On Writing Games

interactive flashbacks

2024-05-14T00:00:00+00:00

In traditional writing, flashbacks serve as a tool to provide crucial information about the story that the reader wouldn’t otherwise know without delving into past events.

As a narrative device, flashbacks retain this function in interactive storytelling, yet their interactivity also allows us to shape and influence future events.

In Dance of The Spirits, we introduce the player to April, the protagonist, through a flashback where they select certain characteristics.

These characteristics subsequently influence the available options and even the reactions of other characters as the story unfolds.

I believe that beyond their informative role, flashbacks offer valuable opportunities to “configure” the game through player choices that impact the main storyline.

What are your thoughts?

sunday prompt #1 - memory swap

2024-05-12T00:00:00+00:00

One goal of this list is to assist traditional writers in crafting interactive narratives.

Taking advantage of the serene Sundays, I’d like to propose prompts specifically designed for this purpose.

If my creativity permits, starting today, I’ll post a prompt every Sunday.

If you’re interested, let me know; and if you find it boring, feel free to say so :)

You’re free to interpret this prompt however you like.

Create an interactive short story, a visual novel, or anything else you can think of.

All I ask is that if you use the prompt, share your work with me. Whether it’s finished or not doesn’t matter; I’d love to know if this is helpful.

Memory Swap: The protagonist has the ability to swap memories with others. Each exchange alters what they know, affecting their relationships and options in future interactions.

There you go.

Now, get writing! :)

probabilistic foreshadowing

2024-05-10T00:00:00+00:00

Something that particularly interests me about the foreshadowing technique is that it is a local element connected to the global story.

By “local element,” I mean that it is something that appears in the story momentarily with the aim of providing a hint toward an important event.

As it is something that “predicts” the future of the story, it is very difficult to do traditional foreshadowing in a game where the player’s decisions matter and significantly affect the story.

In the latest iterations I made in Dance Of The Spirits, I have been experimenting with something that I will call probabilistic foreshadowing.

I am going to heavily rely on the concept of silent evidence for this.

If we consider that the player makes decisions with incomplete information, we can have certain moments in the game where we plant a random foreshadowing.

Imagine a situation where you are talking to many people and they all share their mood:

JULES: I'm feeling great to be honest.

LINA: I'm hella tired, it's been a nightmare week. 
Lina looks over your shoulder.
You get curious and turn around.

Bambie is approaching the Circle of Trust. 
They're wearing a dress that looks like it was made by the devil himself.
You can't quite place the color. It looks like some shade of blue.
But a blue that's been chosen for the end of times.

BAMBIE: Sorry I'm a bit late. You were already expressing how you feel?
Lina rolls her eyes.

[bambie_foreshadow]
  (50%) BAMBIE: I'm...
    set bambie_feels happy
    Bambie looks straight at you. 
    They smile.
    BAMBIE: ... pretty happy.
  (50%) BAMBIE: I'm...
    set bambie_feels annoyed
    Bambie looks at Lina, then at you.
    They sigh.
    BAMBIE: ... annoyed.

We use this random selection to choose silent evidence:

If Bambie is happy, it is because they want you there, and they will help you at some critical moment in the story.

If Bambie is annoyed, it is because your presence there goes against some goal they have, and therefore, later on, they will stand in your way.

The player doesn’t know this; that’s why I say it’s silent evidence. But above all, what’s interesting is that it is defined through foreshadowing.

The interesting thing about this is that it creates a space where the final outcome of the story is not 100% defined by the player and not 100% linear either.

The combination of foreshadowing and their corresponding silent evidence, mixed with the player’s decisions, generates a situation where the player’s choices are not “deterministic” but they do make sense, and that sense was developed in the game with the foreshadowing and subsequent reactions to the state as the silent evidence dictates.

PS: In the code snippet I foreshadowed that Bambie Thug is my favorite Eurovision act this year. If you noticed that, you’re my favorite subscriber, period.

natural selection of translations

2024-05-07T00:00:00+00:00

A third consideration (because we appreciate third positions in this list) is to emulate the book publishing industry.

We can regard the “number of languages the game was translated into” as a metric of success.

Thus, the game is translated only when there is specific demand in a language that justifies the cost.

Internally, I’m torn between translations as a metric of success and universalizing access to my games.

On one hand, I favor measuring success based on demand, but on the other hand, the scarcity of supply can restrict demand for our niche games.

This poses a dilemma for which I’m unsure of a solution.

Independent games typically delegate translation responsibilities to publishers.

Unfortunately, publishers willing to secure rights for specific geographical regions are becoming increasingly scarce.

While there are numerous publishers cross-releasing games between Eastern and Western markets, finding one interested solely in publishing an English-developed game in Germany, exclusively for the German market, is challenging.

This used to be the norm, but with the advent of digital downloads via platforms like Steam, this practice disappeared.

Unlike the film industry, which still exhibits regionalism despite digital platform emergence, we lack such distinctions.

Thus, an intermediate space exists that we must navigate, with each individual determining their stance.

I plan to adopt the following strategy for Dance of The Spirits:

Translate the Steam page into as many languages as feasible, engaging localizers for the three or four most important languages and utilizing LLMs for the remainder.
Assess wishlist distribution and launch with human translations for English and any language with a similar wishlist count.
If the game generates sufficient revenue to cover localization costs (at least 10% of earnings), I’ll localize it for the next language with a substantial wishlist count.

One question I have is whether players would appreciate or object to LLM translations if clearly labeled.

I believe it’s misleading to release a game with automatic translations without clarification.

However, if the language selection features an “AI” tag (I’m not contesting marketing strategies here), would it be as significant? What are your thoughts?

NOTE:

The language order may drastically change post-launch.

In my experience, wishlists often surge, altering language distribution compared to pre-launch.

Moreover, a prominent local streamer playing your game can significantly influence your distribution.

Thus, for point 3, periodic reevaluation of wishlist distribution is necessary to determine the next localization target, particularly after wishlist spikes.

show me the numbers!

2024-05-06T00:00:00+00:00

In the previous approach, we neglected the quality of the translation.

However, I doubt that this mindset reflects the majority.

During one of our weekly meetings, Seba Gioseffi mentioned an alternative approach.

Seba credited Chris Zukowski with this insight, although we couldn’t locate a corresponding blog post; it might have been a tweet or similar.

What we did discover was a blog post where he discusses the impact of translating your Steam page and a comment on Steamworks:

Tip: If you translate the content on your store page into languages that you are considering supporting, you can look at regional wishlists for your game to get a sense of where your game might be popular and which languages might warrant higher priority for translation of your game.

This presents an intriguing approach to addressing the cost of localization with limited resources:

Translate the Steam page using LLMs.
Analyze wishlist distribution.
Conduct human translations for regions with significant wishlists and LLM translations for others (or omit translation entirely for those regions).

I place more trust in an LLM’s ability to translate marketing text tailored for Steam than in localizing an entire game accurately.

What I find compelling about this approach is its utility even when lacking funds to translate into languages with the most wishlists.

Rather than scrambling to find collaborators for numerous languages, you can focus your efforts on languages with the highest wishlist counts.

Based on my experience, wishlist numbers decrease rapidly, with only 2 or 3 languages having significant wishlists while others have very few.

Tomorrow, we will explore the third approach, in which we rely on “old” mindsets.

i don't care, i love it

2024-05-05T00:00:00+00:00

The first approach (of three) I want to talk about is using LLMs to translate to as many languages as possible without worrying too much about the final result.

Given that LLMs have proven to be good enough in the languages I can evaluate, I am going to assume that they will work similarly in others.

Obviously, this is not scientifically correct.

It is known that LLMs have greater abilities with languages that have a lot of text available on the internet, especially given the significant distance between English and other languages.

I do not have enough data, but from what I see, I tend to trust that to some extent the Pareto principle holds true.

80/20 is almost the same as 70/30.

As for the cost of translation, it is a bit difficult to calculate, but possible.

The tokenization process (transforming the text into “numbers” so that the LLM can interpret it) varies depending on the provider and the language of origin.

For OpenAI, English is cheaper than Spanish for example:

English: 1 word ~ 1.3 tokens
Spanish: 1 word ~ 2 tokens

Depending on the model we use and the number of tokens, we can calculate the costs.

I can already say that for an interactive fiction game, the costs are one or two orders of magnitude lower when translating with LLMs than with humans.

I do not want to make ethical judgments on this [yet], I am just presenting the data.

I will not give my personal opinion until the end of the series, for now, I am just presenting strategies.

As I said in the previous email: the result needs to be playtested, unless you really don’t care and love it.

PS: In one way or another, I managed to bring Charli XCX into this 😅.

translation is not localization

2024-05-04T00:00:00+00:00

Today I want to clarify the nomenclature a bit.

A dear reader of the list, Andreas Lopez, responded to yesterday’s email:

My wife is a linguist and fluent in 4 languages (Spanish, English, French, and Japanese). She is actually thoroughly disappointed in the LLM translations because too often they ignore cultural relevance and dialects and mix and match them which can lead to confusion for the people you are making it - the localized versions.

Especially in interactive fiction it’s not about actual translation but localizing it. We have many ‘odd’ phrases in Germany we exchange every day but a different phrase reveals your locale or homestate.

Andreas refers to the fact that translating is not localizing.

There is a difference between putting a text in another language and capturing the essence of that text to reproduce it with local forms and manners.

This is a hard problem to solve, and most game translations suffer from this issue at different levels, even when done by humans.

I will “ignore” this problem for now, because I think this is a problem that LLMs will solve eventually.

I may be wrong, because I’m not aware of the computational complexity of this problem, but I think it might get solved with future LLM generations.

I would keep my hands off of LLM just because the risk of pissing people off with bad writing is just too big for my personal pride and risk levels.

As an indie with $0 budget you might be better off asking for volunteers or just accept you might piss people off due to insensitive translations which you won’t be able to check unless you know the language you have the machine translate into.

Yes. But also maybe not?

You can’t release a game translated with LLMs just like that. That’s a given.

Getting volunteers is not always feasible, especially with games with high wordcount.

The readers of the list can guess my approach: playtesting.

You should never release anything without it being tested, and translations are no exception.

In the next emails, we will discuss some approaches for localization using LLMs (or not).

the unavoidable cost of interactive fiction

2024-05-03T00:00:00+00:00

Interactive fiction is less costly to develop than practically any game that requires a lot of graphics, especially with real-time motion and 3D graphics.

Except in one area of development: translations.

The cost of translation is usually per word, and this makes it prohibitive to translate the game into many languages. There are games that have a small menu, and maybe some text here and there.

If I remember correctly, Nubarrón had a total of 2000 words.

Dance of the Spirits is not finished and already has 20,000 words.

Being in the era of LLMs, this might not be a problem anymore.

I have the ability to evaluate the quality of translation of an LLM in English, Portuguese, Italian, and Spanish; and the truth is that I believe they do a good job.

At least they comply with the Pareto principle, which is already better than many translators, especially those that I could afford as an independent developer.

I am not going to spend time discussing the ethical issue, because I already gave my opinion on it.

In the next emails, I want to share with you some ideas I have on the subject. I was going to include everything in this email, but it turned out to be very long, so I decided to break it down into three more emails to elaborate a little more on each of the three ideas I considered.

llm storytelling in 2024

2024-05-02T00:00:00+00:00

Every once in a while I like to do a review of the state of interactive narrative using LLMs.

In the past few months there has been some progress, but I’m still not sure if we are at a point where we can create unforgettable experiences.

I wonder if this is a natural limitation of LLMs or if we still haven’t found the right way to exploit them.

For those who don’t know how LLMs work: given a series of words (tokens) they determine which is the next word (or token) most likely to make sense, also adding some randomness in between.

My doubt is if the “most likely” approach is so core to the system that the random part will never be able to write very interesting things.

I don’t mean to say that we will never be able to automatically generate interesting interactive narrative, because eventually we will be able to, my question is if the architecture of LLMs is sufficient to achieve this or if we need a different architecture.

It’s not clear that the architecture can generate complex models of reasoning that can surprise us in fundamental ways.

Until we see a new generation of LLMs that can understand essential things like time and space, keep in memory everything that has happened in our story, and maintain coherence, have a sensitivity of our mental models to manage expectations, and other skills that writers use daily, we are still far away.

PS: as a person who likes precise definitions, it’s not surprising that I don’t use the term ‘AI’. In this case I use the more correct term LLM (Large Language Model) which represents deep learning models developed to write text.

second person

2024-05-01T00:00:00+00:00

There is a maxim when one starts writing: do not use the second person.

I have heard this many times, and it is clear that there is a literary tradition that favors the first and third person.

However, interactive narrative as we know it today has as its great-grandmother the collective narrative of tabletop role-playing games, where the second person is mainly used.

You are in this situation; what do you do about it?

I’m not saying that all narrative games have to be in the second person.

But it is an immersion tool that the traditional writer often ignores.

You have a new tool,

What are you going to do with it?

PS: It’s the first of May, Labor Day, so I sat down to write a bit. In recent months, my ADHD has gotten the best of me, and I committed to many very big things. Now I’m trying to distribute my efforts a little better. Hope I can handle it 🐌. I created a now page on my blog and a reminder to update it to make sure I follow through.

character vs. player

2024-03-14T00:00:00+00:00

In the lines where I talked about emotions yesterday, there is a contrasting idea that I wanted to share with you.

In one of our narrative meetings with Nico Saraintaris and Sebas Gioseffi, Nico mentioned that there is a character in his Pixel Pulp for which he receives a lot of hate.

There is a very strong phenomenon of identification in video games that, although it may appear in some books, is very rare and I wouldn’t know for certain if it happens in the same way.

Players, having agency in the game, do not create a barrier between the protagonist and themselves.

This is reinforced by the concept of avatars, digital versions of oneself.

This identification, or lack of separation, generates a very strong negative feeling when the character in the game acts in ways that clash with the player’s intentions.

It is enough for the character one is controlling to say something one disagrees with for a very strong negative emotion to arise.

This bothers the traditional writer.

And rightfully so, this greatly limits the type of character one can develop.

Especially in a time where cancel culture is so strong.

In a way, it makes us feel like we are obligated to create characters that reflect the opinions of the majority.

Or if they don’t, that they are cute and sympathetic “underdogs”.

We can also ignore this and see what happens.

Nico did that, and he received backlash from some players.

The feeling it leaves behind is not nice at all, but it is clearly a not very good consequence of interactivity.

Have you felt this? Do you have a go-to way to approach this problem?

hello, emotions? where are you?

2024-03-13T00:00:00+00:00

In a way, those of us creating interactive narratives are neophiles.

We see interactivity as something positive, as a step in the evolution of storytelling.

But I also believe that, like with any technology, we have to spend some time thinking about the negative consequences.

