Wow, the end of winter break already! We finished the semester in a whirlwind of hard work and going to sleep on time, so we didn’t write any devlogs in December, and over break we took a big step back and didn’t focus our energy on the game, so we also haven’t written any devlogs since!
But now, the start of second semester is on the horizon (in TWO DAYS) so it’s time to look back and reflect and catch our devlogs up.
State of Getting a Blanket
First and most importantly, we got a big blanket! Sinisterly, Muji cancelled our blanket delivery because it was out of stock or something. So we ordered one from LL Bean and it is very big and comfortable! Yippee! I will continue to order from Muji in the future, because this incident indicates that they are popular and have good products.
The State of the Game Right Now, and Future Possibilities
Stickers and Sticker Chat
As of the last devlog, our dialogue system was sort of wrapped up in various iterations, as we tried to figure out the best way for NPCs to understand and respond to stickers, and how we should allow players to use them. I think for the first couple months of the project, it felt like the dialogue system was at the center of the game, so it had to be complex and responsive and interesting, and we were quite stressed about how we could make it feel good and captivating without it feeling like the player was solving a puzzle, or like they were actively solving the mystery.
Shortly after the last devlog, we met with Lawra and showed them some of our prototypes for dialogue systems. They’re response was something along the lines of “What if this was more like a status system, and I just react to statuses with stickers?” And we were like… “What IF it was like that?”
This idea opened up a lot of possibilities, and seemed to serve other parts of the game quite well. A status is way more passive than an active dialogue, which takes off a lot of the emphasis, but still could allow for an NPC who is close friends with the player to update their status message in response to what the player reacted. Additionally, because a status is so passive, it would lend itself well to contextual messages that don’t really require responses. For example, if the player is in a game of war/capture the flag, each character can have a contextual status message for that scene, referencing the player’s goal, that NPC’s position in the game, or just cheering on their team. In these cases, NPCs can express themselves and give the player information without needing to be so direct as actively starting a conversation with the player character. After discussing this change, we decided to give it a shot!
We ultimately found great success with this! One-sticker reactions also meant it was far easier to have NPCs respond to players naturally, without giving a confusing message back, or having no response at all, since players don’t have to create sentences in an obscure sticker language. Our first try of this featured really direct and clear stickers, like thumbs up, angry, scared, etc. This made conversation when an NPC replied really clear, but, after meeting with Winnie and showing her the new prototype, we realized that it both didn’t let the player express themselves as well as the weird stickers we had previously, and also limited how interpretive a player could be of responses they received from NPCs, since the meaning of the stickers they used was so clear. Because of this, we made a new set of stickers that was way weirder and more fun for our second try, including stickers like a cat with milk all over its mouth, a wizard, and a rock. This is the sticker set that we went into the end of semester showcase with, and we found that people had a lot of fun using them! Some people also shared their intentions in meaning when using each sticker, and those varied quite nicely as well.
By the end of year showcase, we also added a personal status message for the player, which people got great use out of as well! It is purely expressive and doesn’t connect to the rest of the game at all (your status message doesn’t even appear over your head in any circumstances).
Also of note is something that Winnie said in one of our meetings at the end of the semester, which was something like “Only 15% of dialogue needs to actually directly respond to what the player says in order for the player to feel heard.” In writing the dialogue for our showcase build, I really took this to heart, and made the NPCs only respond directly to the first sticker the player used, repeating that sticker in their response, before converging back to just a couple branching paths.
The Future of Stickers and Sticker Chat
Such expressive and interpretive stickers call for, in my mind, a much larger set of stickers that the player chooses from for the smaller sticker set that they access with their keyboard. This would require a couple of large decisions to be made.
First: How exactly does this broader sticker pool interact with the smaller pool on the player’s sticker keyboard? Can they swap stickers out at any time? Probably not. Do they have access to the full sticker library from the start of the game? Also probably not. There is a natural trading economy that emerges among kids on playgrounds that I think would be really fun to explore if we introduce a broad sticker pool into the game. Maybe you would swap stickers directly onto your keyboard, or maybe you would have a larger sticker inventory that fills up as you trade, which you choose from at the start of the day.
