This is going to be a long post so strap in. Yes there are annotations, deal with it.
It’s my intent to write a postmortem of Lost Coastlines, as well as my career writing preposterously large text-based RPGs in ADRIFT. It was always my intent to close my text RPG career with this game, and while I intend to submit a smallish ADRIFT game for Spring Thing next year, this is likely my last ADRIFT game of any substance as well as my last IFComp.
A little about my background in ADRIFT: Back in about 2002 I got an idea for a text game in the same vein as Hitchhikers Guide, and located ADRIFT as a way to make that game a reality. I spent more than 15 years abandoning that project and picking it up again before it became Six Silver Bullets, my first ever IFComp entry, in 2018. Six Silver Bullets was an atypical game for me: it’s structured like a puzzly-parser, where the player plays a very specific role and the story has a single definite conclusion, though far afield from what the player expects. Six Silver Bullets was…not well received. A small number of people realized what I was going for, and wrote me to thank me for making an interesting game, but most people saw it as a hyper-punishing slog.
To be fair, games like Six Silver Bullet are not my element: most of the games I tinker with are sprawling role playing games with a significant open-world component. I like to make games where the player can make up their own story, facilitated by a computer-generated world of my own devising.
IFComp got a taste of this in 2019, when I submitted Skybreak!, an extremely large menu-based science-fantasy RPG set in an open world where the player moves semi-randomly through the cosmos, making one and only one decision on each world before moving on. Skybreak! is much, much larger than Lost Coastlines, but its randomized movement system means that a typical player sees only a tiny fraction of the universe on any given playthrough. Lost Coastlines feels larger, no doubt, but this is simply because there are fewer constraints on player movement.*
Lost Coastlines predates Skybreak!, if you can believe it**. It began on a random Sunday walk through Loring Park, in downtown Minneapolis. I realized, while walking, that pairing random objects in my head according to certain patterns gave rise to interesting location names: “Teasquirrel street” and “Seedsewer city” and so on. The process was so fun that I raced home and systematized it, building an engine that could generate thousands of names using the same basic patterns…I then began building a game around it in ADRIFT.
…that was in 2017, following a failed attempt to make Six Silver Bullets and before I’d even heard of IFComp. Before I even started building Skybreak! Lost Coastlines has been in development since that fateful day and now, 4 or 5 years later, it’s finally out there for people to play. Obviously it got a lot more sophisticated over time, but at heart, it’s still at bottom a game about discovering places with weird names.
Following the release of Skybreak!, I redoubled my efforts to make Lost Coastlines, and resolved to make a very different sort of game than Skybreak!. I had rather expected it to be far better received than Skybreak!, simply because it has significant “quality of life” features Skybreak! lacked: for one thing, you can actually choose your destination. The difficulty of rolls is clearly telegraphed, and the player’s skills and inventory more closely hew to a standard fantasy universe (unlike Skybreak!, in which a player’s History skill is as important for navigating the galaxy as Astronomy).
This prediction, it’s fair to say, did not come to pass. I have no idea what score Lost Coastlines will receive in the Comp, but it seems clear from the reviews (some by the same people that enjoyed Skybreak!) that the ultimate reception, and the game’s likely legacy, will be more like Six Silver Bullets: a game with a tiny core of admirers and a lot of people whose overall reaction is “what???”
Am I disappointed? Not really. Between this and Skybreak!, I’d much rather people play and enjoy Skybreak!, a game set in a world I’ve been collaboratively building with a few close friends since I was a child. I’m also well-aware that my games are not exactly mass-appeal material, and I don’t intend them to be. I know these games are not for everyone. I don’t tend to make user friendly games. In part this is because of the ADRIFT system, which is inherently janky. In part this is because I see games like Lost Coastlines and Skybreak! as essentially computer-facilitated imagination: they’re at their best when you, the player, take an active role in dreaming up the world and finding your place in it. I encourage players to, for example, draw their own map or take copious notes on the routes to take to return to exotic destinations. Many people don’t want to do that…
Am I confused? A bit. There is absolutely no question in my mind that Lost Coastlines has a smoother, more user friendly experience than Skybreak!, in which the player can spend half an hour bouncing between the same 3 destinations, encounters don’t even bother to tell you how likely you are to succeed or fail, and half your inventory items are puzzling things like beetles and human brains.
So why did people like Skybreak! more? It’s possible that as a procedurally generated universe, Lost Coastlines felt a little hollow, a little empty and random, whereas Skybreak! is a bespoke setting where each world has abundant lore the player discovers as they play. Several reviews have lead me to believe that it’s also possible I just fundamentally screwed-up certain balance issues. I’m not so sure, but at any rate this merely deepens my confusion because I actually did extensive testing on Lost Coastlines’ balancing, while Skybreak!'s balancing is almost entirely ad hoc.***
It’s also possible that, counterintuitively, adding more player control made the game much more complicated, and therefore less fun. In Skybreak! one doesn’t have to stress over where to go next. There are no supplies to provision out, there is no character inventory to manage. Reviews of Lost Coastlines and Ferkung’s four hour playthrough of my game**** suggest that players struggled with this complexity.