I have the following uncertainty:

Adding interactivity and choices to traditional narratives, does it take away emotional involvement in some way?

What I know is that “gamey” systems erode deepness in some emotions, but help others.

My question mainly comes from whether in the player’s mind, knowing that they can “go back” and change their decisions removes the permanence of the protagonist’s decisions in a story, thereby taking away [part of] the player’s emotional involvement.

I don’t know if this is true or not, and I think only time will tell.

But, assuming this is true, it would mean that interactive stories are inferior to linear stories in terms of generating emotions.

What do you think about that? Is there anything that makes you think this is the case? Or something that makes you think the opposite?

NarraScope 2024

2024-03-12T00:00:00+00:00

Today is a stormy day in Buenos Aires.

I’m lacking ideas and feeling a bit tired so I’m going to take the opportunity to share a piece of news instead of a reflection.

The tickets for NarraScope 2024 are now available, which will have both an online and in-person version.

In their words:

NarraScope is an event that supports interactive narrative, adventure games, and interactive fiction by bringing together writers, developers, and players.

I will be participating as a remote listener this year.

Last year I had plans to participate with a talk about cuentitos, but it coincided with The Canadian Debacle and a season of severe and painful migraines.

So, I had to cancel the talk.

But I’m looking forward to doing something, even if it’s just a Zoom call or something during the event, I will be thinking about it and will inform you. If you have any cool idea let me know.

You can get your tickets at https://narrascope.org/

It is a self-managed event by the Interactive Fiction Technology Foundation and very friendly to empty pockets 👍.

solid new grounds via negativa

2024-03-11T00:00:00+00:00

Today I was thinking about how when one writes a linear story, they act as a curator of a rhizome of possible stories.

This continues to be the case when we create interactive stories, except that our goal is constantly to consider which possible stories make sense, not limiting ourselves to just one.

This selection process is the very nature of art, and it is what, in my opinion, sets the bar between good, bad, and excellent.

While in the video game industry we have some intuitions (which are increasing every day), we are still a very new creative area to have found the sensitivity to curate what works.

As I have mentioned a couple of times, I see interactivity as a new element in the narrative system that came to break the “rules” of the previous system and has not yet stabilized.

Most developers are obsessed with finding the formulas of what works.

I come to propose this to you: nobody has a damn clue.

Why don’t we start focusing on what doesn’t work?

That way, instead of all being stuck in a very limited series of proto-formulas, we begin to see the places where it’s not worth walking and open up exploration of other places.

Instead of watching a video about “The 10 things you need to do to make your game successful,” why not read 10 negative reviews marked as helpful on Steam?

Those are the things that are not working.

Let’s stop doing them and look for new ones.

try not to break the fictive dream

2024-03-08T00:00:00+00:00

In the book “The Art of Fiction”, John Gardener talks to us about the Fictive Dream.

The reader enters a dream when they start reading.

And it is our job as writers not to wake them up from this dream.

If we bore them or break the suspension of disbelief, boom, they wake up.

“In bad or unsatisfying fiction, this fictional dream is interrupted from time to time by some mistake or conscious ploy on the part of the artist. We are abruptly snapped out of the dream, forced to think of the writer or the writing.” John Gardener

This problem, obviously, goes beyond fiction.

In video games we call it retention.

What is interesting to highlight is that retention poses different scales.

There is an element of retention in that we call moment-to-moment gameplay. This retention has to do with the gameplay being satisfactory at all times.

The player has to have a more or less clear intention horizon, and achieve effects and progress through their movements in the game space. All of this has to feel satisfying.

As Gardener would say, we must not bore them, or they wake up.

But we also have other types of retention. We have elements of extrinsic motivation: achievements, unlockables, events, etc.

These types of retention are more long-term, the player returns to play our game driven by that extrinsic motivation.

When I think about how the Fictive Dream manifests in games, I specifically think about moment-to-moment gameplay.

All extrinsic retention tools necessarily break the Fictive Dream and have a more gamey nature, separate from storytelling: the story disappears and the focus is on the game.

Maintaining the Fictive Dream must be the first item to keep an eye on during playtesting.

If you are making a narrative game, you have to treat the rupture of the Fictive Dream as a high-priority bug.

Otherwise, you are leaving your players without what they came to seek.

sorting gameplay

2024-03-07T00:00:00+00:00

Here is an idea via negativa.

A set of small pieces of narrative delivered in a random order is not a dynamic story.

This is what happens in games like Gone Home, Dear Esther, and other walking simulators without reactivity.

If we give the player a series of ordered or disordered nuggets of story, it is irrelevant.

The player always ends up organizing the ideas in their head.

A disordered linear story is as linear as an ordered one.

All we are doing is making the player work a little harder.

This is good because organizing the information in your head is gameplay.

But let’s not confuse sorting gameplay with a dynamic story.

A dynamic story can have sorting gameplay, but having sorting gameplay does not make a story dynamic.

Oh, Logic❤️.

dynamic three-act structure

2024-03-06T00:00:00+00:00

There is something that has me thinking daily: the three-act structure.

On one hand, the escape from linear narrative makes us visualize our stories as interconnected rhizomes; narrative elements affecting each other.

On the other hand, we know that our players experience our stories linearly: one possible combination at a time.

Therefore, the three-act structure still remains relevant in some way.

To tell a good story, we still need to guide the player through the three acts, introduce the world, guide through the conflict, and resolve it.

It is not appropriate to think that because we have dynamism now, the old structures are no longer useful.

Furthermore, the Lindy effect almost assures us that the three-act structure will continue to be dominant.

Therefore, I think that in some way we have to factor this into our narrative design, at least as long as the majority of the population continues to consume stories in this way.

It remains to be seen if TikTok manages to burn a generation’s brain enough for the three-act structure to become irrelevant.

The first thing that comes to mind is the obvious: use the three-act structure and suffer from exponential growth:

We use the introduction to present the narrative problem and the special narrative mechanics of our game, where we may give the player some options.

We will have as many second acts as possible combinations of options in the introduction.

And we can converge on a third act with a single ending, or have multiple third acts with several endings taking into account the possible combinations of first and second acts.

Here is where we see that the effect is exponential if we are not careful.

So far, this is the obvious solution that came to mind when thinking about this problem.

But well, little by little we will experiment and find other answers.

Do you have anything in mind?

nice outfit, you look like a gardener

2024-03-05T00:00:00+00:00

Put yourself in the shoes of a player for a moment.

You start playing a narrative game.

During the introduction, the game presents an interesting mechanic: the story changes depending on the clothing you are wearing.

You had read about that mechanic in the game description and it caught your attention, now is the time to see how it is.

You put on a denim overall. Because the character looks super cool.

In the first conversation you have, you meet an NPC.

“Nice outfit, you look like a gardener.”

Your character gets angry and the game continues.

The game continues.

The story gains momentum, at a critical moment, your character has to jump over a 2-meter pit.

He does it without any problems.

The story ends. All good.

You play the game again. This time wearing a skirt.

“You look so pretty dressed up.” The NPC tells you in the first conversation and your character blushes.

The game continues.

The story gains momentum, at a critical moment your character has to jump over a 2-meter pit.

He does it without any problem.

The story ends. All good.

What is the purpose of being able to choose clothing in this game?

other hidden nonlinearities

2024-03-04T00:00:00+00:00

Simplify, simplify, simplify! I say, let your affairs be as two or three, and not a hundred or a thousand Thoreau

Just as I insist on not starting your first game with complicated structures (multiple endings, dynamic branching, probabilities), I also suggest that you start with a short story.

Narrative complexity is non-linear: a novel is not just a longer short story.

A novel has properties that do not exist in short stories, and said properties arise because of its length.

In order to sustain a long linear story we usually need multiple characters, locations, scenes, motivations, pacing changes, etc.

And each of these things has such a level of interconnection with everything else that it generates a network of dependencies and chain reactions.

The care of this network of dependencies is essential to maintain the consistency of our story.

And all this is just talking about a linear story.

If we add interactivity into the mix, things get way worse, because an interactive story is not a story with choices.

There are properties that arise because of interactivity.

On the other hand, when one has a project and estimates it, it is often said that one must multiply the estimated time by 2 or 3.

There is a mathematical basis for this: time is not volatile, it can only move forward and not backward. Therefore, there cannot be errors that shorten the project (negative time), but there are plenty of errors that can delay it. And this delays are also non-linear, one thing affects the other and so on.

That’s why I always insist on making short games, or experiences that grow horizontally, more variety among the options of a short story instead of very long stories.

interactive storytelling and its nonlinearities

2024-02-29T00:00:00+00:00

During our almost weekly call about narrative and life, Nico Saraintaris made a comment about multiple endings that I found interesting to share.

If we want our game to have multiple endings, and to work well, it’s not that we multiply the work of one ending by the number of endings.

Instead, there’s a compounded effect of work that involves a larger amount (and much larger the more endings we add).

This, for nerds like me, is an example of a complex system where adding a new agent has a non-linear effect.

For non-nerds, let me steal some examples from Nassim Taleb once again: a state is not a big municipality, and a country is not a big state. A human body cannot be explained as a set of atoms.

The properties that emerge from adding elements to a system cannot be explained with the theories that serve for the element.

And that’s where I’m currently working, on the line that separates traditional narrative (a story, without interactions) and interactive narrative, with all the new elements that affect the system we know of storytelling.

Many of the problems we see in games today come from not taking this into account, and believing that a narrative game is a story with options that works the same as linear stories.

Nope.

beware of the filler endings

2024-02-27T00:00:00+00:00

The entire narrative construction is directed towards the ending.

And the main condition for the ending is that it has to be satisfactory.

Unfortunately, it is too common for games with multiple endings to have filler endings.

These endings are quite easy to spot: they are usually branches that are clearly not “canonical”.

I obviously understand how difficult it is to make multiple endings that are all up to the same standard, as writing just one can be a daunting task.

But the player experiences our story as linear, and if the end of the line does not satisfy them, they will never see the quality endings.

In a book, the end of each scene makes it a page-turner.

In video games, the end of a run makes it replayable.

linearity is unavoidable

2024-02-26T00:00:00+00:00

As interactive story designers, we run the risk of hyper-focusing on the dynamic aspects of our story.

We have to pay close attention to the branches, the foreshadowing between branches, and the variables that modify the story.

Why do I say we run the risk?

Because no matter how amazing our dynamic narrative system is, there is a reality we cannot escape:

The player experiences our story in a linear form.

Especially on the first run.

If we want the player to enjoy our story, it has to work as a linear story, no matter what choices they made.

iterative and incremental

2024-02-25T00:00:00+00:00

Finishing a novel is a lot of work.

Very few people start and finish one.

That’s why it’s no surprise that even fewer people finish an interactive story.

Everything we take into account to make a good novel is multiplied by a number that may or may not be under our control.

That’s why we have to start slowly and not try to solve all the problems at once.

If ID Software had done that, we wouldn’t have Wolfenstein or Doom. Just Quake.

In software engineering, the Agile community has a concept that I adopted as a philosophy: iterative and incremental.

What does this mean? It means starting from scratch and doing the minimum to feel like there was progress.

This could be writing a linear interactive story with a single character.

Once we finish that, we can take another step and adjust our process to support something more complicated.

Now we can create a story with a unique ending but with some cosmetic branches.

Then more branches and maybe two endings.

We iterate on this and create a story where the character has two “personalities” and interacts with a character that reacts to this.

And so on.

This doesn’t mean you should release stories to the world that you’re not happy with.

You don’t need to publish those stories if they don’t live up to what you want to be known for.

The important thing is that the byproduct of finishing an entire story is a process.

And by adding something in the next iteration, that process will change.

But there is a hormesis process you need to go through.

Every element you add to your process will damage it.

The process needs some time to find a new balance.

If we try to mix all the elements at once without having a system, the result will probably be exhaustion and cognitive overload.

Remember: iterative and incremental.

the anchor and the hooks

2024-02-23T00:00:00+00:00

Here are two definitions:

Anchor: A structure, system, or genre very familiar to the player. It allows us to assume a certain amount of things as learned.
Hook: It’s an element that attracts the player to our game. It tends to be some kind of distinctive feature.

And how they connect:

To sell something familiar, make it surprising. To sell something surprising, make it familiar. ― Derek Thompson, Hit Makers: How Things Become Popular

If the word “sell” doesn’t resonate with you, think of it as a metaphor for making what you do interesting to someone

This is where the idea of using a specific game or narrative genre comes into play.

From the “box art” of our game, the mechanics, character design, and more; using familiar elements helps the player navigate our experience.

That’s why we want to know what anchors our game has.

Design around these anchors so that the player starts from the familiar.

And let’s not be snobbish, please.

Whether we like it or not, everything we do has anchors.

And if it doesn’t, it will be confusing and discarded.

On the other hand, if what we do is a copy of something familiar, the player will hardly find it attractive.

This is where hooks come in.

We want to have the highest number of elements that are attractive to the player.

This can be a story promise, a writing style, a genre subversion that is novel. Or all of these things together.

The reason we want to be very explicit with the anchor and hooks is very simple: we are helping people understand and talk about our story.

If we can make our story have a phrase like: “It’s X but Y”

Where X is the anchor and Y are the hooks.

We make our players’ lives much easier.

This is commonly known as the “Elevator Pitch”.

This is not strict, but can be with or something similar.

Here are some examples:

Slay the Princess is a choice-driven psychological horror visual novel/dating sim with dramatic branching, light RPG elements, and hand-penciled art.

Anchor: Psychological horror visual novel/dating sim
Hook: dramatic branching, light RPG elements, hand-penciled art.

Roadwarden is an illustrated text-based RPG that uses isometric pixel art and combines mechanics borrowed from RPGs, Visual Novels, adventure games and interactive fiction.

Anchor: text-based RPG
Hook: illustrated with isometric pixel art (since it’s text-based, this is a hook), combination of mechanics from different familiar genres.

A note:

There are more hooks in these games when you go through the descriptions, and more when you play them.

Not everything goes into the Elevator Pitch, only those that are more impactful.

signposting character traits

2024-02-22T00:00:00+00:00

Above many other things, the consistency of actions and personality of our main character is central to maintain the illusion of disbelief.

There is a very specific risk that we run in interactive fiction when we allow our character to have traits or when certain choices define their personality.

The problem lies in the fact that if this choice of personality or trait comes at a later stage in the game, the player may a already have a preconceived notion of how their character should act in different situations.

And when this doesn’t happen, there is a moment of disconnection and even displeasure.

I’m sure you’ve experienced this a thousand times, not only in interactive fiction but also in books.

Luckily, the solution to this problem is practically the same: Signposting.

If we did a good job of “painting the picture” of who the character is up to a certain point and we have an option that adds depth to the personality (for example, choosing 1 out of 5 personality traits), the best thing we can do is write a signpost for each type of personality.

