Week 6, mechanism or bust

This is an edi­tion of Robin Sloan’s video game devel­op­ment diary.

Welcome: to returning readers as well as everyone newly subscribed. If you missed it, week 1 sets up the moti­va­tion behind this project. All the pre­vious edi­tions are avail­able over on my blog.

The k in the hea­dine at the top of the page is dou­bling down.

Wot I got

This past Tuesday was prob­ably the most impor­tant day in the life­time of this project so far. I had been strug­gling with a core question; I addressed it at last; I had an idea; I imple­mented it.

And now I’ve got a game­play mech­a­nism.

The Clois­ters Playing Cards, ca. 1475–80

I always chose blue, of course

Regarding the actual mech­a­nism of this so-called game I am making, my ini­tial strategy was basi­cally to ignore it. Seriously: this was a con­sid­ered strategy! The main attrac­tion of Perils of the Overworld, I figured, would be the writing, and the basic push-pull of making choices — the “what next?” of a Choose Your Own Adven­ture book — would be mech­a­nism enough to keep the player tapping.

I still think there’s a kernel of truth there — “what next?” is one of the most pow­erful forces in the universe — but the no-strategy strategy did not work. Here’s why.

When­ever a nar­ra­tive game’s choices affect its outcome, its designer has to decide whether, or how clearly, to indi­cate what the result of those choices will be.

At one end of the spec­trum, there’s total trans­parency and predictability. For example: in the Mass Effect series of video games, your dia­logue choices affect the devel­op­ment of your character, and the direc­tion of this devel­op­ment is telegraphed bluntly with red and blue text. Selecting a blue dia­logue option reli­ably makes you more of a “paragon,” red more of a “renegade,” like so:

(I need to thank Simon Mattes for that screenshot, which came with the file­name masseffect-best-line-ever.jpg 😗)

Mass Effect’s paragon/renegade dis­tinc­tion has impli­ca­tions for the game’s story; it also affects your avatar’s appear­ance. (The fully “renegade” avatar sports glowing facial scars. Woo!) Not to worry: if you wish to follow the path of the paragon, you can just select the blue dia­logue options. That’s it. You don’t even have to read them!

Way at the other end of the spec­trum you will find the very cru­elest of the Choose Your Own Adven­ture books, in which your choice to, for example, “explore the laboratory” might be rewarded with

You trip a hidden sensor. You are incin­er­ated by lasers. YOU HAVE DIED

And you’re like, WHAT sensor? What lasers??

In CYOA books, death was swift and, worse than random, it could be spiteful: a bite from a deadly snake, its pres­ence in the lifeboat totally unforseeable, might be your pun­ish­ment for choosing the “safe” option.

Game designers gen­er­ally dis­like this end of the spec­trum. The con­ven­tional wisdom is that the player (or reader) should have SOME idea what kind of choice they’re making; otherwise, the con­se­quences feels capricious, unfair, and just gen­er­ally bad.

However, the totally deter­min­istic end of the spec­trum is pretty uninspiring, too. When the results of every choice are fore­shad­owed per­fectly, the story sort of col­lapses into a spreadsheet.

So … I really had no idea where to place POTO on this spec­trum.

I knew I wanted to set up com­peting values of “adven­ture” and “attachment.” I knew accu­mu­lating too much of either would be bad. But I didn’t know if the values were going to be dis­played as like, meters, or as counters, or as little faces (with expres­sions that slowly changed?), or not at all. And, writing Ink — the code that spec­i­fies the game’s branching story — I didn’t know how to com­mu­ni­cate the con­nec­tion between those core values and the choices I was pre­senting to the player.

You have seen it already: the problem was that I didn’t know what the con­nec­tion was between those core values and the choices I was pre­senting to the player.

“What next?” needed a companion: “ … and why?”

The witch’s game

Feeling stuck, I took a step back and con­sid­ered the feel­ings I was trying to pro­duce. I wanted the player to have to bal­ance one value against another, without making it feel like a spreadsheet. (Indeed, I wanted to actively prevent them from treating the game like a spreadsheet.) I wanted to entice the player to reach fur­ther and fur­ther — with the risk, always, that they might go too far. I wanted that walking-a-tightrope feeling.

Turns out, there is a very famous and pop­ular game that pro­vides all this!

It’s black­jack.

I believe that, with any cre­ative project, there are por­tions in which you are attempting some­thing truly new, and por­tions in which you just want stuff to work. For those latter por­tions — this is very impor­tant — you should be shame­less about stealing things that are already proven.