This could also open the door for a lot of interesting integrations with the rock collecting system. Rather than trading stickers for other stickers, you offer kids rocks, and then they offer you one of their stickers (sticker gacha for rock money). It could be even more particular than that in certain cases: Mimi needs more beads? Maybe you found a bead on the ground and put it into your rock pouch, so now you give it to her and she gifts you a sticker as a reward! I think it could be quite fun to integrate some “secret” and “side quest” type of things that let you collect stickers that you can’t get from normal trading.
Secondly, we would have to figure out how NPCs would understand a much larger set of stickers. If there are 50 stickers in the game, there’s no way we can write 50 unique lines for each sticker use case. Nico had an interesting suggestion, where each sticker can be “tagged” with the possible emotions it can express, like anger, sadness, slay, yippee, etc, and then each NPC has a ranked order of the emotions they’d assign to a sticker if that tag is present. So if an NPC has a ranking of 1. Anger, 2. Sadness, 3. Yippee, 4. Slay, and you react with a sticker tagged with Slay, Sadness, and Yippee, that NPC would understand it as Sadness. It would have checked first for anger, and not finding it, then for sadness. In this world of four emotions, the only way this NPC would understand a sticker as Slay would be if that was the only emotion it was tagged with.
I think that this is a cool idea! I think it could easily lead to NPCs feeling really flat, though, if a player just so happens to have a bunch of stickers with one of the same tag consistent between them. Then any NPC with that as their number 1 ranked emotion would respond with the same emotions no matter the sticker used. If we lean heavily on single or double tagged stickers, though, it could work quite well.
Rock Collecting
As of last devlog, Rock collecting was merely a concept, mocked up in black and white for UI class. Now it exists for real. At a base level, it exists as planned. You pick up rocks from the overworld, and then bring them to your pouch to organize them in a physics UI. Awesome!
Actual integration of this led to running up against a bunch of issues. A lot of these issues were solved by having the rock pouch originate not from a UI element at the bottom of the screen, but from the player themselves. To pick up a rock, you would click on it and drag it to the player character, which would open the UI, with you now holding the rock in the physics sim. In order to access your rock pouch without picking up a rock, you can just click on yourself! Using mouse placement to close the UI was a busted idea from the start, since it would inevitably lead to unintentional UI closing, and also if you drag a rock all the way to the top of the UI to take it out of your pouch, now the rock in the overworld is totally out of your rock controlling range! What the freak! We solved this by just closing the UI by clicking outside of it. This seemed pretty intuitive to people!
The Future of Rock Collecting
We were left with lots of hopes and dreams about rock collecting that we didn’t have time to implement, but those hopes and dreams were overshadowed by the excitement of people having fun collecting rocks at showcase! This actually led to a lot of great feedback, and guided our intentions for the future of rock collecting.
The first big thing that we’d like to do is to connect the overworld with the rock UI more. Firstly, this would be done by having rocks that are placed on the ground around the pouch UI appear in the overworld (and leave the pouch) when the pouch is closed. Secondly, rocks that are flung out of the UI would also fly away from the character in the real world!
The second thing is a change to how we want players to leave rocks at shrines. We noticed that some players were dragging rocks around to leave them in different places, and one player, Nico, even dragged a rock over to leave it as an offering at the dead body. This made us reconsider even having a UI for shrine offerings, thinking that it would be more childlike and interesting if those offerings were left in the overworld, around the shrine. We weren't thinking this would be like, codified or built into the games architecture in any way, but instead implied with the design of the shrines, having a circular “offering grounds” around them, as well as placing offerings from other kids on the ground around shrines too.
Starting from Zero
The plans listed in this section were our intentions after finishing the build of the game for showcase, and after seeing playtests at showcase. I think we are still fond of a lot of these plans, but we have also been away from the game for over a month now, and intend to revisit it with more of a “blank slate” mentality at first, keeping our minds open to whatever thoughts and ideas come up, without holding expectations of what we will do next.
