Blog Archives

Still Skeptical of Storybricks

This past weekend Nils and I were walked through the Storybrick demo with Kelly from Namaste in a sort of dual presentation. It was illuminating in a lot of different ways.

A) Storybricks is actually two seperate things: the AI and then the game.

To understand how big a difference this is, think about the difference between the Havok engine and Assassin’s Creed, Bioshock, Fallout 3, Half Life 2, LA Noir, Portal 2, and the Witcher 2. Havok, by itself, is not anything – it is not a game, it is a tool to make games. Storybricks is NOT entirely like the Havok engine though, as it would essentially be a tool within a tool, Inception-style.

This is an important distinction because the build-it-yourself AI part of it is actually pretty interesting. It reminds me of a post by Notch of Minecraft fame where he made what seemed like a mild change (animals flee when attacked) and then discovered an “emergent” AI behavior when suddenly sheep would realistically flee from wolves. Even though I pooh-poohed the thought of unintended consequences in my last Storybrick post, that was more in a story aspect. I am otherwise a big fan of tweaking small variables in a complex system and watching how the dominos fall in unpredictable ways – fans of playing/theorizing the WoW AH pretty much have to be, by definition.

B) The Storybricks GAME doesn’t exist.

This is a particular point I feel was not emphasized enough in other Storybrick write-ups. There is no game, at the moment. Kelly described how they do not want to start building a house, and then halfway through realize they should have been using a hammer instead of a rock to nail everything together. That… sort of makes sense. I mean, I am not a game designer or anything, so I do not know what the “proper” way of game development consists of.

What concerns me, and I got this impression from Nils as well, is that it is always easy to come up with ideas that should be fun; actually having them be fun is another matter entirely. As I mentioned in the previous section, I like the idea of subtle changes making surprising differences down the road. I also like the idea of a Silent Hill meets Fallout 3 style game. There is nothing to really say that either will be fun in practice however.

C) The Storybricks game won’t resemble any MMO.

Making matters worse, as Kelly explained, is that you cannot really even begin to explain a Storybrick game. This is not WoW + NPCs that remembers you. This Storybrick game would not have combat, no NPCs could die, and they are debating right now whether leveling should even exist. The socializing and NPC relationships would be the game. Which confuses me when she explained how all these D&D DMs were excited about it: what exactly are they imagining here? Their campaigns “coming to life?” What is D&D without dice rolls, saving throws, combat at all? It is a purer form of collaborative storytelling, sure. But it is no longer D&D. And they (or their players) probably signed up to play D&D.

So instead of imagining your favorite game with more compelling NPCs, imagine your favorite game without combat, without gear progression, without levels, without a set narrative at all. Then you can start imagining Storybricks and its emergent interactions. The closest thing I can imagine is Myst, except people are the puzzles.


Now that I have kind of outlined the experience, I am going to more briefly list my concerns about the project. And because this whole post is a lot more negative and antagonistic than I necessarily feel, I have a bonus section for the Namaste folk at the end. So my concerns are:

1) I do not see Storybricks working in an MMO setting at all. One of the examples that came up in the demo was how a player might need a key from the guard to unlock a door. The guard might just straight-up give the key to Bob because of their prior relationship, whereas Tom might be forced to run some errand for the guard. The issue is that if Bob and Tom are friends/grouped up, would Bob not be able to simply give Tom the key to use? If that is the case, could you not imagine Bob hanging out in the newbie area and offering to escort new players straight past the “guard gate” (assuming there is some medium of exchange possible)?

2) Related to 1), I have never felt it particularly compelling in D&D when the guy with the highest Charisma or Speech skill is the de facto NPC ambassador. This is not to say specialized roles are not compelling, simply that if Bob is better at making the Storybrick connections (aka making the dice rolls), Bob will basically do them for the group. At most, I have a connection to Bob or the group as a unit, not to the NPCs. In a single-player game on the other hand, each player will have 100% connection with the NPCs.

3) It may be beyond the scope of the tool, but I do very much hope that Storybricks would have some kind of mode or setting that would allow the players designing the modules to be able to direct a player’s experience in a more traditional way. In other words, let people go nuts in the village with NPC interaction, while still having a overarching narrative in the castle.

4) I think serious consideration should be given right now as to what the object of a Storybrick game would be, e.g. the fun part. Sandboxes are all about player-driven goals, of course, but the sand has to be fun to play with. Speaking of which…

Bonus: Ego to Absolvo

In this Storybrick game type, the player explores a mysterious realm populated with souls from Purgatory. The object is for the player to escape the realm by releasing the Pivot Soul, the deceased spirit which binds the player to this place. Souls can be released by either directing them towards an act of contrition (absolve), or by convincing them their earthly transgressions deserves a more permanent punishment (condemn). It is up to the player to decide whether to absolve the Pivot Soul or condemn it; similarly, the player must decide whether to try and release/condemn the other souls (and which ones) before the player leaves. All the souls in each level are connected in some way, either by the event that led to their deaths, or by some other means.

The above concept requires no combat and no particular narrative beyond the Pivot Soul. Better still, you can use the same stock medieval models/setting in your demo while not being tied to a more cliche archetype. And to me, such a scenario is immediately more compelling – assuming the writing is done well, I can imagine seeking out every NPC even after identifying which of them is the Pivot Soul. Players could talk about the “moral” choice they made (absolving a murderer vs leaving them in Purgatory vs sending them to Hell), how many souls they absolved/condemned, which ones, by what means, and so on. And because of the way the Storybrick AI works, you probably could not save them all.

