to say that unless it gets over ruled we are all gonna have to accept the fact that there are areas in the game that just can't be made to develope historically with a relevant degree of consistency. If this is what you want that's fine, it's your mod, but understand that the objective that Doktarr, myself, and others have in mind simply can't be reached any other way.

 

[Norrefeldt]

Where the other problem comes in with what Doktarr is trying to get in the mod, is that these kinds of events are technically "AI cheats." They help an AI accomplish something and we don't get the same special priveledges. And if this is allowed than just think of what other kinds of AI cheats could be allowed too. Making AIs do things that we can accomplish pretty regularly.
Well if this is your final view I think it's safe to say that unless it gets over ruled we are all gonna have to accept the fact that there are areas in the game that just can't be made to develope historically with a relevant degree of consistency. If this is what you want that's fine, it's your mod, but understand that the objective that Doktarr, myself, and others have in mind simply can't be reached any other way.

As you say including this will have far-reaching consequences. We cannot allow these events and not your Portuguese events, nor stopping the proposed independence events. It's more or less the same kind of scripting. Unfortunately it's just the three of us discussing this.
It's not my mod and if it's decided that this is the way to do AGCEEP I won't be in the way.

 

[WiSK]

Unfortunately it's just the three of us discussing this.

But others are reading what you are saying.

However, since Twoflower was the one to initially define 'subtractive events', I'd like to read his opinion.

 

[unmerged(31425)]

As you say including this will have far-reaching consequences. We cannot allow these events and not your Portuguese events, nor stopping the proposed independence events. It's more or less the same kind of scripting. Unfortunately it's just the three of us discussing this.
It's not my mod and if it's decided that this is the way to do AGCEEP I won't be in the way.

I see. I actually meant your in plural(you being part of that plural). I know at least Twoflower has the same view. If it's not to much for me to ask, I was wondering if you guys actually voted on anything like this yet and if so what were the results?

I wasn't certain if you grasped all the far reaching consequences of not allowing these kinds of events in the mod or not. Obviously in our previous debate, only the event I was proposing was being considered. But now that I see you expressing your view in black and white and I've seen Twoflower do this already, I certainly have no need of arguing about just how such a stance will effect the mod. Thanks for clearing this up for me. We certainly are both entitled to our points of view.

 

[unmerged(31425)]

But others are reading what you are saying.

However, since Twoflower was the one to initially define 'subtractive events', I'd like to read his opinion.

His opinion is virtually identical to the one Norrefeldt expresses here. He gives his view in the independence events thread and the Portuguese AI thread. I can remember off handwhich posts.

 

[WiSK]

His opinion is virtually identical to the one Norrefeldt expresses here. He gives his view in the independence events thread and the Portuguese AI thread. I can remember off handwhich posts.

Yes, I'm aware Twoflower and Norreveldt both dislike subtractive events. However, I seem to remember, in the case of the secession of Porto, he argued that you weren't looking hard enough for another solution. I'm wondering if the example of Thrace might make him change his mind. Not entirely changed, but to accept that there might be situations which call for exceptions to his rule.