Boars
At some point we decided that there would be an animal with a heavy presence on the island, and at some point after that we decided that it would be boars. We want to use them to represent something opposite to the structure of society in which things like rumors and sensationalization thrive, and they’ll be quite important in the storyline of our second-loop player character.
Winter Break
Though we didn’t work much on the project over break, we played lots of games and watched lots of movies, so I did think about it a lot, and I did some writing about various things, which I’ll put in their own posts. In the meantime this is some of the stuff I played/watched/read over the break!
Megaman Battle Network 4: Mostly a nostalgia pick, not as much for research, and quit a few hours in because I hadn’t saved for an hour and died. Really a great reminder that nostalgia tends to ignore the bad stuff. The children in this game are absurdly rich and seem to have no parents, truly a kid’s fantasy. Kind of useful for thinking about what styles of decor and web/software design kids would be into in the early 2000s!
Moon: Remix RPG Adventure: A treasure trove of silliness and joy and seemingly anti-player decisions (or maybe pro the games ideology?). Made me feel like a kid, which led me to think a lot about what makes a player feel “childlike,” which I’ll share in a post later.
Despelote: Another time-limitation heavy game like Moon, which is helpful as we consider how to handle the passage of time in our game. I watched Jib play this one, and definitely do want to play for myself after doing so.
Anatomy of a Fall: (spoilers kind of) This was great because it is an “unsolved” crime in a narrative sense, since we never figure out what actually happened. It’s interesting to think about why this works and results in a satisfying movie, and how we can apply this to our game, if at all. I wonder if it may be that the courtroom setting that around half the movie takes place in puts the viewers in the place of the jury rather than the detective, asking them not to solve the incident, but to pass judgement on it. I felt this reflected in how I felt at the end of the movie, convinced that a final, jury swaying testimony was falsified. I think it also helps that there were only two to three possible answers: Either Sandra killed her husband, he ended his own life (or he fell by accident). This binary (an accidental fall is ruled out early on, though I guess not totally impossible) choice for the viewer makes it a bit easier to be satisfied, as their processing of the information in the case can be hyper-focused. By the end of the movie, they either agree with the verdict, or they don’t. In a direct sense, these solutions can’t really be applied to our game. But as we continue to develop it and hone the narrative we can keep them in mind, and hopefully find them to be a positive influence.
Wake Up Dead Man & other murder mysteries: These overall are super different than Anatomy of a fall, as they’re closer to puzzles, where the pieces to reach a solution are laid out before you leading up to the conclusion, and if you’re hyper-attentive, you can solve (or nearly solve) the crime before the detective. I really liked Wake Up Dead Man, and watched a few interviews with Rian Jonson afterwards. I was struck in particular by one thing that he said, which was that in a good murder mystery, the detective is never the main character. I was surprised to find that to be true of a lot of the mysteries that I have read! It made me wonder if our story could use a detective character, one who is completely inaccessible to the player in any way, but who they can see investigating. Maybe that would really hammer in that they are not the one who will solve this mystery.
Overall, reading and watching these, it is interesting to consider the modular/procedural version of this sort of puzzling that we are trying to accomplish. We essentially want the first 90% of one of these stories for each player, where they have a foundation for a belief and some real trust in it. We want, say, 3 of these 90 percents to exist concretely within the story, and variety within each of those that creates fluctuation and variation between different versions of the same belief.
It feels almost out of reach, and I think that the real solution for us is somewhere in between Wake Up Dead Man and Anatomy of a Fall.
Stray Children: Haven’t played it yet, but the people who made Moon released a new game recently, where you once again fall into a digital game world, but this time it’s a world full of children, where all of the adults have turned into evil creatures. It’s up next to play after Moon for me!
What We’ll Do Next Semester
I think that we want to take time to revisit the game with an open mind, as I said before, and plan anew from there. I think it makes sense to treat our big decisions as set in stone for now (as in: the dialogue system is based on status messages) so that we can focus on building out the game from there, and polishing interaction with those things, but we’ll see how we feel after playing through what we have so far once more!