In any case, I do wish the Namaste folks the best of luck. My skepticism is not rooted in malice, or because the tools lack potential. Rather, my skepticism is merely a pragmatic expression of More Show, Less Tell.

Incidentally, if you want someone to brick up a truly convoluted Machiavellian scenario, or want to run with the Ego to Absolvo idea, have your people call my people.


Skeptical of Storybricks

You have probably read about Storybricks from any number of other bloggers. If you have not, well, take your pick. Far be it from me to denigrate free thinking and innovative design (god knows we need some these days), but none of these full-page ads for Storybricks ever seem to answer what is to me the fundamental question:

As a player, why would I WANT this?

Keep that question in mind as you read this excerpt from Too Damn Epic’s Storybrick article:

By focusing on expressive AI, a different experience than currently experienced in MMOs (one that is closer to a tabletop system) becomes available. NPCs will be given drives, emotions and desires. More importantly, relationships between characters will be developed and interactions between the player and characters will affect how characters relate to the player.

For example, imagine a farmer that you as a player have never encountered, an NPC that feels neutrally toward you. This farmer owns a flock of sheep. As a player, you come upon a merchant wanting wool and kill the sheep for the merchant. If you then encounter the farmer his reaction towards you will no longer come from a neutral disposition but one of anger or unfriendliness. And what you experience may be completely different from your friend who did not kill the sheep. Storybricks is the tool that allows players to build these complex relationships into their stories giving the depth to their experiences.”

Put aside for the moment what kind of game architecture would be necessary to compute all these dynamically changing quests and attributes. Would you, as a player, WANT to quest in a potentially labyrinthine “unforeseen consequences” environment? Killing a farmer’s sheep is pretty straightforward of course, but this sort of system could engender a Machiavellian plot wherein finding the Blacksmith’s hammer pisses off the Innkeeper whose daughter is in love with the prince who gives your execution order to the assassin who buys the temper-steel poisoned dagger from the newly upgraded Blacksmith. If someone (non-ironically) wrote a plot that involved every quest you completed as having the opposite effect you intended, you would call that writer a hack. As WOPR would say, “The only winning move is not to play.”

Of course, the Storybrick guys aren’t making a game like that, or any game for that matter; they are simply making the tools. More precisely, tools for a game that does not exist yet, and one in which would require user-generated content submission, such as Neverwinter Nights or Fallout 3/NV. I say “require” because I cannot ever imagine that a game designer would leave the fate of the narrative solely in the hand of player actions. And I certainly could never see it in an MMO. Why? Here is what epic.Ben asserts from the TDE article:

In other words, Storybricks is going to shift the focus in your MMOs. Instead of mindlessly clicking quest text and proceeding through a Pavlovian loop of grinding, achievements, and raiding, you’ll actually pay attention to what’s happening in the world around you. NPCs will display emotional depth, and dynamically react to your experiences in the game world.

One of the hallmarks of the MMO genre is a notion of a persistent world, but that persistence is always in tension with the fact that other players exist. Players say they want a world where consequences matter, that if a town gets burned down it stays burned down. But do they really want a world in which the choice of saving the town is never given to them because some noob 4 years ago logged off in the middle of the quest to put the fire out and the town burned down? “Phasing” was a much-touted Blizzard innovation which amounted to open-world instancing, with exactly zero of the MMO elements intact – the designer reaction to the infinitely frustrating disappearing party member issue of Icecrown was to… phase the background and NPCs in Firelands instead.

Going back to Storybricks, what happens when Bob and Tim want to quest together, but Bob killed the sheep while Tim bought wool instead? Under traditional MMO design, nothing happens because it doesn’t matter how the quest is accomplished (assuming you can even complete a quest different ways). And under Storybricks? I have no idea. Does the farmer have a Schrödinger-sheep that is both alive and dead? Assuming the farmer hates Bob and likes Tim, does he still give out a quest to both, or just Tim? Can Bob get credit if he helps Tim complete it? If the farmer has the same quests regardless of feelings towards Bob, does the farmer having “emotions” matter at all?

This is my fundamental problem with the epic.Ben’s assertion above, and Storybrick’s unspoken premise that “if we build it, they will come.” Or more precisely: “if we build emotive NPCs, the player will care.” Emotional depth and multifaceted character development are the pinnacle of what a storyteller can achieve when crafting a tale, but it always hinges on the listener/player caring about said character first. Why do I care that the farmer is mad at me for killing his sheep? Is he an interesting character? Or am I supposed to care because he holds potentially fun (or required!) quests hostage behind a fancied-up logic gate of prior actions? Much like what happens in real life all the damn time, just because a person interacts with other emotive beings does not mean they care about how other people feel.

I can see and appreciate the work the Storybrick people are doing in building a cart, and hoping that a horse comes along to carry it. Finding a fictional character you have some interest in and realizing that there is emotive, narrative depth to their actions is a truly magical experience. That being said, it still requires someone to have crafted an interesting character to begin with. And if they already went through that much effort to make one, I just do not see the appeal of using Storybricks to fill in the blanks Mad-Lib style. Nor do I see the appeal to the player of being in an environment where potentially random actions have lasting, permanent consequences. If accidentally killing a sheep with a stray Arcane Explosion can be rectified by reputation grinding with the Farmer, that is almost worse.