Thursday, March 15, 2012

Neither Good nor Evil

Modern games always want to put in "moral choices" where you can choose to be good or evil. Some dress this up with something like "paragon" vs "renegade" or whatever.

The problem is that these aren't really choices. They're a choice. You would get the same flexibility with just choosing a good or evil character at the beginning. Occasionally you'll get a player that strays, but for the most part it's both rare and not very interesting to do so.

The answer isn't to choose a different set of personality traits. The answer is to ask questions whose answers will change.

The fundamental personality of a character doesn't generally change. Instead, find a facet of the character that will change with the situation.

How about emotion?

What if your choices aren't about being good or evil, but about whether the situation has made you angry or happy or confused or stressed out...

Talking about your emotions in a video game is probably not about changing the outcome of the event. It's probably more about changing how your party members feel about you. The progression of the fundamental plot arcs can be based on the character's built-in personality (IE, it doesn't change much with player action), while the way your party members act, their powers, and their AI can be based on your character's emotional activities (which are probably largely the same as the player's emotional activities).

Let's give a Jedi-based example. You and your buddy are Jedi. As such, you save a politician from being killed by an angry mob, then respect his diplomatic immunity and have him deported. This is in line with your personality: Jedi wouldn't let a lynch mob get away with murder, no matter how deserved it is.

How do you feel about it, though? Are you angry at the politician for being so horrible and getting away with it? Angry at the the mob for not acting properly to address the concerns legally? Disappointed in the failure of the system to protect citizens? Can't get your mind off the hottie leading the mob's charge? Feeling totally zen about the whole thing?

Your chat with your buddy will allow you to tell him how you feel. In turn, this will change how he considers you. You're angry? Your buddy will focus on playing the calm and collected guardian, and get better in powers like pacify and defend. You're disappointed in the system? Your buddy will learn skills to help manipulate the system - the "red tape Jedi". You're completely zen? He'll allow his own anger to come through...

This seems like it would be a whole lot more interesting than simply choosing "good or evil". Especially since a player's emotions can change! You don't have to be angry all the time! That wouldn't even make sense! If a player selects "angry" even in response to children playing with puppies, you can put him in anger counseling or something.

Wednesday, March 14, 2012

System Content

I've always been a fan of creating custom systems for each game you run, because you can make rules specifically to suit the needs of the game you want to run. It really does make a difference.

However, established systems shouldn't be overlooked. They have two major advantages.

The first and commonly acknowledged advantage is that established systems take the GM from "zero" to "ready" in a minimum amount of time. The GM doesn't have to build the rules, the world, the balance, the combat tables, the monsters, the NPCs, the maps - all that stuff can be pulled from content already in the system, although he's free to create it on his own if he wants to.

However, a system has a second advantage which is not commonly talked about: system content.

A world like D&D contains a huge amount of rules and content. The players can spend a lot of time trying to figure out the best build, debating which monsters are stronger, whether wood elves or moon elves are better, talking about whether the ancient war of Kalvyos actually exterminated the shadow gnomes, and so on.

They can do this without the GM - either on their own just perusing the books, or in tandem with other players. The GM shouldn't dismiss this kibitzing just because it has no direct effect on the game. It's player engagement: it makes the players want to play more, want to immerse themselves more.

Even in a weak ruleset, the system content can make up for it. See: every Star Wars game ever.

Part of the popularity of D&D is specifically because it has lumpy, irregular rules. Players can argue over minutiae like whether armor class should go up or down. Perfectly logical and abstract rules are what most indie developers aim for: "these are the simple, unified rules that you use for everything!" They're actually shooting themselves in the foot! Lumpy rules and content give the players traction.

System content is so valuable for raising player engagement that I think it's important to factor it in to your own systems. Of course, generating enough content is a challenge. Also, the kind of content to generate is important.

I think the content that matters most is rule-linked lists, preferably mixed with a lot of flavor text.

By "rule-linked list", I mean any variety or list of things which interact with the rules in some way. For example, each class has a list of level-up capabilities and restrictions. There's a list of dozens of kinds of guns, each with its own subtly different statistics. There's a list of planets each space nation inhabits. There's a narration about the last war, full of places and combatants and rule-legal results...