Here are some nice prop­er­ties of black­jack:

• It’s simple, and many people already know how to play it.
• It combines strategy with chance.
• It is fundamentally about approaching, but not exceeding, a limit — producing that feeling of “eee, should I?”
• It has a rich history that can be mined for additional material.

That last one I didn’t know before I started googling. People have been playing black­jack and black­jack-likes for hun­dreds of years, and there’s a whole host of related games. There’s kvitlech, played with a deck of just 24 cards. There’s cuccù, which has some truly strange cards. (“Big Pants”??)

And, deliciously, there is hexenspiel, the witch’s game. Look at these cards!

The pres­ence of “the inn” and “sausages” among them made this con­nec­tion feel fated. There was already an inn in my game. There were already sausages!

In short order, I dis­cov­ered that black­jack not only mapped neatly onto the game I was making, but also offered a dis­crete struc­ture — a kind of exoskeleton — I’d been missing.

Let me walk you through it. The pre­vious struc­ture was:

In every loca­tion, you made a series of choices, divining the meaning behind the text, hoping for the best.

The new struc­ture is:

In every loca­tion, you “draw two cards,” which is to say, you play through two short adven­tures of unpre­dictable intensity/weirdness. From those, you receive adven­ture points — the equiv­a­lent of your black­jack hand.

Let’s say your total is 14. So far, so good.

Then, the over­world deals itself in. The game “draws two cards,” which means you play through two short episodes that pull you into the com­mu­nity of the place you’re exploring. Those episodes incre­ment a dif­ferent tally: attachment. And, as in black­jack, only one com­po­nent of that point total is shown. The other remains hidden.

Let’s say the over­world holds a 10 and a mys­tery card.

Then comes the crit­ical moment: do you “draw another card,” which is to say, seek out another adven­ture? Doing so can nur­ture your sense of ambi­tion and restlessness. It can also send you tum­bling over the edge.

The alter­na­tive is to hold steady, con­fi­dent your 14 will be sufficient.

Decide what you will. If you beat the over­world’s total, your sense of adven­ture pro­pels you onward to a new loca­tion of your choosing. If you fail, your game ends with a pleasant vignette.

(If you “bust,” going over the limit — which might or might not be 21 in the final game — then you die a classic adven­turer’s death. Dagger in the night, pit of spikes, that sort of thing. A touch of the vin­tage CYOA frisson.)

Chance plays an impor­tant role, but, as in black­jack, the risks are all calculated. You know what you know, and you know what you don’t know. You do your best.

This mech­a­nism is now up and run­ning in the game, if crudely, and: it’s fun?!

Here is a quick cap­ture of the game­play as it exists now. As with pre­vious cap­tures, I am showing this to you incan­des­cent with the knowl­edge that it looks like shit (we have ahem not yet achieved per­fect typesetting) and, additionally, that it is tough to follow along when it’s not you doing the tapping, advancing the text at your own reading speed. So, sorry about that. And, excuse my cursor.

Press play to check out a sliver of black­jack-as-quest:

Sound snack

You know what part of the cap­ture above doesn’t require any caveats? Jesse Solomon Clark’s won­derful tap-by-tap score. My system for selecting and playing Jesse’s “plonks” is still very rough, but even now, his music is working with the text to deliver some lovely moments!

This week’s sound snack is just one of those plonks, isolated; my favorite one. Every time this was selected to play with a bit of text, I hon­estly got chills:

As I’ve wired these plonks into the game, it’s been fun and inter­esting to hear the effects that emerge. If you tap quickly, the plonks pile up, pro­ducing sur­prising new chords. If you tap slowly, they become a kind of ten­ta­tive pulse. That unpre­dictability is the whole point. A dynamic, reac­tive score is the state of the art for games these days (rather than a sound­track played straight through) but/and I do think POTO’s approach, molding the music tap-by-tap to your actions, pro­duces a novel feeling; almost as if you’re the conductor.

The power of music and moving images together is widely understood, widely sought; I think music and text together are a bit underrated. Jesse’s devel­op­ment of POTO’s “plonk score” deserves its own edi­tion of the newsletter, and it will have it.

Postcards from possibility space

Those of you who took a journey across the uneven ter­rain of pos­si­bility space: thank you sincerely. It was a weird request, but/and the votes you sub­mitted are very helpful, and the spe­cific phrases that many of you emailed to me are won­derful. I haven’t been able to read all of your selec­tions yet, but I pledge to do so.