That is, write scenes or descriptions that help us make sense of the character’s future actions thanks to their new personality or trait.

In this way, we include a transition stage between how the character behaved before, and how they should behave afterwards, improving our chances of not falling into actions that the player finds dissonant.

my stance on 'AI'

2024-02-21T00:00:00+00:00

Yesterday I was asked about my stance on AI.

First of all, the question is wrong.

What they wanted to ask me is what I think about Large Language Models (or LLMs, like ChatGPT), “artificial intelligence” is a marketing term and very broad. Anyway.

The answer, as with everything complex, is: I’m not sure.

I don’t believe in copyright as a system for protecting rights.

I believe in a system of collaboration and the universal generation of knowledge.

I believe that we are not as original as individuals, but rather a collective construction.

No matter how much self-help books try to convince us otherwise (and liberal individualism, which is the zeitgeist).

This same belief is what turns my answer into “I’m not sure”.

Because on one hand, LLMs are a first step in building a global cultural tool.

But on the other hand, I am not very happy that two or three companies are governing the outcome of this.

It’s the same old discussion, GNU/Linux vs Windows all over again.

In a couple of points, I believe that:

It is hypocritical to use copyright as a flag against technological progress, given that we are all a collective construction. No one lists all the books that influenced them, maybe we mention one or two in interviews.
I am concerned about technological concentration, a few companies dominating the field.
I look with much love at what is happening with open models (LLaMA 2, BLOOM, etc), although I would like them to be more open: to open not only the weights, but also the datasets, and processes.
I am concerned that writers with technological aversion and stubbornness may not be able to adapt to the new models and end up without work. Same thing with visual artists and other areas that are affected.
I am concerned about the second-order consequences of technological progress in general, but this is just anxiety, I can’t do much about it.
We are going to have a new period of information deterioration (articles generated by LLMs, fake news, a lot of garbage).

And above all (although this is the bottom): - I believe LLMs are the future of interactive fiction.

The use of RAG (Retrieval-Augmented Generation) in combination with models fine-tuned to our writing style will become the norm.

That’s why I think it’s important that, while the world is having the ethical conversation (whether we are part of it or not), we have to learn to use these tools and be part of the exploration.

If you don’t feel comfortable making money out of the results, then don’t charge, but explore anyways, tinkering is what creates gems.

choose your own story beats

2024-02-20T00:00:00+00:00

I was remembering that choose your own adventure books used to have a magic that I can’t find in many current games.

So I dusted off one of them and started to review with a designer’s eyes what was generating that magical feeling for me.

This has nothing to do with it, but: it was quite interesting that one makes choices blindly in these books. It is not clear what one is choosing by following an option and obviously that is not necessarily bad, otherwise these books would not have had the success they had.

Now, the most relevant thing in my opinion is the variety of paths that are generated by choosing these options.

We can find ourselves on a spaceship making friends with aliens or underground leading a revolution of subterranean beings, all in the same book.

Obviously, the writers of choose your own adventure already understood something that many writers today forget: the intense variety of story beats generates an agreement between the player and the designer that invites to replay the experience.

Another agreement that these books have is that each story is relatively short: another property that works very well for replayability.

How is it possible that we have forgotten these two characteristics?

Why do we fall into minor variations when we could invite players to explore alternative stories but as alternative as they did in the choose your own adventure axis?

I’m not going to tire of repeating it, I prefer a game with 10 one-hour stories than a game with a 10-hour story.

And don’t come to me with production efforts and that it’s very expensive: a 10-hour game is just a little less effort than 10 one-hour games.

And if you run out of money, it’s better to have 3 releaseable one-hour stories than a third of a 10-hour one, which can’t be released.

the objective

2024-02-19T00:00:00+00:00

The player generally has a clear goal, and uses the game mechanics to achieve that objective.

Sometimes it is very explicit, like “Collect 20 bat wings”.

But other times, it is not so explicit, to the point that there are games where the objective is simply to go through the experience, as is the case with walking simulators.

When we design our stories, we want to make a decision regarding the objective.

Is exploring the options a sufficient objective to entertain the player?

Do we want to make the objective explicit?

Do we want multiple specific objectives that unlock in different branches?

Or do we not want the player to have anything else in mind other than our story?

Do we have micro-objectives or more long-term ones?

PS: I’m dealing with a burnout crash. This is a known cycle for me (as it is for many ADHDers), I load up on commitments and then crash. That means that I have to navigate my energies as better as I can. I’ll try to write every day, but I can’t promise it right now, hope you’re still happy with what I write and stick with me. Have a great week!

main character's background

2024-02-09T00:00:00+00:00

In traditional media, the reader has a single role, that of the audience.

In interactive stories, this becomes complicated because there are three roles at the same time: audience, actor, and player.

This introduces a somewhat profound problem in relation to the information available to them.

A character in a book (the actor) exists in the world. They have a pre-existing background that helps them function coherently.

In a video game, the player usually doesn’t have this background, they bring their own, so there is a disconnect in their role as an actor.

This problem exists to a lesser extent in traditional media: the viewer doesn’t have the same information as the actor. But it is usually subtly resolved in the introductions of the stories.

I believe that this is one of the reasons why narrative games tend towards the mystery genre: because neither the player nor the character have the background, but they have to discover everything.

Another go-to is heroism: there’s not much background needed to play the heroic role, it’s very well understood.

If we can create tutorials to teach very complex mechanics, why is there little work done in the construction of the character’s background?

I think the solution to avoiding the easiness of mystery/heroism lies in learning to write good introductions to our stories.

Do you know any game that doesn’t rely on mystery/heroism and that provides a very good introduction to the world and the character’s background to give you the tools to become a good actor?

the programmer writer

2024-02-08T00:00:00+00:00

Many writers have a bit of aversion to technology.

The same goes for many game designers.

I am of the opinion that if you are going to work on creating interactive experiences, you have to learn how to code.

Mind you, I’m not referring to learning how to build an operating system from scratch, with a bootloader and all.

No, I mean understanding the mechanics of programming.

Being able to express a logical idea in some kind of language.

The reason for this is that there are certain mental constructs that programming teaches that greatly help organize your ideas and enable how to express them.

You need to understand the basic mechanics of decision making (if), how to express truth values (true, false, and, or, xor), and grasp the foundation of program flow (while, for).

This is necessary to make good use of variables and to react to state.

The ability to abstract in terms of decision logic is a very valuable tool for creating dynamic stories, and it needs to be trained.

If you know how to code, good, not much for you here.

But if you have no idea about programming, why not take a free introductory course, something simple? What do you have to lose?

Eventually the mental models will form, and things will be way easier.

and a pinch of chaos, please

2024-02-07T00:00:00+00:00

When you want to make hot chocolate, you put milk (plant-based, obviously) and cocoa in a glass.

The order is up to you so as not to encourage pet peeves.

But it’s not enough to just put cocoa and milk.

You have to stir.

Every static system needs an external and chaotic force to reach a new balance, better than the previous one.

Why does this matter to you?

Because we are already tired of the hero’s journey.

We are already tired of the Disney and Marvel structure.

We are already tired of “choices matter” that don’t really matter.

Most of today’s narrative games are milk and cocoa, but few have been stirred.

What in your life experience would inject a little chaos into predefined structures?

It is impossible for our game to “be talked about” if we use the same forms as everything before.

Why tell the same story, in the same way?

I am rewriting some sections of Dance of the Spirits with this in mind…

PS: This came up yesterday during a conversation with my respected friend Mark Cordell Holmes, follow him on LinkedIn, his insights into visual storytelling are out of this world.

user investment in branching narrative

2024-02-06T00:00:00+00:00

Some time ago I read a reddit post where the author commented that, for them, games with branches suffer from an inherent design problem:

… the knowledge that a game’s story responds to your choices makes you less invested in the current story, because there’s always the knowledge you can go back and do something else.

I have never experienced this, I disengage from the story if the writing is bad.

But with dynamic writing, good story beats, and a good treatment of agency, I don’t feel this detachment from the story at all.

In fact, good interactive narrative work, makes want to replay to see what other stories there are besides my first run.

It’s possible that there are some players who decide to do option spamming.

But those are not the players we have in mind when we write our stories.

What do you think? Is there any truth to the detachment from stories due to the possibility of “rewind”?

my problem with voice acting

2024-02-02T00:00:00+00:00

It is assumed that a game with voices is preferable to a pure text one.

I believe that most “talkie” games have an inherent problem that remains to be solved.

I don’t know if it’s just me, or if it’s a problem that many of us suffer from, but many games with multiple dialogue options break immersion big time.

When I am exploring different dialogue options, the intonation is designed to work with a given order, a cadence as one delves into the conversation.

And many times, the order in which one chooses the options makes the intonations sound unnatural.

This is especially true when there are many “root” options, meaning options that are not 100% related between them, and can be chosen in any order.

Here I found an example (not very good) in Stray Gods.

Notice how every time an option from the root of the dialogue tree is selected, it feels like a new conversation is starting from scratch.

This is understandable, voice actors record all the dialogue once and they don’t know the intonation that the conversation should have at any given point.

Maybe this is something that will be solved when text-to-speech becomes ubiquitous, but for now there isn’t much of a solution.

Unless our game is pure text.

We can greatly improve immersion by considering where the conversation is coming from and how we can continue it in a more natural way.

Instead of having all these discrete paths, react to the state, knowing which options were chosen, and change both the questions and the continuations of the dialogues a little so that they flow better.

I don’t know, just an idea to explore…

Do you suffer from this issue with talkie games?

I’m wondering if I’m just too picky 🤣

truly branching narrative

2024-01-31T00:00:00+00:00

The more I try different games that claim to have branching narrative, the more I realize that the term doesn’t give much information to the player.

In general, what this means is that there is a system in which the player can choose options and those options allow them to go down one branch or another.

The problem I see is that in most games that promise branching narrative, the branches are practically cosmetic.

At most, they result in one ending or another, but the story beats remain the same.

I believe there are 2 types of branching narrative that are not cosmetic:

Where the choices significantly modify the story, changing not only details but also which story beats are part of the story.
A linear story without variations, but where each branch provides information that is not available in the other branches.

Can you think of any other type of branching that works well and doesn’t leave the feeling of “lack of agency”?

last words on Stray Gods

2024-01-29T00:00:00+00:00

Ok. I finished Stray Gods.

I liked it, but I think it’s because it’s the first game that tries to bring Broadway from the manual to video games.

Unfortunately, I have to say that the game works, but not as a great showcase of interactive fiction.

The story is ok, it’s not Sandman, but it’s captivating.

The main problem is, like in many narrative games, there’s no agency, few reasons to replay, etc.

It’s almost a linear game, there are few things you can change, and definitely a waste of opportunities to do memorable things.

It saddens me because the game’s lore works, the characters are good, the art works and everything.

Unfortunately, it doesn’t go beyond.

The game lasts 7 hours, and it’s very ambitious. I don’t understand why they didn’t make it last 2 hours but with 3 very different stories.

Anyway, I enjoyed the first run a lot, but more because I’m a theater kid than anything else: if you like musicals, give it a try, otherwise… next!

I still think that we need more games that think of stories as spaces for exploration, that design using the rhizome as a central element and leave behind the linearity of traditional media.

PS: You may have noticed that I didn’t send emails over the weekend. I’m trying this new writing format from Monday to Friday and resting on weekends. We’ll see how it goes.

interactive musical

2024-01-26T00:00:00+00:00

Those of you who know me know I’m a sucker for musicals.

It’s not rare to find me cooking while singing my heart out to The Next Right Thing, Burn or Breathe.

Yes, falsetto and all.

So it shouldn’t be a surprise that I’ve always been waiting for a Broadway-style musical game.

This morning I was finally able to find some time to play Stray Gods.

I still don’t know much about the game itself, but at least I was happy to see that they implemented the basic mechanic that I had been wanting:

That there are timed options to choose what to sing during the song ❤️

This was going to make or break the game for me. So I’m happy they did this.

I could have not imagined an interactive musical without this mechanic.

Now I’m excited to keep playing it, so far I’m really enjoying it, despite it being a murder mystery (I’m getting a bit tired of the genre).

But I’m also enjoying the adaptation of Greek mythology to the new world.

For now, I recommend it, if only to play that mechanic.

We’ll see if it holds up or not.

PS: Any of you know Lin-Manuel Miranda? If so, I’d love an intro 🤭

poor design decisions in Late Shift

2024-01-25T00:00:00+00:00

Yesterday I decided to slowly tackle my narrative game backlog.

I grabbed my Steam Deck and played through Late Shift.

According to the Steam Store:

Late Shift is a high stakes FMV crime thriller. Forced into a brutal London heist, your choices matter in this interactive cinematic experience with adaptable storylines that lead to one of seven conclusions. Your decisions are you.

I’m afraid to say that this is one case of marketing being better than the actual thing.

This game suffers from almost all the problems I complain about on this list:

It’s video-based, so it takes time to go through dialog and scenes. You’d expect that, at least after the first playthrough, you’d be able to skip scenes. Nope. You are forced to re-watch every single video.

Come on Wales Interactive, Lucas Arts identified this in the 90s and gave us F4, this is not acceptable now (yes, the game is from 2017, but 2017 > 1990).

So, this makes the game impractical to replay.

But well.. I may accept that if the game’s very well crafted.

As it turns out, most of the choices are cosmetic. I’m wondering if this is by design, given that the description says:

Forced into a brutal London heist

Forced being the keyword here.

The options they give you in one way or another always lead you to the same place.

When you do a play-through with diametrically opposed choices, you end up pretty much in the same place.

One would think that at least it has an influence on changing one or two story beats, but no, there’s no way.

Again, Forced is the keyword.

In addition to this, the writing has problems.

For example, there are moments where you hear the inner voice of the character. And it’s so obvious that the designer wants to tell us what to think that it breaks the suspension of disbelief.

It’s a bit sad because it’s clear that a lot of effort was put into this game.

Something I do want to highlight that felt very good:

FMV games usually have player choices inserted at the end of a scene, where the flow of the game stops, you choose, and it continues.

Late Shift puts the options with a timer during the scene process, and it doesn’t feel as clunky.

For this type of game, it’s a good design decision.

In fact, if you don’t choose anything, the game plays the canonical version of the story. I liked that.

Unfortunately, it’s a design decision that is less important than everything I mentioned before.

If you played Late Shift, do you have a different opinion? Can you think of other games that suffer from similar issues?

distance between player and designer

2024-01-24T00:00:00+00:00

There is a problem that I haven’t seen solved yet, and I don’t even know if it’s solvable:

There is a disconnect between what the player intends to do when choosing an option (horizon of intention) and what the designer has defined will happen when choosing the option (horizon of action).