Basically, "rule-linked list" just means "don't stuff your guidebooks full of backstory and worldbuilding that is supposed to be interesting on its own: Make sure it ties into the rules and mechanics."

A common trap is to just create huge stretches of samey text. So, when building content, make sure to switch it up. Common elements to combine:

Narratives ("the elves struck back by...")
Explanations ("Elves live for thousands of years...")
Stats ("Elf: +2 dexterity, -2 strength")
Templates ("Generic elf warrior:")
Commentary ("{{ABE}} I always play an elf, 'cause they're smexy!")
Guidance ("It's usually best for the GM to present elves like...")
Images of characters/monsters/closeups ("this is an elf')
Images of the setting ("this is an elf village")
Technical-oriented images ("here are common elf weapons...")

So make sure to switch it up. Don't worry about balancing the types: you just want to make sure the player never spends more than five minutes without switching gears, or it can be boring.

But how can you build such a large amount of content?

Well, dedication is the classic way. Just keep making more until the game feels ready.

It's not very realistic, though. Another way is to make it a community project, and allow the early audience to give you content of their own creation. This'll probably work if you have an early audience.

Yet another way is to steal content from another, established setting. This is problematic because you need to build a lot of "glue" to translate that content into your system, but on the other hand, translating more content is something which players often love to do.

The last way I can think of is heuristically. In many situations it'll be possible to automatically create many kinds of details - for example, you can automatically generate thousands of pages of maps covered in nations, and even write up the summary of each nation automatically. You could also create an image system which could create schematics for any kind of gun (with many random visual variations), and combined with a gun-randomizer, create a thousand different kinds of guns.

The problem with such algorithms is that they create very repetitive, bland content. One option is to massage them manually, creating flavor text and leaving out the most boring ones. Another option in this internet age is to package the algorithm.

Why would you need to print out a blank character sheet? Link to a web site that lets you build the character. This other web site generates guns. This one generates nations. Put a small selection in your book, but then challenge the players to use the randomizers and generators on-line.

Anyway, however you approach it, I think you need to consider system content when developing a new system. A system that is too smooth and sleek doesn't give players any traction when they want to make the move from "newbie" to "fan".

Tuesday, March 13, 2012

Designing for Players

This is a technical game design post about live games (tabletop RPGs, LARPs, etc).

Someone from a long time ago is thinking about one of the games I ran back in 2006, thinking about running a version with different mechanics but the same dynamics. I figured, now's a good time to talk about it. Long story short: you can't just switch out the mechanics and expect the same dynamics. Long story long:

When you're doing a live game, there's the matter of GM time management.

There are more players than there are GMs. Sometimes ten times more. Therefore, you can't allow the players to eat an equal amount of GM time. If five players each want to spend five hours playing your game, you can't allow that to cost 25 hours of GM time.

The basic way to control this is to turn the players into a single unit. A "party". So the five players all occupy the same physical space (or digital space, whatever). Five players spending five hours on your game only costs you five hours instead of 25. Well, 5 + whatever it took to develop the content for those hours. This is the typical tabletop RPG model.

But this doesn't scale very well for two reasons. First is that parties have a maximum size. Past about six people, it's wayyyy too chaotic. The other problem is that some players will spend six hours a day on your game. Even if every other player is loaded into that one player's time, that's still six hours a day! Wayyyy too much.

So the next step is to develop games which make the players play with each other. IE, either without a GM or with only a portion of the GM's attention. So if a party of players spends five hours playing, maybe the GM only spends half an hour at it. Or maybe he still has to spend five hours, but he can spend it running lightly between half a dozen parties also spending five hours.

These were the sorts of games I began to develop. Plenty of experiments and missteps followed, but I learned a lot. Here are some of the things I learned:

Mechanics & Dynamics
If there is some special feel you get from a specific kind of play, it's not uncommon to approach game design from an angle of "I want the players to do this, so I'll run the game and make them do this". For example, in a game called "inhabitants", the players formed into cooperative subgroups that worked semi-independently to help the home base and survive. But that was achieved by having the right mechanics, not by simply saying "that's what I want everyone to do".

