...You know, I didn't intend to create a game about isolation, madness, and plagues during a worldwide pandemic, but uh....here we are.

Looking back at 2020 and the work on Black Heaven, I've learned a hell of a lot. This game has gotten amazing responses from nearly everyone who's played it, and I'm really grateful for all the people who have worked on it and supported it.

I thought it'd be cool to share some of the stuff I learned, in case some of you are interested in what it's like to develop a game, or are thinking about making your own. Either way, happy 2021--I hope things get better, and that Black Heaven goes back to being just a dating sim, instead of an eerily prophetic preview of our collective, downward spiral into the depths of our own minds.

Cheers,

- Chris Mahon, Creative Lead

Do Your Homework

In the immortal words of Gin Rummy, "...there are known knowns and there are known unknowns. But there are also unknown unknowns; things we don't know that we don't know."

What?

Here's an example: most VN developers know that creating a visual novel requires getting some character art, but it's an open question how much a piece of artwork is going to cost. Still, they know they have to pay something. Those are "known knowns" and "known unknowns."

"Unknown unknowns," on the other hand, are things you didn't even think about.

When I started Black Heaven, "unknown unknowns" included:

• What size that character sprite should be
• Whether a sprite should be a .jpeg or .png file
• How to write a contract for an artist that includes provisions for revisions

There are so many times I had to go down rabbit holes to fill in the gaps in my knowledge, whether it was about Ren'Py coding or forming an LLC. And the discouraging thing was that the more I looked, the more I realized that there were giant pieces of the picture I was still missing.

It can take weeks or months to suck up enough knowledge to get a rough picture of your game's development. It's extra frustrating because you never know when you're finished, and you're probably going to run into new, unexpected obstacles when you're actually doing it. 

But here's the thing: the more research I did, the more I could describe the holes in my knowledge. I could start asking people specific questions like "I'm trying to use persistent variables in Ren'Py, but I don't know which file I should be looking at--script or options." People became much more willing to help me when they saw that I'd already put in some work and done my homework.

Bottom line? If you do your homework, most of the big "unknown unknowns" get broken down to "known unknowns," which can be tackled more easily. It's a long, hard road to building a complete picture for your game's development, but it ends up paying off by making it easy to ask other people for help. 

You Can't Do It Alone

This one has been the hardest for me because I like to work on projects by myself. Until now, I've hated collaborating on pretty much anything (apart from a kung fu rap album I worked on), and I was dreading all the networking that I'd inevitably have to do. But--surprise!-- it's not so bad working with people,  and Black Heaven has taught me that you really can't make a game by yourself.

Finding a team of people was the first hurdle, and looking back, I really lucked out--Linnea Kataja is incredibly professional and easy to work with, Lim Chin Yang puts out incredible work quickly, and Marek Domagala is a creative genius when it comes to turning vague briefs into compelling music. BaiYu was conscientious and thorough, and Re:Alice did a great job (despite my inexperience with laying out UIs).

The most valuable thing, I realized, has been professionalism--someone who delivers when they say they will, and puts in the work to create quality. I've worked with a lot of incredibly talented and creative people, but most of them weren't organized, consistent, or...entirely coherent. For Black Heaven, loosening my grip on things and trusting team members to do their own thing was a big step.

Apart from that, holy shit is the VN community supportive. I've never been part of a scene where so many people are ready to help new people, give advice, or showcase others' work. Lemmasoft Forums, DevTalk+, LoveVN, and the NYC Gaming Discord have all be hugely helpful (and encouraging) at getting the game this far.

As I mentioned, I don't like networking--I can't shake the feeling that I'm doing something vaguely disingenuous when I do it. But talking to the excellent people who surround the VN scene has shown me that asking for help is essential. Better yet, it's shown me that a lot of experienced people will jump at the chance to give the advice they wished they had when they started out.

Designing Branching Narratives Is Maddening

The first chapter of Godel, Escher, Bach: An Eternal Golden Braid starts by explaining how bananas it was that Bach was able to improvise a six-part fugue for King Frederick. It explains how a fugue is made up of musical threads called canons, which are all variations on a single theme (which is a melody).

Gradually, canons are layered on top of one another until they weave together into an incredibly complex pattern, where a theme can be repeated backwards, in higher keys, and at different speeds. And Bach could just...do that on the fly.

Creating a branching narrative isn't that hard, but it's in the same ballpark.

For Black Heaven, I'm dealing with five narratives:

• The PC's past relationship with Ru
• The PC's past relationship with Izagi
• The PC's past relationship with Lyse
• The timeline of events at Itzon that led to the plague (and the school's destruction)
• The PC's experience of returning back to Itzon to collect his classmates' souls

...but in addition to those, I'm introducing points in the narrative where reality, memory, and the PC's fantasies get mixed up, causing some ambiguity of the truth of certain events. These represent branches from the main narrative, and end up creating multiple iterations of later scenes.

I've done a lot of writing over the past six or seven years, which has always included outlining and revisions. When it comes to branching narratives like this, however, I had consistent trouble just...getting a handle on all of the different iterations and branches. I ended up using Twine to help me chart the three narratives of Ru, Izagi, and Lyse, but the larger narrative seems to big and cumbersome to chart out without a program like Scrivener.

However, while I was working on Black Heaven this year, I was also working as a contractor for Tales, a mobile publisher for game-like, interactive stories. I had to learn their coding language and editing software, but the toughest thing of all was writing ten-minute, episodic stories with significantly branching paths--I might write 8,000 words, but a player might only see 2,500 of them in a playthrough. 

Working with Tales taught me a  something that helped with Black Heaven: how to think in modular chunks. Regardless of what path a player is on, there are certain milestones they should experience, and those milestones are the points where all the paths converge, then diverge from again. 

For example, if the PC has made a decision to take Lyse to a fancy restaurant (rather than a low-brow place), both paths might have a moment where the PC has to choose to reveal a piece of gossip about Izagi. That decision may not influence the way the rest of the night plays out, but it will change what scene the player sees two scenes from now.

Conclusion

There are so many talented people in the VN scene (and in the indie video game scene as a whole), and for the first time in the history of video games, almost anyone can make and distribute their own games. Still, the barriers to making that dream a reality are the same as always: money, organization, planning, and experience.

I've watched really promising projects just fall apart from the inside, and I hope this advice helps new game devs avoid that fate. Black Heaven wouldn't have gotten this far if I didn't have dozens of internet articles to learn from, and I hope I can add to the collective hivemind.

Good luck!

Note: I didn't proofread this as carefully as I would have liked, so please excuse any typos or grammatical errors. I'll send you a picture of my Chicago Manual of Style, if that helps repair my credibility.