This disconnect can be minor if there is a strong enough sense of construction and the designer knows with certainty what the player intends to do when choosing an option.

But it can also not be, and the player may have an intention that is opposed to what the designer thinks.

Many games use ‘flags’ to solve this:

SHOPKEEPER: Here's your bag sir.
 * Thanks for your service (sincere)
 * Thanks for your service (ironic)

In my opinion, these types of solutions are sub-par.

I don’t think it’s a good option to take the player out of immersion with these flags.

But at the same time, the risk of falling into a very large gap between intention and outcome is very high, and generally generates more rejection than a simple flag.

What do you think? Have you seen this problem solved in a different way?

a rhyzome of game narrative archetypes

2024-01-23T00:00:00+00:00

Did you see how many video games have a “character creator” that presents a finite set of hair, eyes, nose, face shapes, etc.?

What if we take this concept and apply it to the archetypes of our characters or NPCs?

We can generate a combinatorial explosion in our favor, and be able to create characters that work in unexpected but consistent ways in different narrative situations.

I think this could be implemented with some variables, like D&D does with alignment:

Chaotic / Neutral / Lawful
Good / Neutral / Evil

With 6 values, we get 9 very clear combinations of how characters behave.

What can we do if we apply this to archetypes?

I would be interested to see someone explore this… do you know if someone has already done it?

telepathy and narrative agency

2024-01-22T00:00:00+00:00

A few days ago, Tony Howard-Arias (writer of Slay the Princess) wrote an article talking about details of the game.

Some of the things he mentions are what I discussed in the list: the use of variables to reduce combinatorial explosion, that narrative is enough to make a game fun, etc.

But there’s something that caught my attention:

[…] we as game writers don’t have a direct pathway to our players’ thoughts, which put us in a difficult position where we could either ask our players for those thoughts (a big, immersion-breaking no-no in our book) or to find a way to vocalize thoughts that were at least tangential to whatever decisions the player has just made.

While Stephen King gave us the image that writing is a form of telepathy, when it comes to video games, misused telepathy can break the immersion.

The attempt to force the designer’s ideas onto the player is the main criticism that is made to graphic adventures as a genre.

That kind of telepathy, the one that takes away agency, is the one we have to avoid.

I hadn’t realized that The Voices of Slay the Princess were a design element to solve the problem of communicating ideas to the user without taking away agency.

Genius 👏

time off

2024-01-09T00:00:00+00:00

Hi fellow writers, I’ve been struggling this past week to keep the list going.

The main reason is that I’m not home, with limited time to write and think and with terrible internet.

I don’t want to clog your inbox with sub-par discussions.

I’ll pause the list until Jan 22nd.

Don’t hesitate to get in touch!

That said, it would be extremely helpful for me to get an email from you with your thoughts about the list up until now.

Let me know if there are topics you think I should be talking about, and if the list has been useful to you.

I have some ideas for the list, but I’d like to hear what you think/want first.

This is all about you in the end.

(see? playtest!)

🙃

writing endings that are not lame

2024-01-07T00:00:00+00:00

For some reason, many game writers tend to create good and bad endings.

If the player does the morally right thing, or what the writer wants, they get the good endings.

If they do something else, they get the bad ones.

I’m not talking about early endings, I actually like them if they are well written.

I’m referring to endings where the player succeeds in their objective, where everything ends well.

Slay the Princess laughs at this, and if you do everything the narrator tells you, you get the “Happy Ending”.

I really liked that they did that, especially since the most interesting thing in the game is navigating between trusting and mistrusting the narrator.

Why not use the opportunity of different endings to give more and better information about a complex plot?

Why not let the player make decisions and have the ending be the projection of those decisions?

Let each decision have a consequence, whether it’s good or bad, but not be morally judged beforehand?

One way to approach this is with rhizomatic foreshadowing, but you don’t have to be so meticulous.

It’s enough to have endings designed to not be so binary.

Players are already expecting things that go beyond basic morality of good and bad, success or failure. Let them eat cake!

interactive planners vs interactive pantsers

2024-01-06T00:00:00+00:00

Is there room in interactive narrative for the planners vs pantsers dichotomy?

With the narrative complexity presented by the rhizome, the discussion of planners vs pantsers becomes kinda outdated.

While there may be different types of systems to which planners and pantsers can orient themselves (rigid structures for planners, and perhaps more organic and random elements for pantsers), we will never succeed in finishing a game without planning.

At the same time, I am convinced that the best ideas arise from free writing, a practice that belongs mostly to pantsers.

Therefore, I believe that in interactive narrative, in one way or another, there needs to be a symbiosis between both approaches to writing.

That is why, if you identify with one of these groups, one should start listening and delving into the methods of the other, to see if this facilitates progress in your game.

But in the end, the key is always the same, even if it gest boring: test and playtest!

limits for ease of writing interactive stories

2024-01-05T00:00:00+00:00

Another strategy to tame the rhizome is to define limits.

We all want our games to be deep, and to give the player the ability to do whatever they want.

Perhaps soon we will be able to do this with LLMs, but for now the results are not so good.

For now, we have to accept that our ability to create variety is limited.

Defining the limits of variety from the beginning can help us write better stories.

How long do we want the story to last?
How many words should it have?
How many possible arcs will each player character have?
- And the NPCs?
How many locations does our game have?
- In how many ways can they be modified based on the player’s choices?
How many main or secondary conflicts will be addressed?

In turn, to prevent our games from getting out of control, I suggest minimizing all these numbers, setting low limits.

We can create a version with these minimized quantities, and only when the game fails, we can try to increase them.

For me, defining the limits even helps me maintain momentum and not lose motivation so much.

Thinking about writing a 200,000-word game is overwhelming, but knowing the main limits helps to tackle the necessary work.

dating sims are the new zombies

2024-01-04T00:00:00+00:00

Today I woke up with the sad feeling that all narrative games fell into two categories: mystery + detective or dating sim.

So I sat down to analyze the first results on Steam to see if this feeling was real or not.

To my surprise, not only was it not real, but I found some games with interesting proposals, obviously this from the marketing on the Steam page (I didn’t even consider the reviews); I haven’t played them yet, but I wanted to mention them here in case someone has played them and has any comments.

The first one is Needy Streamer Overlord, although I’m clearly not the target audience for this game, I found it interesting to use the need for approval from viewers as the main conflict of a game. Maybe I’ll play it, maybe I’ll watch it on YouTube, who knows.

Suzerain is a game about diplomacy and politics. Basically, the premise is: You are the president. It could be incredibly interesting or incredibly boring. I’ll have to try it out.

Milk inside a bag of milk inside a bag of milk was a hit in 2020. It’s a short story about how difficult things are in everyday life. I can relate.

Hira Hira Hihiru is a visual novel about a disease that brings the dead back to life. I’m not sure if it falls into the “mystery” category or not, but apparently it doesn’t.

MINDHACK has a very interesting premise: hacking the brains of the bad guys and turning them into happy people.

In addition to these games that I found interesting, there were 2 mystery games, 6 dating sims, and 3 RPGs on the list.

Yes, 6 dating sims 🤦. Evidently, Steam likes to flirt.

Emily Short's Dialogue Expressiveness in Mask of the Rose

2024-01-03T00:00:00+00:00

Today I took some time to review pending blog posts.

The latest post by Emily Short about the game Mask of the Rose caught my eye.

I haven’t played the game yet, but the article talks a bit about the systems they used to improve player expressiveness when making choices, and I found it interesting to share with you.

In the article, she mentions that you can define certain “traits” of the character (for example, if they are melancholic), and this enables a certain branch of options that you can choose in dialogues. The same thing happens with the clothes you wear (this is also central in Woods Folk).

In Dance of the Spirits I do something similar, setting character traits early on in the preface, and then reacting to that.

This is an example of defining certain variables and designing the options around them to facilitate exponential explosion: at each moment of dialogue, we go through our variables and make sure that all possible combinations are taken into account.

It is the basis of reaction to state and consistent narrative design.

In addition to this, she talks about a hypothesis generation system, in which the player can select various options and create their own hypothesis of what happened in the crime.

What I found interesting about this is that the player’s selection also enables dialogue options.

It seems like an excellent metaphor for confirmation bias.

Has anyone played it? Is there anything else notable?

It made me want to download it and play it in the next few days.

writing interactive games because we can

2024-01-02T00:00:00+00:00

Writing is telepathy. - Stephen King

What better advantage than the flexibility of words to create images in our minds.

The production effort for a video game to visually represent complex things can be prohibitive.

Words, on the other hand, are cheap.

With just a paragraph we can, in less than a minute, create visual images that would take hundreds of hours to draw, animate, etc.

Even with a static image, we can set the visual anchor and bring it to life with our words.

Let’s use that superpower!

I propose something to you:

Write a paragraph that can serve as a trigger for a new game.

But let it be a game that would be very expensive to produce.

Either because the art would be very costly or because the programming would be very complex.

For you, that is.

It doesn’t have to be something that a professional can’t do, or that Microsoft can’t pay for it.

Just write something.

Let’s start the year with something written.

You in?

Here goes mine:

I never thought that temporal paradoxes were so annoying.
I liked them so much in movies.
It was last week, I think.
I fell asleep.
And I woke up 10 years ago.
When I went back to sleep, I woke up 20 years in the future.
On and on.
Every.
Damned.
Day.
Just by breathing in the past, the future changes drastically.

Feel free to share yours!

happy new year!

2024-01-01T00:00:00+00:00

2023 was a somewhat eclectic year for me.

I started the year researching new deep learning techniques, to see if that was my next move in my career.

In April, we moved to Nova Scotia, Canada. We stayed there for 6 months and then we returned.

It didn’t work out, and upon our return, we had many problems with Hidden People Club, which ended up draining a lot of my energy for the year. I’m kinda shutting it down for now. Not sure what I’ll do with it.

This list came about as a way to distract myself: I wanted to spend more time thinking about interactive narrative and less about everything that was happening.

And it was one of the best decisions I made in 2023, along with returning to Argentina and buying a plot of land in Mar del Plata to build a bioconstructed house this year.

Because I enojoyed this so much, I want to grow the list in 2024, to initiate new and better conversations with all of you.

The list was born on November 11th and quickly became an activity that I really enjoy.

Not only did it help me synthesize many of the things that were swirling in my head, but it also allowed me to get to know many of you and establish new and very interesting relationships.

Thank you for being a part of this and I hope the list brings you as much satisfaction as it brings me.

Happy 2024!

See you tomorrow!

the cake and the rhizome in the wild

2023-12-24T00:00:00+00:00

From the same interview to Roadwarden’s Aureus I’ve been talking about these past few emails:

Have players surprised you in any way since the full game launched?

Some players grew attached to the tribes a bit more than I expected. They’d like to have an opportunity to abandon their main quest for good and simply stay in the peninsula, betraying the city and helping the Northerners maintain their way of living.

This is an example of what I usually say with let them eat cake.

Aureus defends the position that the player cannot abandon their main role to join the tribes.

I believe that if a reasonable number of players want this, it would be deserving of an update if it wasn’t captured in playtesting before.

I’m a firm believer in expanding the possibilities of narrative instead of listening to our own’s storyteller ego.

The interesting thing is that later in the interview, Aureus supports this (edited for clarity):

I think some video game writers hope to give themselves a chance to shine, striving to astonish others with their fantastic ideas, but they fail to leave any space for the player to carve their own way.

So I’m a bit confused.

On another topic:

I also didn’t expect so many people to start the second playthrough right after they finish the first one. It makes sense considering most players fail to complete many of the quests, even the major ones, during their first journey, and therefore some of the world’s mysteries remain uncovered. Still, I expected them to wait for a few months, or maybe a year or two, before riding back north. I wish there would be a “second playthrough” option that could cut the familiar introductions and description by half, or even completely, reducing the time spent with repeating text.

This, in my opinion, is a consequence of not having done enough playtesting.

Achieving replayability during playtesting sessions of a 30-hour game is complicated, I get it.

But still, replayability, in the modern video game industry, is one of the first things to take into account.

Apparently, Aureus did a great job with rhizomatic foreshadowing.

But he failed in the design of the experience by not considering that people need tools to navigate the rhizome without having to go through the same text over and over again.

Slay the Princess hides the options that you have already chosen in a previous loop, for example, to help you navigate through new paths.

It is essential that we design the replayability experience before launching our game, there is no doubt about this.

PS: I’ll take the next week off. I’ll resume the daily emails on January 1st. Happy Holidays! <3

gameplay and plot devices

2023-12-23T00:00:00+00:00

A plot device can make or break a story.

A well-written one can give you an unparalleled way to resolve a situation.

But a poorly written one, and there are many of these, can bring down everything you previously built, no matter how good it was.

That is why, as writers, we have to be especially careful when including them in our narrative.

And how is this affected when we talk about interactive storytelling?

As always, we come back to managing expectations: we have to be very attentive because a plot device can become a gameplay expectation.

In the interview I shared with you in the previous email, Aureus comments the following:

In Roadwarden, there’s an abandoned watchtower standing by the eastern road. I locked it behind wooden doors, but it made no sense to me that a player couldn’t enter it another way.

One of the options he added is that the player can destroy the door (a plot device that allows them to enter the tower).

Then he continues:

And since I make the door destructible, I then had to make similar choices whenever there’s another barrier of similar “destructibility.” One secret passage may be opened with a special amulet. But it can also be shattered with a set of tools if the player was to find them.

This level of attention is what interactive stories demand from us.

If a door can be broken, it becomes a gameplay element, breaking doors i a mechanic now. and we have to consider it in all the doors (literal or figurative) that we put in our game.

Did you feel betrayed by any game that built an expectation to not satisfy it consistently afterwards?

and it was exhausting

2023-12-21T00:00:00+00:00

This email is a reality check that, sooner or later, we have to face as game creators.

I’m going to quote Aureus, creator of Roadwarden, who put it very clearly in an interview:

I prefer to avoid a lot of preparation before writing, and instead discover and explore as I go—and that’s terrible for video games. I wrote quite a large chunk of the game having only a few vague concepts in mind, and that did wonders when it came to throwing new hooks at the player or adding more mysteries to the world. And as time went on, the more the stories in the game felt detached and nonsensical, with no end in sight.

I had to tell myself to stop and tie things together. I spent months updating not just the script, but also the code, the illustrations, and the game systems.

And then, indeed, work went much faster. Using my previous experience and the clear goal in sight, I started adapting old areas, writing [about] new settlements in a much shorter time, and adding more answers than questions. The game had its skeleton, I just had to add the flesh to it.

And it was boring.

Since I already knew what specific characters ought to do for the story, I just wanted to be done with them and get back to planning new, weird things. In other words, I got better results the more I worked against my nature. And it was exhausting.