I’ll soon take those thou­sands of city descrip­tions offline, and there’s some­thing a bit melan­choly about it. Is this a new 21st-century feeling, or is there some his­tor­ical analogue? You pro­duce a moun­tain of mate­rial so large you can’t pos­sibly review all of it — no one can — and so, in the end, inevitably, you have to just … throw most of it away.

Here are a few high­lights skimmed from my inbox.

Jasper found this very com­plete and appealing setup:

A trav­eler in Facto Shi­bokawa had to beware the witch and the dragon, the dragon who devoured the people of the city, and the witch who knew all about the sac­ri­fices that had been made to gain the city’s wealth. One must always beware of the witch and the dragon in Facto Shi­bokawa.

Kip found this gnomic line:

Inside Ambiacco, the magic was so cun­ning that it had the appear­ance of a grey hair falling on the belly of a hog.

Thomas found this passage, which is, hon­estly, a tiny hybrid William Gibson story:

Therefore, Forta Altilla was a major con­cen­tra­tion of training for anyone wanting to stretch one’s kinesis edge in all pos­sible direc­tions. Yet when the young Teck guys like you trav­eled from place to place, they would be anomalous: their out­er­wear was exquisite. They favored silk.

“They favored silk”!!

Márton high­lighted this evoca­tive description:

Because Borosnes was ancient before it was built, it was essen­tially a human city from before the rise of men. Borosnes was old.

This dis­covery of Dan’s is not, I don’t think, right for POTO — 

The city of Spade­valen was located some­where in the North, and around the lake and below the waves it had fifty one-story residences. It was sur­rounded by a wall to keep out unwanted foreigners. With here and there one or two exceptions, the whole pop­u­la­tion was treated like ser­vants and taxed. Inside Spade­valen, there were two fam­i­lies that lived in sev­eral apartments. Both fam­i­lies were suf­fi­ciently supe­rior to one another.

—but I would absolutely read that short story.

Samantha spotted this line, fit for a fable;

The river had little to do, and the moun­tain was happy.

Finally, Annie found this:

The main export of Ptera was candor. Its res­i­dents were of all ages and professions. All of them were gentle, kind, and honest.

Ptera sounds like a dan­gerous place for an adven­turer.

An adven­turer could get attached to Ptera.

This week

A bit of a twist: I’ve just signed on to a very cool writing project with a VERY fast timeline, neces­si­tating two weeks working at Warren Ellis speed, or as near as I can muster. (Can I hit … 0.6 WEUs?) That task will com­mand all of my attention, so this will be a bye week for POTO work. There will prob­ably not be a newsletter next Sunday; I’ll resume on Sunday, May 24th.

I don’t mean to be cryptic; I’ll describe this project — this rival for my time! — as soon as I’m permitted, first through my Society of the Double Dagger newsletter, and eventually, I’m sure, on Twitter.

Wonder weeks

The Clois­ters Playing Cards, ca. 1475–80

Just one more note before I go. I want to amplify some­thing from the very beginning: the expe­ri­ence I had, ear­lier this week, of a single Tuesday com­pletely trans­forming and unlocking this project.

A lot of advice about cre­ative work — writing in particular — calls for a com­mit­ment to the steady day-to-day appli­ca­tion of effort. “You’ve got to get in your daily word count” and so forth. And: that advice is good! However, it elides some­thing I’ve observed, which is the extremely uneven impor­tance of dif­ferent days’ efforts to the final product.

In early child­hood devel­op­ment, there’s this con­cept of “wonder weeks” scat­tered throughout the first year of a baby’s life, during which they exhibit sudden, star­tling leaps in their aware­ness and abilities. Maybe this has been totally debunked or some­thing; I don’t know. But even if it has, I will hang on to the analogy, because, hon­estly, I have expe­ri­enced “wonder weeks” in the course of every project I’ve ever pro­duced.

Of course, knowing that wonder weeks occur, it’s not like you can plan for them or call them forth: “Oh, sorry, I’m busy; I put a wonder week on my calendar.” But I think you can learn to rec­og­nize them — they’re not always weeks — and, having done so, lean into them: clear the decks, stay up late.

Just an obser­va­tion from my own life and work. Tuesday was a really impor­tant day.

From Oakland,

Robin

This has been an edi­tion of my video game devel­op­ment diary, sent by email every few weeks. You can subscribe: