Print

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

[Jeebus]

Thanks fore the discussion everyone!

I have another scenario that's been on my mind that seems related to this: the gain effects of Gatekeeper and the Exile mat are opposite each other.

So far I lean toward preferring the rule "any new thing that gets added will also trigger" but doesn't that lead to a weird thing between these two effects?: A gain happens but only one can trigger. But as soon as you apply that effect, the other one triggers and essentially undoes the other's effect.

I see that Gatekeep has the FAQ answer: "Gaining a card that you do have a copy of on your Exile mat is unaffected, and lets you discard copies from your Exile mat if you want to as usual." But doesn't "any new thing that gets added will also trigger" contradict that FAQ answer?

Good catch. The reason is that the trigger itself is not retroactive. Let's say you gain a Silver when you already have a Silver in Exile. For any when-gain abilities what you gained was "a Silver that you had an Exiled copy of [and which had a certain cost and certain types etc.]". That defines the trigger window. Only abilities with matching triggers can trigger in this window, no matter if the Exiled Silver is discarded.

Another example: You have played Livery, and then gain a Fisherman costing $2. When you resolve when-gain abilities for this gain, the Fisherman costs $5. But what you gained was "a Fisherman costing $2 [and that had certain types etc.]", and only abilities that trigger off of that will trigger. So you don't gain a Horse from Livery.

EDIT: I mistakenly wrote $3 instead of $2, and "do gain" instead of "don't gain".

[Jeebus]

No, this is my interpretation, that I came up with, looking at the different rulings presented here. Nothing that Donald X. said, just trying to make sense of for myself.

I see. You did write that.

The enabler here was Innovation, enabling a card to be played while still in it's "when-gain" window. The Livery-Innovation example is possibly what made the previous ruling paradoxical in the eyes of players: The Livery is played and sets up a "for the rest of this turn" trigger condition, that (under the rules as communicated before) is now applied to a time when that effect clearly couldn't have existed: when that Livery had just been gained.

That does make a compelling argument to reverse the ruling in that case. How that generalizes, is again my idea is presented above.

I agree that Innovation opened up a lot of possibilities that might seem counterintuitive to players. But in my mind, this already happened with "while in play" abilities like Goons and Haggler entering play after the "triggering event" (the buy in this case). Duplicate and Moat are more intuitive because they don't spell out a time frame for when the "triggering event" needs to happen. Haggler says that the "triggering event" needs to happen while the Haggler is in play, which it actually didn't. The same goes for Livery. It's possible to interpret Haggler and Livery such that they work like Duplicate and Moat, but it's not intuitive for many (probably most) players. But again, if Livery doesn't work like that, neither should Haggler.

In this thread, two years ago, we were both actually saying that it would be better if the Haggler scenarios did not work.

[DaveS]

Good catch. The reason is that the trigger itself is not retroactive. Let's say you gain a Silver when you already have a Silver in Exile. For any when-gain abilities what you gained was "a Silver that you had an Exiled copy of [and which had a certain cost and certain types etc.]". That defines the trigger window. Only abilities with matching triggers can trigger in this window, no matter if the Exiled Silver is discarded.

Another example: You have played Livery, and then gain a Fisherman costing $3. When you resolve when-gain abilities for this gain, the Fisherman costs $5. But what you gained was "a Fisherman costing $3 [and that had certain types etc.]", and only abilities that trigger off of that will trigger. So you do gain a Horse from Livery.

Hmmm, gonna have to chew on that idea a bit.  The light bulb has not yet turned on for me Also, in your Livery scenario, I think you mean $2 and "not gain a Horse"?

[Jeebus]

Hmmm, gonna have to chew on that idea a bit.  The light bulb has not yet turned on for me Also, in your Livery scenario, I think you mean $2 and "not gain a Horse"?

You're right! Thanks, I fixed it.

[DaveS]

The reason is that the trigger itself is not retroactive.