So, if you're designing your game, the first thing you should ask is "what do I want the players to do? How do I want them to behave?"

In the case of Inhabitants, I wanted the players to form into task forces and figure out the details of what they wanted to accomplish, how. With 12+ players, it was necessary that their time was not spent 1 to 1 with me. In addition, I wanted the players to think about the game over the course of the week: I wanted the game to be meaty enough that they would be able to discuss it amongst themselves with little to no interaction from me.

Later, with Kung Fu the Card Game the RPG, I polished this a lot more, so let's talk details about both games and the mechanics I used.

Both the in-game mechanics and out-of-game mechanics require attention. Not simply which dice you roll for what, but also how the players get together, the physical layout of the meeting areas, and the amount the players are likely to meet over the course of the week in person or over email.

Taking these into account, I decided that in order for the players to want to play largely independently of me, there were a few factors required.

1) GM-free rules are obviously a requirement. If the players have to go through you in order to do anything, then you have to be available all the time. Especially in a LARP, this is not a good situation. Therefore, design your rules so that the players can resolve most conflicts themselves, and only have to come fetch you if they want to do something outside of the common rules.

In Inhabitants, I did this by making most of the gameplay about simply choosing between options. How do you lay out your base? How do you fill out everyone's schedule? There are no die rolls, no randomness: this is up to you. Write down the result, and then we'll grind through it all. Choices which are difficult or cumbersome to make but easy to understand when written down were ideal, so deciding schedules for characters played a huge role.

In the card game, I did this by making most of the gameplay about dueling. Basically, like Magic the Gathering, you don't need a referee to play the game, even though everyone's decks are so different and may contain cards you've never seen before. The rest of the rules I built to work asynchronously: submit a quest request by email, I'll get back to you. This works well.

2) Small groups - average of three players per group. I didn't need to aim for the maximum size of 6, because I am not managing the groups 1 to 1, just circulating and touching bases. In Inhabitants, I pushed to keep groups small by scaling the number of things that required attention to always be about 1 per 3 players. In the card game, I did it by making the rules support mostly 1-on-1 battles.

3) A central information system for the world state. In Inhabitants, I used a white board and a table covered in sheets of paper. In the card game, I used everyone's decks of cards, and encouraged them to make custom cards with beautiful images if they wanted to. Clarity is important, but so is detail. You could do this using a computer program, but flexibility is very important: this is where the players will express themselves.

4) An extended information system for players who want to explore.

In the case of Inhabitants, I used a flash program containing all the still-frozen inhabitants who could be woken up from sleep. There were around five hundred, if I remember correctly, with dossiers and stats and such. Exploring them kept the players who wanted to do that sort of thing busy, but instead of them pulling ahead, the busy players simply came in with ready-made lists of who to wake up, why - facilitating the less active players' experiences. Also, there were loads of quiet details hidden in: the military uses this particular catchphrase to mean this particular kind of thing, they are trying to keep particular events quiet, but you can figure them out by the patterns of who was transferred when, and what their specialties were...

In the card game, I used "snippets", which is a simple technique that pays off beautifully. With snippets, you write down the whole plot, but scatter it such that a player can't tell what's going on from just the bits he has. Players gathered snippets from each other by winning duels, and were only allowed to discuss snippets with each other when they both had the snippets in question. Snippets is ablative armor: the players will burn through them. Creating tension between the players (setting them up as competition for a limited pool of winners) extends how long that ablative armor will last, as does confounding rules (not allowed to discuss snippets), although the rules and tension will both be ignored in the fullness of time.

5) Amount of tension - I've run games with players at each other's throats, and games where everyone was clearly on the same team all the time, and everything between. You need to choose how much tension to create. The more tension there is, the less the players will work together. This can be very valuable if your game could be solved by everyone working together.

In Inhabitants, all the players were flatly on the same team. This was insured by having an inclusive end condition (everyone can survive) and by widespread amnesia (nobody starts with conflicting or tangential goals). A strong amount of continuous pressure from the game world also kept them from developing many internal tensions, even though some of the players really didn't like each other in real life.