I think that in one way or another, all creative disciplines have this situation, but in video games, at least for me, it’s brutal.

Especially because it can last for years if the video game is big enough.

This resonates a lot because I’m currently in that moment with Dance of the Spirits: things are taking shape, but it takes a brutal effort to get through the boring parts.

And well, not everything can be fun and games.

To not end it in a sad note:

If you’re in this situation, get together with other people, play your game with them, a little praise/feedback helps move things forward.

character arcs

2023-12-20T00:00:00+00:00

A tool to ensure consistency in our story is character arcs.

In traditional narrative, we usually design a single arc, and our characters evolve with it throughout the story.

Something we can do to facilitate the design of our interactive story is to define a manageable number of character arcs.

Then, we can define a set of variables that serve as decision-makers for the character arcs and design the options we give the player to modify these variables.

In this way, we ensure:

1) Agency: we are giving the player the concrete possibility of modifying the world in an important way. 2) Consistency: if the options modify the variables in a coherent way, the reaction of our characters and their arcs will be consistent. 3) Variety: it is easy to realize if the amount of variations is varied enough or not. If we see that different arcs are very similar, we can work on differentiating them more.

Undoubtedly, writing multiple character arcs is a challenge, which is why it is best to start with two or three, and gradually increase the possibilities as the game requires it while playtesting.

writing npcs

2023-12-19T00:00:00+00:00

In the area of character development, writing interactive storytelling opens up an immense design space.

I’m going to focus on the characters that are not our player’s, which is known as NPC (Non-Playable Characters) in game design jargon.

A traditional way of writing these characters is by defining their motivations and personality.

Once we have this, we can easily derive their behavior and dialogue with greater ease and consistency.

When we transform our stories into something dynamic, this principle can be modified in two ways.

The simplest way is to react to the state.

Our NPCs have pre-defined motivations and personalities, but their behavior changes according to the state of the game.

An example of this is when we design a zero sum reputation system.

But it doesn’t have to be as big at that, we can create small branches where different things happen, or modify the probabilities of the NPC saying or doing certain things based on the player’s previous decisions.

The other way in which the principle can be subverted is by not fixing the NPC’s personality and goals, but transforming them into variables and having all their behavior derived from these variables.

An example of this in Dance of the Spirits:

# start
  [fede]
    [(50%) fede_honest]
      set fede_liar false
    [(50%) fede_liar]
      set fede_liar true

At the beginning of the game, I define whether Fede, one of the characters, is a liar or not. There’s a 50/50 chance.

This will define many things that happen in the game and allow or disallow certain endings.

In this case, I am modifying Fede’s behavior based on a single variable for simplicity, but this could also be a combination of several variables.

the handstand walk

2023-12-17T00:00:00+00:00

Writing stories with multiple possibilities is leaving behind the uchronia.

It’s taking responsibility for all possible versions of the universe.

This is what Jorge Luis Borges posed as overwhelming in The Aleph.

It is no coincidence that this is difficult for us.

Our brains evolved with a strong illusion of the passage of time.

Of a single timeline.

Every time we try to work on multiple timelines or time loops, our brains are veering off course.

I’m writing this to convince myself that feeling overwhelmed while working on interactive narrative is more than expected.

It doesn’t say anything about how good or bad we are at storytelling.

This will be determined by the players later.

Telling solid stories with multiple paths is like walking on our hands.

We can do it, with a lot of effort and training. But it’s hard.

If writing is overwhelming you: scale down.

Fewer timelines, fewer characters, fewer variables.

There’s always time to expand later, when you have more trained muscles.

re: games & numbers

2023-12-16T00:00:00+00:00

Fellow list member Guillaume Ardaud wrote (shared with permission, edited for brevity):

One topic I didn’t get to ask you about that I’d be curious to hear your take on is that of the relationship between games, and the numbers they use to represent properties. It doesn’t matter if it’s your character’s health, how fast you move, how much damage your sword causes, your reputation with a faction in a RPG - it all boils down to a number in the system representing that property.

Most games show you the numbers as they are. If your character has 75 life points, you see the number 75 on the screen.

Very few things in reality work this way. The way my friends, coworkers, etc feel about me isn’t a -100 to +100 “sentiment” number - it’s a complex combination of myriad factors.

Some games will push those boundaries a little bit. It’s become popular in FPS games to not display any health related metric, but when the internal number gets too low, the character starts breathing heavily and the peripheral vision goes dark.

Another twist is to have numbers with little indication as to what they do - some RPGs will have a “luck” or “miracle” stat that you can increase, but without knowing precisely what it will improve.

If you have lots of numbers that aggregate to a single metric, you can have deeper, finer grained behavior - for instance, Dwarf Fortress doesn’t have a single “health” counter for dwarves; how healthy a dwarf is a complex combination of all the numbers that represent the state of its organs, etc, which interlocks nicely with lots of other mechanics.

I’m curious as to how you see that intersecting with your work around narrative design. Does having obvious, simplistic numbers (like “health”, or “reputation”) break narrative immersion?

Is there a point to hiding the numbers and their interactions if the players just end up figuring out how to read the game to tell what the underlying numbers are, and go to a wiki to make up for the game’s lack of transparency in the interactions between its numbers?

I appreciate this email as it touches a topic I haven’t talked about here and it’s a point of constant conflict I have with most game designers.

There is a trend in game design, which I believe gained strong momentum with the game Journey, that places great value on communicating things to players without text, or through the use of diegetic indicators.

I’m going to be harsh: I think it’s poor design to hold this opinion.

Why? Because I believe that designers, both narrative and game designers, should be open to exploring all possible spaces and find the best way to communicate and create ideas in the player’s mind.

There are many games where it’s unclear what’s happening due to pretentious design that fails to communicate, but succeeds in inflating the designer’s ego.

I prefer to focus on finding the best way to communicate, and the best way is to try all possible methods during playtesting.

Playtesting will tell you if a number is breaking immersion or not, and in general, in my experience, it’s more the mental constructs of “gameplay” rather than the numbers on the screen that break the illusion.

I know this is an answer that isn’t an answer, so I apologize, Guillaume, but at the same time I believe you’ve touched on a topic that doesn’t have an answer.

Each case is unique, and again, playtesting will tell you if, on average, a design decision is good or bad.

The important thing is to create concrete experiments to validate (or invalidate) those decisions.

playtest! playtest! playtest!

PS: Bahnsen Knights is out, and it’s very good! ❤️ Thanks Nico Saraintaris for another great game.

PPS: This is not sponsored, I just like his work so much!

enframing writers

2023-12-15T00:00:00+00:00

As a child, I used to repeat: “The problem is not [insert technology here] but how you use it.”

I had a lot of faith in humanity back then.

Many years have passed, and I understood that the tools we use shape our thoughts.

That’s why I am very careful nowadays in choosing which tools I use.

Beyond his controversial political orientation, Heidegger posed a very important thing: The primary danger is not in the specific technologies themselves but in the way their mode of revealing conceals other possible ways of relating to and understanding the world.

One consequence that matters to us from this is that, normally, writers do not have access to programming.

That is, they cannot create their own tools, and therefore are forced to work under someone else’s decisions, and above all, under those other people’s ways of thinking.

The tool you choose to use can be one of the most important decisions you make, and you might not even realize it.

(Dear reader: here’s the answer to why you can’t find me on social media, but I’m always an email away)

Heidegger said something interesting that we can use as a rule of thumb: It’s only when a tool breaks or malfunctions that it becomes “present-at-hand.”

If a tool becomes “present-at-hand” too often, should we replace it?

Is becoming “present-at-hand” an indication that the tool is trying to change the way we think?

overbored!

2023-12-13T00:00:00+00:00

Yesterday, I had one of those days where everything goes wrong.

I ended up having to format my computer and ran out of time to write the email. 😢

But well, here we are. Back again.

First of all, I wanted to mention that tomorrow Bahnsen Knights, a narrative game from a friend of the house, Nico Saraintaris, is coming out.

You might remember him from emails like “re: stretching & wildcards”.

There’s always something interesting in what Nico does, so I strongly recommend his games.

Returning to today’s topic, I wanted to talk about something I felt while playing Overboard! by Inkle some time ago.

According to its creators:

Overboard! is a tense, highly replayable detective game from the creators of 80 Days and Heaven's Vault.

It’s essentially a dialogue-based game where we have to save ourselves from being accused of murdering our partner.

I find it a bit hard to speak negatively about a game that I really wanted to like. Especially knowing the work behind launching a game like this.

But if we don’t criticize, we don’t learn, so I’m going to comment on something I felt playing the game that was interesting.

The game is based on a certain set of conditions (which character is in which part of the ship, who you talk to, etc) that unlock narrative situations.

The idea is that you play the same loop many times, choosing different options to save yourself from being accused of murder.

We have a certain amount of time before we are called to give our version of the events and it is decided whether we are murderers or not.

So far so good, the first runs feel interesting, but on the fourth or fifth loop, I felt that I was spending more time thinking about the mechanics strategically than caring about the story.

To such a strong point that I wasn’t interested in what the characters were saying, but I wanted to quickly get to the outcome to see if the options I had chosen were useful or not.

If the game was more mechanical, whatever, but this is a narrative game.

Not only that: the game encourages this kind of play through random “missions” it gives you in each run.

For example:

Can I make a thousand pounds?
Can I frame Clarissa?
Can I force Carstairs to help me?
What is Anders’ secret?

Smells familiar?

That’s right, another example of extrinsic motivation.

It makes me feel like the designers knew this game was not worth looping after 2 or 3 times that it needs to force you to find solutions that you wouldn’t care about otherwise, just so you keep playing…

When I realized what was happening, I closed the game and moved on to something else.

A shame.

The art is very nice and the story could be up to par, but the game design choices turned it into something so mechanical that the story faded behind the logic.

Did you play it? Was this your experience as well?

eat your own food

2023-12-11T00:00:00+00:00

It’s interesting how one can have strong opinions, talk and think a lot about something, and not apply it to their own creations.

Yesterday, I was playing Dance of the Spirits a bit, the game we’re working on, and I noticed that no matter what you did during the game, the final dialogue was always very similar and didn’t have much connection with what was happening before.

So, I added a small system:

Fede, the last character you talk to, is on the stage playing the harp throughout the game while you talk to the rest of the ghosts at the dance.

I decided to randomly integrate Fede into this situation.

Every time you go to talk to a character, there’s a chance that Fede is watching you. There’s also a chance that the game will notify you of this.

If Fede is watching you, he’s going to “pay attention” to your conversation, and if you empathize with characters he doesn’t get along with, it will lower your reputation with him.

The opposite will happen if he sees you talking with the characters he gets along with.

Abril, your character, can use this to their advantage…

At the end of the game, depending on the reputation you have with him, and what you have found out, you will have certain options and certain reactions from Fede, enabling different endings of the game.

I thought it would be interesting to tell you this so you can see how a single decision integrates several things I’ve been talking about:

Randomness
Reaction to state
Designing around narrative needs

What do you think of this example? Any suggestions on how to improve it?

probability and state

2023-12-10T00:00:00+00:00

Today I was thinking about the origins of cuentitos.

We wanted to make a game with a system of random events, but knowing myself, I couldn’t allow this system to have the same “error” that many others do.

I needed a narrative coherence that I didn’t see in other roguelike games.

For example, Monster Prom has a random event system where the things that happen affect the final outcome in some way, but they are very disjointed from each other.

Another game that suffers from this problem is FTL, where each event has a small narrative connotation, a result, and that’s it. There is no connection between events.

Maybe people who enjoy more systemic games don’t have a problem with this, but I, who enjoy stories more than pressing buttons, do.

What I found was that none of the tools available allowed me to create this system easily.

To satisfy my needs for probabilistic management of events and reaction to the state, I had no choice but to create my own language.

Towards the end of the development of the current version of cuentitos, Roadwarden came out, a text RPG that implemented many ideas similar to what I did, but they did it in Ren'Py, a tool for writing graphic novels made in Python.

I don’t want to hear how much work it was to create the content for Roadwarden, I’m eager to track down the creator to geek out about this (if any of you know Aureus, an intro is welcome).

The game lasts more than 30 hours, so doing a very deep analysis could take a lot of time, but I’m going to see if I can do something soon.

Going back to cuentitos, I’m eager to set up a web platform so that anyone can create dynamic stories, with probability and reaction to the state in a simple way.

If this is something that would interest you, please respond to this email, I’m going to start making a list of beta-testers to start testing the platform.

Until tomorrow!

intrinsic vs extrinsinc

2023-12-09T00:00:00+00:00

In yesterday’s email, I mentioned that Slay the Princess falls into a case which I consider to be extrinsic motivation: when the princess in her spiritual state tells you that you have to bring her 3 vessels.

This is what’s known as a “fetch quest”. A request from a character for us to go find something and bring it back.

I am not against fetch quests (or not completely aganst them?), but basing an entire game’s progress on “bring me 3 of something” feels a bit lazy, especially when other parts of the narrative are so good.

Intrinsic motivation comes from an internal desire of the player to do something; this can be a question they want to answer, or simply an interest in seeing how the story unfolds, what consequences their actions may have, etc.

Extrinsic motivation is when there is something external that makes us want to advance. A fetch quest is an example of this: the player has no interest in obtaining that item, except because an external agent asked for it.

Another example could be completing an “Achievement” in a game.

Slay the Princess has many achievements, and just seeing them there can make a player more eager to replay, but it’s not the game that is generating this motivation, but the list of achievements.

For me, intrinsic motivation is much more valuable, it’s the holy grail of keeping the player captivated in our story.

But when we’re in trouble, and there’s little time, planting extrinsic motivation is a tool we always have at our disposal, especially for replayability.

slay the praise

2023-12-08T00:00:00+00:00

During this week, I’ve been analyzing various aspects that I think Slay the Princess did very well and found interesting to analyze.

Today, I’m going to take a different approach and try to leave some notes about two things I didn’t like much.

While playing, I experienced a strong dissonance with how the game implemented progression.

The first time the princess is absorbed by “the hands,” she tells you that you need to “bring” three vessels (this is how the game refers to the different incarnations of the princess we encounter when making certain decisions).

Why don’t I like this decision of the 3 vessels? It feels too gamey.

Promising me progress if I go through the loop 3 times.

It feels as if the designers think their loops aren’t interesting enough to go through them anyway.

It’s almost an extrinsic motivation (I’ve never talked about this, but I will), and extrinsic motivations break my suspension of disbelief.

The other thing that struck me quite a bit and took me out of the flow was when the princess is in her “spiritual” form, the writing style suddenly becomes very philosophical.

Not only that, your responses often seem out of tune with the writing altitude that the princess has.