So regarding how to process effects, is this algorithm correct:
When a gain occurs, maintain a collection of effects to be resolved (2BR) and a collection of resolved effects (RX).
1. At the moment of the gain, find all gain effects in the mix that are not yet in either collection and add them to 2BR. (A Sheepdog near the top of the deck is NOT YET in the mix, but both an active Gatekeeper and the Exile mat always are).
2. For each effect in 2BR: if its trigger is not enabled, add it to RX; (it won't ever be re-evaluated for THIS gain). This is how we avoid both Gatekeeper and the Exile mat both from ever both enabling.
3. An effect in 2BR is chosen and executed. Add it to RX.
4. Now, we check for any NEW effects in the mix (like a drawn Sheepdog). Rinse and repeat from step 2.

[Jeebus]

The reason is that the trigger itself is not retroactive.

So regarding how to process effects, is this algorithm correct:
When a gain occurs, maintain a collection of effects to be resolved (2BR) and a collection of resolved effects (RX).
1. At the moment of the gain, find all gain effects in the mix that are not yet in either collection and add them to 2BR. (A Sheepdog near the top of the deck is NOT YET in the mix, but both an active Gatekeeper and the Exile mat always are).
2. For each effect in 2BR: if its trigger is not enabled, add it to RX; (it won't ever be re-evaluated for THIS gain). This is how we avoid both Gatekeeper and the Exile mat both from ever both enabling.
3. An effect in 2BR is chosen and executed. Add it to RX.
4. Now, we check for any NEW effects in the mix (like a drawn Sheepdog). Rinse and repeat from step 2.

I don't think what you're describing covers it.

1. At the moment of gain, define what was gained, for instance "a Silver with type Treasure and cost $3 that you have an Exiled copy of".
2. Find all when-gain abilities that have a matching trigger and put them in 2BR.
3. An ability is chosen and resolved, and moved to RX.
4. We check for any new ability: go to step 2.

So if the Exile mat is resolved in step 3, Gatekeeper still won't be added, because its trigger doesn't match what was defined in step 1.

But if a Duplicate is gained, it's a card costing $4 (which is "up to $6"), so if you resolve Innovation's when-gain ability, playing the Duplicate (step 3), then when you go back to step 2, you find that Duplicate's own when-gain ability has a matching trigger, so it is added to 2BR: You may call Duplicate to gain a copy of the Duplicate.

[Jeebus]

Donald, I wonder if you could respond in this thread?

You stated on Discord back in February last year that Livery/Innovation should work like Haggler/Innovation, and I agree. You didn't make a public decision about Livery then, but in April you had made the decision that Livery works differently than Haggler, but I don't see that you gave a good explanation.

I remind you that you previously ruled that Silver/Merchant/Innovation works - which is equivalent to Livery/Innovation.

Haggler's when-buy ability, like Livery's when-gain, is limited by timing. Haggler only works from the moment it's in play, Livery from the moment it's played. In the Innovation scenarios, the gain/buy happened before Livery was played and before Haggler entered play. It's fine to rule that Livery/Innovation doesn't work, but then Haggler/Innovation should be reversed too.

And if Livery/Innovation does not work, neither can Hireling played at start of turn.* (The same goes for Prince, Lost in the Woods and Key played/taken at start of turn.)

A good model, as described by Ingix, is that all abilities trigger when something happens, and then no more abilites trigger during the window: You gain a card - all 10 Sheepdogs and all 10 Duplicates in the game trigger, but only those Groundskeepers that you have in play, and only those Liveries that were already played. If a Moat finds its way to your hand in the window, or a Duplicate to your Tavern mat, you can resolve them - they already triggered. But a Groundskeeper that enters play in the window didn't trigger. The same goes for Liveries that get played in the window. (When buying a card, only those Hagglers that you have in play trigger.) At start of turn, all 10 Ratcatchers trigger, but only those Hirelings that have been played.

The other way is that abilities can trigger during the window - Hagglers, Hirelings and Liveries...

---
* "Start of turn" is a moment in time (not a phase), just like "when you gain a card". It happens right before you start playing Action cards. All abilities trigger at the same time, and then you choose which to resolve first, etc. In both cases, new abilities can be added during the window. I absolutely don't see how you can say that these two windows are different.

[Donald X.]

Donald, I wonder if you could respond in this thread?

I'm there for you.

You stated on Discord back in February last year that Livery/Innovation should work like Haggler/Innovation, and I agree. You didn't make a public decision about Livery then, but in April you had made the decision that Livery works differently than Haggler, but I don't see that you gave a good explanation.

I remind you that you previously ruled that Silver/Merchant/Innovation works - which is equivalent to Livery/Innovation.

Haggler's when-buy ability, like Livery's when-gain, is limited by timing. Haggler only works from the moment it's in play, Livery from the moment it's played. In the Innovation scenarios, the gain/buy happened before Livery was played and before Haggler entered play. It's fine to rule that Livery/Innovation doesn't work, but then Haggler/Innovation should be reversed too.

And if Livery/Innovation does not work, neither can Hireling played at start of turn.* (The same goes for Prince, Lost in the Woods and Key played/taken at start of turn.)

A good model, as described by Ingix, is that all abilities trigger when something happens, and then no more abilites trigger during the window: You gain a card - all 10 Sheepdogs and all 10 Duplicates in the game trigger, but only those Groundskeepers that you have in play, and only those Liveries that were already played. If a Moat finds its way to your hand in the window, or a Duplicate to your Tavern mat, you can resolve them - they already triggered. But a Groundskeeper that enters play in the window didn't trigger. The same goes for Liveries that get played in the window. (When buying a card, only those Hagglers that you have in play trigger.) At start of turn, all 10 Ratcatchers trigger, but only those Hirelings that have been played.

The other way is that abilities can trigger during the window - Hagglers, Hirelings and Liveries...

---
* "Start of turn" is a moment in time (not a phase), just like "when you gain a card". It happens right before you start playing Action cards. All abilities trigger at the same time, and then you choose which to resolve first, etc. In both cases, new abilities can be added during the window. I absolutely don't see how you can say that these two windows are different.

The first thing is, that I should ignore Haggler for now, focus just on Livery. The best solution to "Haggler should work the same as Livery" ultimately is rewording Haggler to match Livery. I have come around to the idea of doing that, when next it's convenient. That doesn't mean that current-wording Haggler somehow doesn't require rulings; but, I only have so much time, I am saving time by not considering those rulings.

And next: what's most important, by far, is having simple rules to explain to people that cover the actual common cases they see in games. For example "this is how Livery works" is vastly more important than "this is how Livery plus Innovation works." Any ruling will ignore the result, which I don't care about, and just focus on, what falls out of the best explanation for normal people about the most common situations.

By that metric I can easily rule out Ingix's model. I would feel like a madman trying to explain that to people.

This question always comes back to Moat / Secret Chamber. I think people in general find it weird and confusing that you can trigger Sheepdog and draw a Sheepdog and also play that Sheepdog. But it's a weird confusing thing with a lot of weight behind it in the history of Dominion. There could be separate rules for hidden triggers and non-hidden ones; it seems simpler to have them all work the same, though that may not be true.

My current ruling is that Livery / Innovation gets you a Horse.

[Jeebus]

My current ruling is that Livery / Innovation gets you a Horse.

Thanks for your reply. So just to clarify, you're changing the ruling you gave back in April? It's not so clear from the rest of your answer, so that's why I'm double-checking.

I support this ruling, because then the rules are consistent with Haggler/Innovation, and Hireling at start of turn, as it currently stands. But I still want to respond to what you wrote about Ingix's model.

And next: what's most important, by far, is having simple rules to explain to people that cover the actual common cases they see in games. For example "this is how Livery works" is vastly more important than "this is how Livery plus Innovation works." Any ruling will ignore the result, which I don't care about, and just focus on, what falls out of the best explanation for normal people about the most common situations.

By that metric I can easily rule out Ingix's model. I would feel like a madman trying to explain that to people.

I don't think the model says anything about "this is how Livery works" (without Innovation).

Consider that this model already has support in Dominion. We know that cards trigger wherever they are: Hunting Grounds in the supply trigger when you trash them with Lurker. So a Sheepdog triggering before you draw it is actually how Dominion already works. But more importantly, if you wanted to rule that things that haven't been activated yet don't retroactively work, like with Livery/Innovation, this would be exactly what this model explains. (It would also mean that Haggler, Hireling etc. work differently in these special situations.) In short, if you actually needed to explain why Sheepdog can be played but you don't get a Horse from Livery, basing it on this model would be less complicated than anything else I've seen.

This question always comes back to Moat / Secret Chamber. I think people in general find it weird and confusing that you can trigger Sheepdog and draw a Sheepdog and also play that Sheepdog. But it's a weird confusing thing with a lot of weight behind it in the history of Dominion. There could be separate rules for hidden triggers and non-hidden ones; it seems simpler to have them all work the same, though that may not be true.

Then I would go for separate rules for Reactions.

[Donald X.]

Thanks for your reply. So just to clarify, you're changing the ruling you gave back in April? It's not so clear from the rest of your answer, so that's why I'm double-checking.

Time does not permit going back and looking at old rulings.

I don't think the model says anything about "this is how Livery works" (without Innovation).

I can't further consider Ingix's model at this time. Ingix is super helpful and is improving the game and that particular idea I could immediately reject.

Then I would go for separate rules for Reactions.

I try not to tie myself to bad precedents, but it seems late for that. I'm not sure it would be as simple as that either (what's distinct here is that the triggering card was hidden, not that it was a Reaction). And then, if Reactions work differently, well, that still leaves you playing all of your Sheepdogs. It can't be e.g. you put all the Reactions on the table and then resolve them one at a time (which fixes Sheepdog confusion); that couldn't fit on the card, it would have to be tied into e.g. a card type, and then the card type Reaction couldn't mean that because Reactions aren't all like that, they are different things. So then I'm making a new type to tie these rules to it (or a keyword unexplained in the text), or going to microtext. If it were day one then of course that could be what Reactions were, and they would all work that way.

[GendoIkari]

My one issue with a different rule for reactions is that it’s pretty arbitrary whether or not something is a reaction. Fortress certainly could have been. As I understood it from years ago, the idea was it was a reaction if it triggered from a hidden place. But now we have Patron. I believe the rule is that it’s a reaction if Donald feels like it’s a good idea to help people pay attention to it at abnormal times by giving it a different color. Which, is basically pretty arbitrary. I like that reactions, like attacks, are just a helpful indicator and not something with specific rules attached.

[Jeebus]

Then I would go for separate rules for Reactions.

I try not to tie myself to bad precedents, but it seems late for that. I'm not sure it would be as simple as that either (what's distinct here is that the triggering card was hidden, not that it was a Reaction). And then, if Reactions work differently, well, that still leaves you playing all of your Sheepdogs. It can't be e.g. you put all the Reactions on the table and then resolve them one at a time (which fixes Sheepdog confusion); that couldn't fit on the card, it would have to be tied into e.g. a card type, and then the card type Reaction couldn't mean that because Reactions aren't all like that, they are different things. So then I'm making a new type to tie these rules to it (or a keyword unexplained in the text), or going to microtext. If it were day one then of course that could be what Reactions were, and they would all work that way.

When you mentioned separate rules for hidden triggers and non-hidden ones, I thought you meant keeping Moat / Secret Chamber and Sheepdog as is. Saying that the current "window rule" only applies to Reactions would solve that, I think. Faithful Hound, Patron, Tunnel and Village Green would be the only ones that are not hidden, but I don't think the "window rule" could ever apply to them anyway, since they only trigger on themselves.