Part of the disparity, though I’m convinced not a very big part, is probably simply because Lost Coastlines was a buggier game. While this game underwent extensive playtesting, it’s simply too complex to be fully debugged (indeed, it may be the most complicated game ever created in ADRIFT, at least, I know of no other ADRIFT game that has done so many tricks with procedural generation). Naturally, this detracts from the user experience, but Skybreak! was also buggy, so that’s not a full explanation, especially since Lost Coastlines’ bugs were less likely to softlock players, a notorious problem upon Skybreak!'s first release.
Did I do a good job? I think so, though it could have been better, and a bit more polished. I am hoping that, as time goes on, this game will have the “unheralded master work aspect” that at least one reviewer saw in the rough. As this reviewer correctly pointed out “no one is going to play this game” but “the people who do” will enjoy it for years. if you happen upon Lost Coastlines or this review, even years later, drop me a line at Theodidactus@gmail.com if you enjoyed it…it will make me very happy to know that the game brought joy to someone.
What now? As aforementioned, I’m largely retiring from making new games in ADRIFT, and I likely won’t submit to another IFComp. I’ve reached the point in my career where I need to get serious about making games that I can try to sell. I do intend to continue debugging Lost Coastlines and Skybreak!, and releasing new expansions to both (the fourth, and largest, expansion for Skybreak! will be out January 1st, 2023). I’m expecting this project to continue for many, many years, which means that someday this game will be far bigger and more complex than it is now.
I’m going to leave below two lists of spoilered text. The first are tips to play the game that I think many people missed, or misunderstood. This is not to be critical of the fine reviewers, because it’s incumbent upon a designer to make this stuff obvious, and I clearly didn’t. The second is a list of questions about stuff I hope you found. Write me if you have a good answer.
Some tips:
-
Lost Coastlines is a game about exploration. It’s not easy to build up pleasance (which functions as both a way to tabulate a high score, and the game’s currency) while moving between a few cells. It is likewise difficult to get rewards from travelling over long distances aimlessly. I strongly encourage players to KEEP TRACK OF THEIR FINDS and EXPLORE CREATIVE WAYS TO TAKE ADVANTAGE OF THEM
-
In particular, keep track of places that have a high probability of giving you treasure: there’s a beach full of diamonds right at the game start, and a market to sell them as well. The map is scattered with ruins that are overflowing with plunder.
-
You cannot die in this game. Instead, the game continues as long as you want. While you will slowly accrue Sadness, Madness, Fury and Worry from bad decisions, there are ways to discharge all of these. You will find them as you explore. I encourage players to be patient and methodical in their exploration. There are places out beyond the horizon where you can lose all your worries and exchange your fury for pirate treasure.
-
That said, there are significant consequences for accruing too much of any one form of unpleasance. A lot of reviewers felt like there were no consequences, especially for failing to feed their crew and repair their ship. If you neglect either for too long, or if your unpleasance grows too high, terrible things will occur.
-
Your character will continue to grow more powerful as the game goes on: you’ll collect more items that increase your abilities and situation bonuses, and your character will encounter more and more experiences that increase their might, talent, seacraft, shadowcraft, and dreamcraft. The game is punishingly hard at the beginning, especially the nightmares you encounter, however, as the game goes on, you’ll find yourself prevailing over more and more encounters, until by the end, even scalpel-hands man won’t frighten you.
-
Don’t open the spider door
My questions for you, the player.
- Did you fall in love with someone wonderful?
- How many Gilmore Girls references did you spot?
- What was your favorite mixed drink?
- Did anyone open the spider door?
Footnotes
*Lost Coastlines has about 4,000 unique actions that affect game state (though many are complex, involving hundreds of subroutines). Skybreak has 7,000 (and far, far more lines of prose text)
** Technically, the universe Skybreak! is set in has been under constant development since the first time I started doing silly fantasy worldbuilding circa age 6, but I didn’t start putting the world into ADRIFT until 2018.
*** Lost Coastlines had a dedicated tester whose only job was to try to break the game’s scoring system and provide me with extensive notes on game economy. This tester was given extensive notes by me on how easy or hard certain things should be. Skybreak!'s only balance tester is myself, and I have absolutely no idea how common or uncommon brains should be.
**** For which he deserves some kind of medal, particularly for correctly pronouncing Schneckloth’s name, and reading all his dialog in a crypt-keeper voice. Did you know that Schneckloth’s name comes from a criminal case very important in my day job? Now you know.