I believe that both your character and the princess should have a similar progression, because, after all, both are being profoundly transformed by the experience.

However, your level of dialogue becomes out of sync with that of the princess in her spiritual mode.

This reminds me a bit of Evangelion, where something similar happens.

There’s a trend that believes that by writing things with a spiritual tone (already written a billion times), the game gains “depth”.

I don’t share the feeling.

I think these two are the things I liked least about the game, and this is where this series ends. Starting tomorrow, onto another topic.

Do you want to tell me what you thought of this series?

Is there any game you would be interested in me analyzing like this?

slay the state

2023-12-07T00:00:00+00:00

Slay the Princess takes reacting to the state very seriously.

Practically every major action you take has an impact on the story.

There are very subtle and interesting examples.

For instance, it’s not the same to attack the princess as soon as you enter the room as it is to try attacking her after you’ve been talking with her.

The game reacts to the fact that you take the time to talk to her.

Unfortunately, the game only uses boolean variables (true / false) to define the state. (except at certain times when it considers how many times you ‘revived’, but it’s not very relevant).

Despite this, it has so many other strong aspects that the fact we are just following one path or another doesn’t come across as a disadvantage, but undoubtedly it is a structure that has been seen a lot.

Narratively, I think the greatest value of this game lies precisely in its reaction to your actions.

I believe there isn’t a single action that doesn’t generate some kind of reaction, whether it be at the level of the story, or with the Achievements.

If I had to choose one thing to recommend about this game to any writer, it would be this.

Players expect the game to react to their actions, the more dramatic the action, the more dramatic the reaction they expect.

We should never forget this, let’s avoid the frustration of giving options that have no effect, please.

slay the repetition

2023-12-06T00:00:00+00:00

Today I want to talk a bit about content repetition in Slay the Princess.

As the game is based on a very repetitive loop, the risk of quickly boring the player is very high.

Slay the Princess did not fall into this problem, and I believe the greatest production effort is focused on solving it.

To start, the story itself is based on the fact that each important choice one makes changes both the princess and the player.

There are 10 distinct princesses:

The Eye of the Needle
The Stranger
The Beast
The Nightmare
The Damsel
The Spectre
The Tower
The Razor
The Witch
The Prisoner

Each of these princesses has their own visual style as well as in the tone of the dialogue.

Additionally, at the end of each princess’s story, there is a closing instance where some philosophical aspect of being is exposed.

On the other hand, each chapter also modifies the player through “the voices”, of which there are also 10 (in addition to Hero which is the first one we start with):

Voice of the Hero
Voice of the Broken
Voice of the Contrarian
Voice of the Cold
Voice of the Cheated
Voice of the Hunted
Voice of the Opportunist
Voice of the Paranoid
Voice of the Stubborn
Voice of the Smitten
Voice of the Skeptic

Each of these voices is a personality that comments on your choices and what the Narrator says.

Thanks to these two variations in each chapter, the feeling of repetition posed by the gameplay loop is reduced.

By limiting the options to 10, the game’s creators tamed the rhizome.

They controlled the consistency, non-repetition, and novelty of content with 10 variations (both of princesses and voices) and managed to create a sufficiently large and interesting gameplay space.

The separation of novelty between the princess and the voices also generates a positive combinatorial explosion: there are chapters where the princess repeats but the voice changes, and this generates enough variety in the dialogue to not need another princess.

This is a basic tool of game design, we’ll discuss it in another email.

slay the numbers

2023-12-05T00:00:00+00:00

In yesterday’s email, I did a small analysis on how the elimination of the moral dilemma in Slay the Princess manages to solve the problem of player agency.

Today, I want to focus on something I mentioned in that email: the branching in Slay the Princess feels sufficient to justify the game.

This caught my attention.

The game has no mechanics other than choosing options.

Moreover, there are people in its community asking for a printed version of the game in a Choose Your Own Adventure format.

That’s why I’m interested in analyzing the numbers and its structure beyond its content, there’s no big idea behind this email, beyond: mh… so with these numbers a branching game can work.

Obviously, there’s much more to STP than this. But as a starting point, it’s great.

Something to note: I haven’t seen among the reviews a significant number of people saying that the game is too short, or that it lacks enough content.

The critiques have more to do with the writing itself than anything else. Later on, I might do a more concrete analysis on this.

The game has approximately 150,000 words and 2500 images.

This is not little, but it’s also not enormous. It’s a scale of writing that one person can handle. This is good news.

The same with the art, it’s a lot, but it’s also a scale that one person can tackle.

Why do I emphasize this: in the last three years I had many problems getting a publisher for my games because I didn’t want to grow my team beyond three or four people. Here we have a game that sold over 100,000 copies made by two. 🤦.

The first chapter is always the same, the second, third, and fourth use the same loop, but there are things that change according to what you did in the previous chapter.

I didn’t take the trouble to check everything myself, but between the diagrams they made on Reddit and the Fandom wiki we can more or less see the extent of the game.

In total, there seem to be 23 chapters: chapter 1, 10 chapter 2s, 10 chapter 3s, and 2 chapter 4s. (again, I didn’t check this with the game)

All chapters are designed around the game’s concept: to kill or not to kill the princess, which simplifies its writing quite a bit.

Obviously, the text and images of each chapter change, but it’s interesting to see how having a solid concept, it’s not necessary to vary mechanics or things that take a lot of time: it’s enough to vary the narration and support it with images.

Sorry for this frayed email, these are many data points I gathered and my head is processing to see if there is any structure underneath that we as writers can take advantage of in our games…

Did seeing these data spark any ideas for you?

slay the ethics

2023-12-04T00:00:00+00:00

Slay the Princess introduces us to the game in medias res:

You’re on a path in the woods.

And at the end of that path is a cabin.

And in the basement of that cabin is a princess.

You’re here to slay her.

If you don’t, it will be the end of the world.

The main conflict of the game is presented in the first 5 lines of text.

You have to decide whether to kill the princess or not.

Now, it would seem the game poses a moral dilemma.

Logic would suggest that you make a decision and execute it.

You either save her or kill her.

But Slay the Princess takes a different path: both the narrator who asks you to execute her and the princess herself are not trustworthy, something you quickly discover in the game.

The effect this has on your position is interesting: by making both characters unfriendly, you have no preference for either.

By quickly eliminating the moral burden, it also removes expectations.

As players, we don’t have much idea about the consequences of the choices we make.

That is: we don’t really have agency, beyond choosing this path or that one.

This means that morality is left behind and one is limited to exploring and dealing with the consequences of this exploration.

We are blind.

But it doesn’t matter, because we have no preference for any path.

As long as what we choose brings us something new narratively, we’re interested.

Slay the Princess functions more as a narrative fog of war than a moral dilemma, and this makes it very interesting.

This is one possible way of dealing with the player agency issue:

Remove it with style.

It left me thinking about what other techniques could be used to generate a similar effect, where the horizon of intent becomes more exploratory than interventionist.

Any ideas or references that resonate with this?

slay the princess

2023-12-03T00:00:00+00:00

When a narrative text-based game sells 100,000 copies, one must stop and analyze it.

And when more than one subscriber mentions it in less than a week, one must stop and analyze it.

I believe that to be a hit, you need luck, and luck can’t be designed.

But I also believe that certain things need to be well-done, otherwise, it can’t make it.

It’s a combination of skill and luck.

This means that I don’t know why Slay the Princess did well.

No one can know that.

It’s like knowing the shape of an ice cube from the puddle it left after melting.

What I can say is that they satisfactorily resolved the major problems of today’s narrative games, and this is worth analyzing.

To avoid dropping a spoiler in the first email, be warned: there are spoiler risks.

Although they will probably be kind of meta, as they will emerge from analyzing the game’s structure.

If you haven’t played it and want to, don’t open emails starting with ‘Slay’ in the coming days.

Come back after you’ve played it.

If you don’t mind spoilers, read them; they won’t ruin the experience either.

The first thing I want to address today is that I often say that branching structure is no longer enough.

Slay the Princess is a branch-based game. There’s no randomness or anything strange.

And it’s enough. Or at least, I felt it was enough.

So I eat my words. At least in this case.

There are elements that we will discuss that make Slay the Princess a very interesting experience.

What interests me most is that we can find very specific answers to the problems I mentioned in the list before:

Replayability
Content repetition
Player agency
Reaction to the state
Taming the rhizome

In the next emails, I will analyze each of these problem areas and how they are resolved in Slay the Princess.

Hope you enjoy it!

a playtesting session

2023-12-02T00:00:00+00:00

I am seriously thinking about creating a playtesting session like this:

1) Give the player my game. 2) At a random moment, stop the game. 3) Ask them: What do you know about the story? What do you think is going to happen?

And what if what they think is going to happen is better than what I had in my head?

Let them eat cake

Have a great weekend!

frustrating results

2023-12-01T00:00:00+00:00

Something’s been on my mind lately, summarizing some of the things we’ve been talking about:

When writing narrative games, we have a unique opportunity to explore the intricacies of the real world.

We can work complex moral and ethical dilemmas into our stories, we can create experiences that are thought-provoking and reflective of the nuanced world we live in. Or not.

For instance, think about the tough choices some people face in resource allocation in real life, like in healthcare or during a natural disaster.

That’s something we don’t experience daily, unless we work on those areas, and many people might find that interesting.

Many video games default to violence as the primary conflict.

It’s the easiest thing to do, obviously.

Two opposing sides, war, and that’s enough.

Many of us are tired of this and are looking for different ways of interpreting conflict.

The selection of options opens the possibility of generating a type of conflict that does not exist in linear media, but is very strong in human experience.

And I’m not referring so much to moral dilemmas, which are great to explore.

I’m referring more to the conflict that arises between making a decision and the outcome that follows.

This is slightly related to Brian Upton’s concept of the horizon of intention and horizon of action.

The intention that the player has when choosing an option, and the action (or result) can be different, and a great source of conflict.

Obviously, we need to be careful and subtle with this, it’s not about removing the player’s agency, but about exploring the frustration that things turn out to be more complex than one imagines as a design goal.

Have you played any games that perform this? I think I have not, but I’d love to…

taming the rhizome

2023-11-30T00:00:00+00:00

Game designers have a basic design tool: the mechanics.

These are things that the player can do to change the state.

A typical example is “jumping”. Or “running”.

In narrative games, “choosing an option” is a mechanic.

A very common practice for designers is to choose a mechanic and design around it.

That is, to explore all the ways that this mechanic can be used and that are interesting.

Once they exhaust a mechanic, if it’s not enough to sustain a game, they explore another mechanic and eventually the combination of both.

This strategy is precisely to solve a problem that we share when writing dynamic stories: taming the rhizome.

Game designers could create endless rules for special cases, and these rules generate first and second order consequences that greatly complicate the design.

Many games suffer from this problem and become incoherent.

If we write without thinking too much, and we constantly create ad-hoc variables to represent such and such a state, it becomes very difficult for us to maintain coherence throughout our story.

It’s already difficult to be coherent in a linear story, imagine if there are 120 variables that we have to keep in mind at all times.

That’s why it’s a good idea to sit down and think: what is the minimum set of variables that I can have in my story, so that it’s not too difficult for me to reason about it but at the same time it’s interesting for the player?

In Dance of the Spirits, I’m experimenting with 3 very important variables, which I’m using for the definition of the story and the consequences of actions, and 5 or 6 that are less relevant.

The important variables are defined quite early in the story, one is randomized and the other two are tied to some player decisions.

The less relevant variables have more to do with enabling or disabling options to the player according to the actions they took, but they do not significantly affect the story.

My estimate is that the game is going to have about 35,000 or 40,000 words.

I’m still not clear if the 3 variables are enough to generate something interesting, but I think they are.

I share this to start identifying some kind of parameter of what might be needed depending on the size of the story.

If you have worked on a narrative game that did something like this (designing the story around a bunch of variables), I’d love to hear your experience! Reply to this email and let me know.

rhizomatic time loops

2023-11-29T00:00:00+00:00

Another game that successfully managed to use rhizomatic foreshadowing is 12 Minutes by Luis Antonio.

This is a puzzle game whose main mechanic is narrative, making it a very interesting subject of study.

The game is designed around executing a series of actions with a time limit (12 minutes).

After that time you’re forced to go back to the start.

Forcing you to play it over and over again in increasingly creative ways to get more information.

You can try different dialog lines, and perform different actions to unlock subtle branches that slowly start to paint the picture of what’s going on.

That might already be enough to make a whole game, but the designer took it a step further: the character remembers everything he does in each loop, same as the player.

This works very well for two reasons:

It makes the actions you take believable. Why would the character hide in a closet if he doesn’t know someone is coming for him?
It enables additional paths that only get unlocked after you got a piece of information.

This last point is very interesting to me because it’s a form of reaction to the state that (beyond Groundhog Day) I had not seen executed so well in video games.

It seems to me that there is a lot of room to explore with temporal loops beyond 12 Minutes.

I feel there is an entire genre to be explored there.

Do you know any other game that uses temporal loops as a way to control the exponential explosion of the rhizome?

I would love to play other similar experiences that are not necessarily detective-oriented…

rhizomatic foreshadowing

2023-11-28T00:00:00+00:00

Today I want to talk about one of the writer’s best friends: foreshadowing.

The premise is simple: subtly hint at things that will happen in the future of our story to capture the reader’s attention.

In linear narratives, it is generally used to create suspense, manipulate the questions the reader asks about the story, or simply provide colorful details that are reinterpreted with the important events of the story.

It can be subtle or even meta with this, as is the case with Ted Chiang in “Story of Your Life,” from the beginning:

Your father is about to ask me the question. This is the most important moment in our lives, and I want to pay attention, note every detail.

We still don’t know anything about what’s going to happen, but the author is already giving us clues that something strange is happening with time. It’s not moving linearly.

When we know nothing about the story, these lines are confusing. But in the end, this makes a lot of sense.

Then we have foreshadowing that is better forgotten, as in the entire work of the boy wizard student from She-Who-Must-Not-Be-Named, where the professor named Lupin ends up being a werewolf 🤦.

Now, foreshadowing works the way it does because of one property: the linearity of the medium.

What happens when we introduce non-linearity?

Foreshadowing as we know it breaks down.

Unless we use it in a linear way, forcing events to happen, we have no way of knowing if such and such a narrative event will occur in the future.

This presents both a difficulty and a unique opportunity.

If at all times we are aware that we do not know what paths the player will take, can’t we use foreshadowing to pose questions that may or may not be resolved depending on what happens and make this an element for replayability?

This is exactly what Flavourworks did in the game Erica (the game used to be a PS4 exclusive, but they released it on PC too, yay!).

The game is an FMV with branching narrative, nothing surprising there.

What caught my attention is how they managed to make it so that when choosing a path in the game, there are questions posed that as a player you intuit will be resolved in other paths.