That would eliminate all Innovation related scenarios like this. But, since the Reactions Sheepdog, Falconer and Black Cat allow any card to be played on "when-gain" (via Way of the Mouse), you could still in rare cases get Haggler etc. entering play.

My one issue with a different rule for reactions is that it’s pretty arbitrary whether or not something is a reaction. Fortress certainly could have been. As I understood it from years ago, the idea was it was a reaction if it triggered from a hidden place. But now we have Patron. I believe the rule is that it’s a reaction if Donald feels like it’s a good idea to help people pay attention to it at abnormal times by giving it a different color. Which, is basically pretty arbitrary. I like that reactions, like attacks, are just a helpful indicator and not something with specific rules attached.

There are rules for Reactions. You can resolve a Reaction in your hand several times. And the newish rule that the players can keep resolving Reactions as long as someone does something.

[GendoIkari]

There are rules for Reactions. You can resolve a Reaction in your hand several times. And the newish rule that the players can keep resolving Reactions as long as someone does something.

But these rules deal more with hidden knowledge than with the reaction type. The official rule may be pertaining to only reactions, but it also only pertains to reactions in your hand, and it's there because you can't prove whether you have a second Secret Chamber or just the same one twice. With cards that aren't in your hand, like Patron vs Fortress, a rule that made one act differently than the other just because one happens to be a reaction would be putting more weight on the fact that one happens to be a reaction than currently exists. Right now it's a little inconsistent that one is a reaction while the other is not, but it's an irrelevant inconsistency that has no rules implications.

[Jeebus]

The rules we're talking about only concern Reactions. It's easier and better to refer to "Reactions" or "Reactions in your hand" than "cards in your hand that do something based on something else happening" or similar.

And I'm pretty sure that Patron being included if we only say "Reactions" also has no rules implications.

[GendoIkari]

The rules we're talking about only concern Reactions. It's easier and better to refer to "Reactions" or "Reactions in your hand" than "cards in your hand that do something based on something else happening" or similar.

And I'm pretty sure that Patron being included if we only say "Reactions" also has no rules implications.

Sure, I didn't mean to suggest that the current reactions rules should be re-worded to avoid making them reaction-specific. I'm just saying that for a potential new rule that deals with the scope of triggering windows as it related to things like Livery, Priest, Way of the Seal, etc; should probably be related to the fact that those have on-play abilities that don't exist until the card is played. Whereas Sheepdog and Fortress just have things that are always active and available.

In other words, if there's a rule that makes Sheepdog different than Livery, then that exact same rule would and should make Fortress act like Sheepdog as well. So it can't be Sheepdog being a reaction that makes Sheepdog different than Livery.

[Jeebus]

Sure, I didn't mean to suggest that the current reactions rules should be re-worded to avoid making them reaction-specific. I'm just saying that for a potential new rule that deals with the scope of triggering windows as it related to things like Livery, Priest, Way of the Seal, etc; should probably be related to the fact that those have on-play abilities that don't exist until the card is played. Whereas Sheepdog and Fortress just have things that are always active and available.

In other words, if there's a rule that makes Sheepdog different than Livery, then that exact same rule would and should make Fortress act like Sheepdog as well. So it can't be Sheepdog being a reaction that makes Sheepdog different than Livery.

Then I think you misunderstood the part of my post that you responded to. I was just replying to Donald's comment about "separate rules for hidden triggers and non-hidden ones", thinking that he meant keeping the "window" rule for all hidden triggers (meaning all Reactions in your hand) - as it has been since Moat / Secret Chamber - but removing the rule for EVERYTHING non-hidden. So it was not about how to separate Livery from Sheepdog and Fortress. And to be clear, this was not something he said he would do now. It was a side-comment to the main topic.