This presents a very concrete motivation to play the game again. The much-desired replayability.

I am terrible. If you don’t give me a solid reason to go through the story again, I won’t.

That’s how I played games like Until Dawn or The Quarry only once, they didn’t generate enough interest for me to return to the story.

But Erica did it so well, that it not only left me with posed questions: when going down another path, it answered one question and even reinterpreted things that had been told to me in the initial branch.

This made me want to explore all the paths to achieve an understanding of the entire story.

(They failed in the game design front, by introducing frustrating interactions, but that’s another topic)

I’d love to hear your opinion if you play(ed) it.

going back to the start

2023-11-27T00:00:00+00:00

While one can read a book (or watch a movie/show) infinite times, its design is oriented for one to do it only once.

The “plot twist” is used up. The cliff-hanger is resolved.

Video games can also be designed this way, but the expectation is that one can play them again.

This property, in the game design jargon, is called replayability.

Why do we care about replayability?

There are expectations from the players: most associate $ with hours of gameplay. And playing again is a good way to increase hours of gameplay.

And also from those who produce games: it is so costly to generate content that creating branches that players never see is a waste of time and money.

I’m not saying we have to force players to replay our games.

It’s all good with a play-through being self-conclusive and the player not wanting to play more.

What matters most to me is the arrow in the other direction: what happens when the player wants to explore our story through another path.

This is where most game writers fail.

The reason is simple: doing a good job requires thinking like a game designer and not just a writer.

One of the most common problems is that the player has to re-read a lot of the same text to go through alternative paths in the story.

No options are provided to facilitate this (skip text, shortcuts, etc).

This demotivates the player, leaving much content unseen, and a negative feeling about something that, if better designed, could have been a great experience.

It is the responsibility of the writer to ensure that the narrative design favors replayability.

This includes working with programmers and game designers so that the necessary features for this are developed.

It’s not an easy job.

But nobody said it was easy, right?

Tomorrow I will write about a technique to address this that I’ve been thinking about for several years.

And I hope we can figure out more solutions, because it’s an open problem.

Is this something you’ve experienced as a player?

What other problems do you identify that a writer might have to ensure the replayability of their stories?

Do you have any interesting examples of how a game resolved replayability?

re: stretching & wildcards

2023-11-26T00:00:00+00:00

Today I wanted to share a small article written by my friend and fellow subscriber Nico Saraintaris.

For a bit more background, Nico and I worked together a few years ago on Bloody Service, a game in which I developed a visceral frustration with the state of narrative engines (we used Ink for the game), which triggered the creation of cuentitos.

Nico is the creator of LCB Games, along with Fer.

They have been working on a series of games within a genre they call Pixel Pulps.

On Friday they published an article about their creation process. If you notice the article’s name is similar to this list: yes, we are both a bit nerdy about Stephen King, what can we do.

In the article, they introduce two concepts: stretching and wildcards.

About stretching:

In the Marvel Method, Stan Lee cannot stretch ad infinitum the text of a panel. Space is finite. But we? We can! We can keep a single image on screen and render as much text as we want. This process, which I call Stretching, allows us several things from a narrative point of view: to manage the reading pace, to build expectations, etc… But, more importantly, Stretching allows us to let players see Fer’s work on screen for longer!

This is a great example of how we can use a property of interactive narrative that we don’t have available in printed narrative (although we do in audiovisual).

While stretching is a good concept to dwell on, what interests me most personally are the wildcards.

Nico explains that, when he’s writing the outline for the game, he sends it to Fer with some notes, and details that are not fully fleshed out.

Fer then sends Nico some artwork based on the outline:

Once I receive Fer’s storyboard, that’s when things get interesting. The detail appears as an unavoidable presence, and many times I wonder why I chose what I chose, or why Fer chose what he chose. Wildcards is letting go of the steering wheel for a while and then pick it up and straighten the car. (emphasis mine)

So. Here’s what’s interesting to me.

Nico has long said that games are inevitably unique and that one shouldn’t worry about doing “the same” as others, that each game has its touch because the people making it have a unique combination of interests and backgrounds.

I think this is the first time Nico has publicly talked about a concrete implementation of this idea in his production process.

What I find interesting is that the synergy they created between the two of them, in terms of writing, is nothing more than an injection of probability.

What do I mean by this? When one writes a lot, things tend to converge to a more or less stable point. What Nico and Fer achieve with their process is to add a point of randomness that undoubtedly makes the game’s outcome better.

This also gives Fer a level of authorship that is not usually present in narrative video game projects, where the writers have all the control and the artists are limited to drawing.

As the late Sophie 💔 would say:

If you need that something, but don’t know what it is Shake, shake, shake it up and make it fizz

Great article, Nico, thank you so much for sharing your process 👏.

The interesting thing is that Nico developed this process because he has the expectation that whatever Fer introduces will create a better experience.

Can we do the same thing, but within our stories?

silent evidence

2023-11-25T00:00:00+00:00

I came across the concept of silent evidence a few years ago while reading Black Swan by Nassim Nicholas Taleb.

The main idea is that humans tend to narrate the things that happen, explaining them with chains of cause and effect that we take for granted.

Many times we predict incorrectly because of these narratives we take for granted because we ignore the presence of evidence that would make us change our minds.

This hidden evidence is what we call silent evidence.

He uses it to justify his idea that we cannot predict.

In his words:

Silent evidence pervades everything connected to the notion of history. By history, I don’t mean just those learned-but-dull books in the history section (with Renaissance paintings on their cover to attract buyers). History, I will repeat, is any succession of events seen with the effect of posteriority.

As writers, we often, sometimes unconsciously, use silent evidence to our advantage.

All information that we as authors have and the player does not, creates a very important area of work for the generation of meaning.

What does the player know at this point in the story?
What do I know that he doesn’t?
Are there opportunities for foreshadowing?
Or better: are there opportunities for probabilistic foreshadowing? (more on this in another email)

These are some narrative design questions we can ask ourselves at any point in the game that are directly related to silent evidence.

In what ways have you used or you plan to use silent evidence in your stories/games?

Have you played any games that use silent evidence in novel ways?

I do, but that’s for another email, sorry. 🤭

open re: playtest! playtest! playtest!

2023-11-24T00:00:00+00:00

Fellow list member Kamran Ayub (who maintains and works on Excalibur.js and runs Keep Track of My Games) shared an interesting thought about yesterday’s email:

One thing it seems like narrative games would benefit from playtesting is for inclusivity. Your game could be played from all types of people from different cultures different to yours, so you may unintentionally introduce language or elements that don’t land how you intended.

This.

So. Much. This.

Inclusion should be a priority for us.

It is practically impossible to have a complete view of where we are being non-inclusive.

At least not without many years of experience in the subject.

Clearly, playtesting is a good tool to identify inclusivity issues.

I’m not going to delve into the ramble of cultural Marxism, what matters to me is making the world accessible and inclusive, which is why I was excited about the response.

Although Kamran is focusing on content issues (e.g., whether a word or another might be offensive, etc.), I believe this is also true in matters of form.

For instance, a children’s book has less text than a book for adults.

This is true in video games as well, the length of the texts is greatly reduced, we will talk about this in an upcoming email.

And not to mention those of us with ADHD.

There are games I cannot play because of the amount of information that needs to be managed at the same time without getting distracted.

This speaks to the importance of playtesting with a variety of players.

It’s clear that this extends to all types of playtesting, not just narrative, but it’s good to emphasize it.

Thank you, Kamran, for your comment!

Question:

Do you know any games that would have benefited from having done narrative playtesting on something in particular? Why do you think that?

playtest! playtest! playtest!

2023-11-23T00:00:00+00:00

Game designers have been banging their heads against the wall for years with closed-door creation.

In fact, this happens to a greater or lesser extent in all creative disciplines, but in video games the consequences are brutal.

What do I mean by creating behind closed doors? This:

I have an idea for a game. I think it’s excellent, so I start creating it.
I spend 3 years creating the game I want, it’s perfect.
I launch it on Steam and it receives bad reviews, so bad that my game is categorized as “Mixed” and therefore Steam stops showing it to new players.

What happened? We designed behind closed doors. We never showed our game to anyone during the development process.

The process of testing our games with real players is called playtesting (emphasis on real).

The central idea of playtesting is to help develop our games iteratively and incrementally. The loop would go something like this:

I design and implement a part of the game.
I ask myself specific and general questions about this part of the game.
I do playtesting of this game segment with other people.
I allow players to give a general opinion and then ask them the specific questions I want to answer.
I collect the data, analyze it, and make decisions: is the game okay or does something need to change?

This may be an oversimplification of the process, but it more or less goes like that.

I want to propose here that narrative games are no different from other games, although they pose their own problems. I’ll leave you some questions here, if there’s one that catches your attention and you have opinions about it, please respond to this email and tell me!

What things are important to playtest when we talk about narrative games?
If narrative is about creating a mental model in the player’s brain, how can we test that? Are we generating the mental model we want in the player’s mind?
If we have branching or probabilistic narrative elements, how do we test that all paths/possibilities are interesting?
What do we do if a player tells us: “here you forced me to do this, but I wanted to do something else”?
How do I test part of my game if the stories need to be complete to see if they work?

I will be addressing some of these issues in future emails because they seem crucial: there’s a chance that one might write a book and people like it so much that it doesn’t need to be modified, but, in my experience, this doesn’t happen with video games.

open response about yesterday's email

2023-11-22T00:00:00+00:00

A fellow gamedev friend from the list, Andreas Lopez (creative director and studio owner for Vaelynn) replied to yesterday’s email commenting about binary branching (shared with permission):

As a studio owner and fellow narrative focused individual, the biggest problem I see with your design choice of putting chance in these sort of outcomes is that it depends how you use them.

Some random, uneventful conversation? Sure, cool! But if it affects and gate keeps special items, quest lines, whether or not an NPC will join your party/base, etc. then all you will cause is people to do save scumming in order to get the outcome they want.

Regarding your comment on save scumming (saving all the time), perhaps you are heavily influenced by RPG game design, and maybe what I propose is not yet mature enough to use in RPGs, which is a fairly rigid genre. But we already have other genres like rogue-likes where the player expects randomness as part of the system, and doesn’t resort to save scumming when they get a bad result. I believe this is more about managing player expectations than something innate in the players themselves. If they always expect to get a “good” result, yes, they will be saving all the time.

On the other hand, the game design can simply accept this, and allow the player to rewind to any point in the game. I added this functionality in cuentitos because I think it’s important to have it in case the game designer wants it.

Players don’t want realism, they want to escape into the world. Like in baldurs gate 3 they want perceived control and clearly understand of when they don’t.

For example they might need to make a persuasion check. They see the die roll and they see the outcome.

Many still do save scumming for these instances, but many are okay with moving on and letting it go. Why? Because they saw how they got there and lady luck wasn’t on their side.

If you hide the logic, then you’ll just frustrate people once they find out because they realize they could’ve save scummed.

Here, you make a point about how game design solves this problem in RPGs and how it manages player expectations.

Saving throws are a game device to incorporate randomness and for the player to expect that what they are doing is a gamble. I completely agree.

Roadwarden, for example, places a die to the left of the options that have random outcomes.

It’s part of the game design to ensure that the player understands that there is a bet or that they have the possibility to rewind. Frustration arises when things happen that the player does not expect.

I definitely believe in branching and I appreciate randomness. But the player always needs to understand how or why the outcomes happen as they are - especially if they have further consequences beyond a greeting.

Yes! The problem here is not probabilistic branching, but rather the player’s mental model and their expectations when encountering it. If a player encounters binary branching when they do not expect it, they will also feel frustrated. This is a problem of game design and feedback, not so much of probabilistic narrative.

And if they only affect the greeting then you have to save the result so that as you talk to the same NPC 10 times over that you don’t get 7 positive, 2 neutral and 1 ‘i hate you’ type of greetings. Just something to keep in mind.

This is so true, even with binary branchings: one has to have a consistent reaction to the state. We agree.

In my opinion, not being consistent is a bug.

I enjoy the overall concept you are designing here though.

❤️ I am very glad that this is the case!

Thank you very much, Andreas, for writing and helping me clarify these things.

the binary branch frustration

2023-11-21T00:00:00+00:00

There is something that frustrates me a lot: we are using extremely powerful computers to allow the player to choose between one path or another. Between two binary options, or as we programmers like to say, boolean.

Most tools like Twine make it very easy to write narratives with branches associated with choices made by the player, so it doesn’t surprise me that most writers choose this.

I am interested in advancing interactive narrative, which is why I created this list, and my language cuentitos (it’s not ready to be used by just anyone!).

What do I mean by binary or boolean branches? Look:

You see the high orc priest in front of you.
  He greets you and you see a smile drawn on his face.
    req reputation_with_orcs >2
  He looks at you with a hatred you have rarely seen in your life
    req reputation_with_orcs <-2
  He just looks at you, as if wondering what you want.
    req reputation_with_orcs >= -2
    req reputation_with_orcs <= 3

This example suggests that, based on my reputation with the orcs, one of the 3 possible paths will happen.

Which option is executed depends on what the player has done before reaching this moment, so it’s a way of reacting to the state, and we want that 👍.

This example gives the player 100% control over the outcome. Maybe the player doesn’t know it the first time they play, but their previous choices determined which option was chosen.

I think this is fine, but it limits the experiences we create.

I am particularly interested in having tools where we can play with the player’s agency, validating their choices, but not giving them 100% control over what happens in the world.

This comes from a frustration I have that goes beyond narrative:

Most of us associate the quality of our decisions with the outcome.

This is a serious mistake because we can make good decisions (Crossing the street when the light is red for cars) and still get a bad result (If a car doesn’t stop at the red light and hits us). Or vice versa.

I believe that if we consider that all experiences educate us, video games where the player has 100% control are reinforcing a broken pattern of reasoning.

That’s why I created cuentitos, to be able to write probabilistic narrative.

You see the high orc priest in front of you.
  (1) He greets you and you see a smile drawn on his face.
    freq reputation_with_orcs >2 +10000
  (1) He looks at you with a hatred you have rarely seen in your life
    freq reputation_with_orcs <-2 +10000
  (998) He just looks at you, as if wondering what you want.

In this case, as we saw in the article on zero-sum reputation systems, the 3 options can be chosen, only each one has a probability based on the state of the game.

This is much more interesting to me, because in this case the player only has a 90% control, 1 out of 10 times they will get a result that is not what would have come out in the binary version of the narrative.

What do you think about this? Do you think this is a good idea? Or do you think there are reasons why this shouldn’t be done?

game over?

2023-11-20T00:00:00+00:00

I just had a game over in a massive multiplayer game.

In this game, all players analyze a lot of information at their disposal, which is basically a rhizome of small ideas, opinions, and predictions that each one chooses to believe in.

The game has 3 stages where you have to select an option based on the information you gathered during the exploratory phase. These options open branches that change the next stage.

In the last stage, you have 2 options, and that determines the end of the game.

This time I got game over.

(I assume you know what I’m talking about, but just to clarify: The 2023 presidential elections in Argentina)

Now I started thinking, how can I channel everything that happened in the mailing list?

Here it is:

Today I was listening to some people who voted for the candidate I did not support, and basically, they were saying that they didn’t agree with most of his proposals, but they wanted a change. Something different from what they already know, because what they know isn’t working. They also say that anyway, all those things he proposes, he won’t do.

They are right that it’s not working, but I find it very interesting that they choose an option believing that something is not going to happen.

This reminds me of a game design concept proposed by Brian Upton in Aesthetics of Play (yes, the same book I mentioned a few emails ago).

He talks about the horizon of action and the horizon of intention.

The horizon of intention is the group of actions and effects that the player expects to happen or believes could happen.

The horizon of action is what the game effectively allows them to do.

In this case, the horizon of action was choosing between two candidates.

The most interesting thing is that the choice was made based on what each one interprets as the horizon of intention.

This intention horizon was fabricated, intentionally or accidentally, by everything that person has lived through up to that day and by the campaigns of the candidates.

I’m not sure where I’m going with this comparison, but I found this intersection between game design, narrative, and reality interesting, and I thought it could be a good starting point to think more in the future about the options we give to our players, the probabilistic analysis of why they choose one option over another, and how we can play with expectations and outcomes to create interesting narrative situations.

I hope the most interesting thing that happens for us is that Argentina begins a process of economic reconstruction where people don’t lose rights, but the truth is, as a society, we threw a d20 yesterday, and we don’t really know the probabilities of each outcome.

zero sum reputation systems

2023-11-19T00:00:00+00:00

We have already established that narrative rhizomes can become difficult to manage due to combinatorial explosion and the impossibility of deriving with total certainty the path that a player took to reach a given point in the story.

That’s why it’s crucial that we develop strategies and tools that allow us to encompass this problem, helping ourselves with the state for this.

One of these tools I refer to as zero sum reputation systems. It’s a very simple concept, but very powerful and very gamifiable.

The idea is that we are going to define a variable that receives a number, at the beginning of the game it will equal to 0.

set my_reputation 0

This variable can be positive or negative. And we are going to consider that at both sides of 0, there are 2 opposite poles in constant competition. For example, when my_reputation is less than 0 we say that it is our reputation with Elves, while in positive it is with the Orcs.

Throughout our story, the player will choose options that will modify this variable

There's an orc looking at you. You don't feel very threatened by him. Or her.
  * I ATTACK the damn beast!
    set my_reputation -5
  * I wave my hand and smile
    set my_reputation +3

And eventually we will be able to use thresholds to make decisions:

You see the high orc priest in front of you.
  (1) He greets you and you see a smile drawn on his face.
    freq my_reputation >2 +10000
  (1) He looks at you with a hatred you have rarely seen in your life
    freq my_reputation <-2 +10000
  (998) He just looks at you, as if wondering what you want.

In this case, we have that the first 2 options have a very low chance of being chosen (1 in 1000, that is 0.1%), but if the condition that the player’s reputation is balanced towards the orcs is met, the first option becomes the most likely (10001 in 11000, that is 90.91%). The second option increases the chances when the opposite happens.

DIGRESSION

In this case, I’m not eliminating the opposite chance, it’s just highly improbable. I’ll talk more about this in a future mail, as this is a special interest of mine (and the reason why I created cuentitos).

END DIGRESSION

In this way, we can handle decision making, even random ones, and we can control the rhizome a bit more: if we see that the first option came out, then we know that the player has a good reputation with the orcs, and we can use this information to continue navigating the rhizome.

I refer to this technique a reputation system because it’s easy to understand it that way, but it can be used for other things that are not reputation, the important thing is that whatever is defined works in a polarized way, where the positive value is one thing and the negative its opposite.

This system has been used a lot even in branching narratives, for example in games like Mass Effect, or The Age of Decadence for an indie title.

Can you think of other ways to tame the rhizome? I’d like to hear them :)

the rhizome is here to stay

2023-11-18T00:00:00+00:00

Since the introduction of the concept of storylet by Emily Short, the narrative design community had a concrete word to describe small narrative fragments that can be recombined by a narrative engine to generate unique stories and situations.

This narrative architecture, unlike more linear ones, integrates very well with gameplay.

It’s no coincidence that among the first messages on the mailing list were variables and state.

These two elements are the glue between gameplay and narrative, as they allow both systems to make decisions together.

If there is a narrative architecture that has ideal characteristics to combine with game design, shouldn’t we always use it?

Yes and no: we need to analyze the difficulties it presents, the cons, as it is not necessarily ideal for all types of interactive narrative experiences.

What we do need to understand is that, whether we like it or not, this is a reality that writers will have to get used to: rhizomes are here to stay.

interactively tangled

2023-11-17T00:00:00+00:00

I find it interesting that we stop to think a bit more about the rules that change when we move from traditional linear narrative to interactive narrative.

And, since we were talking about sense-making, there’s a big property of dynamic narrative that I think is nice to bring to light.

“What is writing? Writing is telepathy.”

― Stephen King

Sense-making is an incremental process.

In traditional narrative, we start from a “blank sheet” (to simplify, obviously we never start from zero), and little by little our words are building the ideas that as authors we want in the reader’s mind.

This same property (the controlled construction of meaning by the author) remains in place if our game has a linear or branching narrative.

In narratives with branches, the complexity simply increases because the author has to take into account all the possible paths of the narrative tree, but the process of sense-making has a unique direction: at any given point, we can go back and know exactly what the player knows about the story.

Maybe not their interpretation, but we do know what information was given to them.

The real problem starts when we deal with non-linear narrative systems.

Here the trees transform into rhizomes (see comment about Deleuze below) and it is not possible to backtrack what the player knows or does not know, because there is no pre-established order of the flow of information.

I feel this is very abstract, so I’m going to give a concrete example:

MOTHER: What do you want to eat?
  (20%) APRIL: Vegan Chili
  (20%) APRIL: Lentil Soup
  (20%) APRIL: Tofu Stir-Fry
  (20%) APRIL: Curry
  (20%) APRIL: Avocado Toast

[...] The next day

MOTHER: What did we eat yesterday?
APRIL: I have no idea...

In this example, I’m using the probabilistic functionality of cuentitos: the lines that have a percentage are part of a “bucket,” from which one is chosen, in this case, all have the same probability (20%), but they might not.

As authors, we cannot know what they ate the day before when the mother asks, because that is randomly decided during the game.

We could, however, define a variable and react to the state.

MOTHER: What do you want to eat?
  (20%) APRIL: Vegan Chili
    set dinner chili
  (20%) APRIL: Lentil Soup
    set dinner soup

[...] The next day

MOTHER: What did we eat yesterday?
  APRIL: Vegan Chili Mom, are you losing it?
    req dinner chili
  APRIL: Lentil Soup Mom, it was very cold.
    req dinner soup
  [...]

Rabbit hole alert: For those who want to get a bit woowoo with this, this is a manifestation of what the post-modern philosophers Deleuze and Guattari call the Rhizome in the book A Thousand Plateaus.

Have you ever encountered this problem?

What did you do to mitigate it?

about players making sense

2023-11-16T00:00:00+00:00

narrative in all its forms is nothing but an ultra-effective strategy for generating meaning in our brains.

humans tend to group knowledge and remember through the connection of ideas, and nothing is better than narrative to aid this process.

those of us who have designed many games throughout our lives are hyper-aware of this.

Brian Upton managed to synthesize it better than anyone in his book Aesthetics of Play.

a game is not merely the rules, or in the case of stories, the events that happen, nor even the dynamics and argumentative arcs.

no, half of the game is the player’s interpretation of everything that is happening: the process of meaning generation.

we have a great tool at our disposal if we can master semiotics.

exploration and iteration are the bases for finding unique things, but it is good to have a framework to guide these explorations.

next time you sit down to write a scene for your game, why not think about what the player knows at that moment and what they should know next,

then you can try exploring different paths to arrive at the same destination of meaning generation in their head.

don't run into the woods

2023-11-15T00:00:00+00:00

there are many things that can go wrong with traditional storytelling:

characters can be underdeveloped
the story can be predictable
we can screw the pacing
tone might be all over the place
meh conflict
poor dialog

… just to name a few.

but in games, oh my.

before games, you controlled pretty much everything. in games, you don’t.

in a book (or movie or tv show) the reader can either keep reading, or stop. that’s it. that’s as far as their agency can go.

well, that was true until choose your own adventure showed up, but still 99.99999…% of books are not CYOA.

when we write for video games, interactivity opens up a vast array of possibilities, but also a thousand potential problems.

the biggest issue is that the rules change.

not all of them, but enough so that an experienced writer may have to start almost from scratch.

therefore, it’s advisable to start small, beginning with what’s familiar.

there’s no need to start with a dynamic narrative roguelike RPG with reputation systems and randomly-generated NPCs based on modular archetypes.

it sounds like a great challenge.

but it scares me just thinking about it.

don’t run into the woods.

why not start with a linear narrative?

use all the techniques we already know and feel them in video game form?

have you heard of walking simulators? They’re just that, with one or two twists. Yet… ❤️❤️❤️❤️

I’m not saying you should show it to anyone if you’re not ready.

though it’s bad advice, if it helps you get started, it’s all good.

I’m suggesting that you write and feel it.

just like you did with your novels or short stories without even realizing it.

there’s time to gradually give the player the agency they deserve.

there’s time to get more complex and work around the pitfalls, one by one.

and remember King Stephen’s words…

The scariest moment is always just before you start. After that, things can only get better.

talking in code: state

2023-11-14T00:00:00+00:00

yesterday we talked about the concept of variables and their use in narrative.

today I want to focus on the concept of state.

programmers, or at least the older ones like me, tend to think a lot about this.

both a game and any other computer program have states.

if I am writing in a word processor, all the text I wrote is part of the current state of the program.

if I selected an option, for example an italic style in the text, it is also part of the state.

but not only what I wrote, the position of the mouse, which menu is open, whether the file is saved or not, the marks of the spell checker, all this is state.

as a simplification when creating stories:

all the variables that we have at our disposal and their value at a given point in the game are the state of our story.

as in the example of the word processor, this may not be limited to the variables that we create to guide the narrative.

if our game has gameplay elements or resources (for example ‘Pieces o’ Eight’), we can use the current state of that gameplay variable to define available options in our game:

SHOPKEEPER: The Sword is 100 Pieces o' Eight. The Shovel is 75 Pieces o' Eight.
  * I'll take the Sword
    req pieces_o_eight >99
    -> buy/sword
  * I'll take the Shovel
    req pieces_o_eight >74
    -> buy/shovel
  * I don't have that kind of money.
    req pieces_o_eight <75

in this example, you can see how I limit (using the req command, for require), which options I show: if the user has the money, I show the option to acquire the object, if they don’t, I don’t show it.

you can (and honestly should) use this to bring your game to life.

an example would be adding alternative options when the conditions are not met:

* I don't have money for the Sword, but... could I wash the dishes?
  req pieces_o_eight <99

this is what I usually call reacting to the state.

it’s vital that at every moment in the game we consider the possible states in which a player can arrive at a line.

think about if there’s something interesting that can be done at the narrative level by reacting to the current state.

is there any game that you think is very good at reacting to the state? Hit reply and tell me!

talking in code: variables

2023-11-13T00:00:00+00:00

what do programmers talk about when they talk about variables?

and even more importantly, why are we talking about variables if this list is about *game writing**?

let’s start with the first question.

a metaphor that helped me as a kid when I started programming was to think of a shelf full of jars.

one can store something in a jar, and it stays there until the next time you want to use it.

when we program, these jars are called variables. you can even label them, just like jars.

we use them to store information that we consider will be useful in the future.

why is this important for narrative?

because it’s essential to use variables to change our story.

to react to player choices.

let’s analyze a small example from the game I’m writing now, using cuentitos, the narrative engine we created at Hidden People Club.

Jaime: I am Jaime.
Jaime stretches out his hand to greet you.
  * I take his hand
    set tactile_sensitivity false
    You shake his hand and he smiles.
  * I look at his hand
    set tactile_sensitivity true
    You get caught up looking at his hand.
    He looks at you and puts his hand in his pocket.

[...]

Mom: April! Thank goodness, I was worried. Now I understand why you were late.
She hugs you and kisses your forehead.
  req tactile_sensitivity false
She puts your blanket over you and sits beside you.
  req tactile_sensitivity true

in this example, we’re using the variable tactile_sensitivity.

if the player chooses the option “I take his hand,” we store the value false. If they choose “I look at his hand,” we store the value true.

throughout our game, this variable can be used to modify the story, in small or large ways.

in the example, April’s mother, who knows her very well, hugs her if she doesn’t have tactile sensitivity, or sits beside her without touching if she does have sensitivity.

this is an example of reaction to state which is something we’ll discuss in a future email.

to recap:

a variable is like a jar of information: it stores data that we can use in the future.

it’s essential to understand that variables are the starting point for creating dynamic stories.

no choice

2023-11-12T00:00:00+00:00

in yesterday’s email, I proposed the importance of the player agency in narrative games.

that is, their actions should have consequences.

but for some reason, many video game writers ignore this.

and I’m not talking about ignoring it in major story arcs, where sometimes it’s hard to keep the promise of agency.

what I mean is giving you choices, which are not really choices.

we’ve seen this many times:

BARTENDER: Good day, traveler, what are you looking for on this beautiful day?
 * A good beer
 * A clean bed

[The PLAYER chooses 'A good beer']

BARTENDER: Sorry, but I've run out of beer for today.
BARTENDER: How about a clean bed instead?

is there anything more frustrating than this? I doubt it.

giving the feeling of agency only to take it away the next moment is one of the worst narrative experiences we can design.

unless, of course, you’re using it narratively, seeking that effect. as a designer.

have you ever done this? why?

did you have a good reason? did you do it by mistake?

or is there something I’m missing, and it’s an essential narrative design pattern and I’m just the weird one here?

anything is possible

the challenge

2023-11-11T00:00:00+00:00

the main challenge in writing games vs. regular storytelling is interactivity.

as writers we need to develop an understanding that now the medium is not one-way anymore.

players expect agency (having the ability to modify the world somehow).

they love it when their choices matter.

as a game writer, you need to keep this in mind always.

telling stories linearly may work sometimes, but it’s usually not enough.

next time you’re writing a game, pay close attention to how you’re letting the player modify the world and make that a storytelling priority.