In Bastard Jedi, the players were always at odds with each other. This was accomplished both by giving the players backstories that set them slightly at odds, and by making the fundamental rules of the game such that different character builds wouldn't get along very well. This is an extremely effective technique, to the point where I felt confident that the players wouldn't get together and talk about their backstories. If they had, they could have solved the game right from the start - but they didn't. It was a long, long time before they started to realize that their backstories and private experiences were actually pieces of a jigsaw puzzle.

In Kung Fu the Card Game the RPG, the players were set at odds by the rules (combat-based resource collection, limited number of winners) but were actively brought together by the other mechanics of the game (dissecting the rules, discussing the plot, the fact that none of them had a strong stake in the plot). This is a good combination, because the players will feel that they are "beating the GM" when they collude on the sly. This worked really, really well: the players actually banded together and secretly created a task force to protect a midboss so that nobody would win the game before they figured the plot out.

6) Reactive touches are important. While the game can unfold largely on its own, nothing ever behaves exactly like clockwork. The most common application of reactive touching is giving your most excessive players more content to keep them fed and involved, while at the same time helping them to enhance everyone else's play experience. Basically, the idea is to give them stuff to do that, if they finish it, will allow them to help other players. Do not give them content which just flat-out makes them stronger or changes the game world, because then the less aggressive players will get left behind.

In Inhabitants, I did a lot of this explicitly. Individuals and small task forces would be given a variety of options and clues which they could grind through to figure out a good arrangement for the base as a whole. Math and pattern recognition challenges are a good choice. This was supported by having the huge list of "everyone still frozen" - combining data from different sources is always fun.

In the card game, I built this capability right into the game with the snippet mechanic. I knew that some players would drop out, so I gave their snippets to the most aggressive players (via email). Since everyone was really working towards the same end (despite the smokescreen of fights and limited victory slots) I didn't have to worry that they would make the other players miserable. I also insured that the most aggressive players were also going to be the ones that were interacting the most with other players, so you couldn't have someone soloing the game.

This mechanic still needs some polish, because there were a few shortcomings. One shortcoming was that party formation was too strong: several players would form a solid party and there would be a social barrier between them and the less aggressive players or players from other parties. Another shortcoming is that the character advancement system was somewhat faulty, and allowed for aggressive players to alienate less aggressive players in the endgame.

7) Interplayer effects are also valuable. Allowing players to make contributions to the experiences of other players is always going to take a load off your shoulders. Still, this is a delicate and difficult thing to build. I don't really have any advice as to what mechanics you should use.

One thing to be careful of is player obsession. Obsession is a major driving factor, and it's largely fueled by the amount of stuff the player can do when he gets obsessed. Which is typically when he's in his bedroom, alone. So focusing on interplayer effects can really raise the barrier if he has to talk to other players. On the other hand, content such as character sheets, flash programs, diagrams, and snippets of text are all available at such times, and are great at validating a player's obsession.

8) Enough space to physically split into groups, as being too close would produce too much crosstalk. In Inhabitants, this was a large room. In the card game, this was "all of campus".

I hope this has been an interesting article. There's more to say, but that's plenty for now.

ME3 Review

This review contains spoilers for Tali's mission, and extremely minor spoilers for everything else.

Let's talk about Mass Effect 3, one of the very few AAA games I played and enjoyed in the past year. This is a review in the literature sense, so don't expect a rating or fanboy gushing.

Firstly, I'm not playing to get a perfect. I'm pretty much playing through with a "whatever happens, happens" mentality. With that in mind...

One of the most irritating things about the ME franchise is that it imports your saves... but if your saves were on a different kind of machine, there's no easy way to replicate the choices you made. This wouldn't be a huge problem if canon-Shepard didn't always make precisely the opposite choice that I made in literally every situation.

For example, it's revealed that canon-Shepard killed the last Rachni queen. Really? My default personality is "unapologetically murders helpless species"? You want me to play a character that literally killed a whole sentient species while it was helplessly imprisoned? After it had been tortured for years and was talking, lucidly, about just running off and finding a planet somewhere? You want me to play someone who unabashedly murdered a torture victim and the whole sentient species they belonged to?

As personality traits go, it's hard to think of a more horrifying one.

Well, whatever.

Playing the game, I'm trying to figure out why I'm only touched by certain scenarios. The writing is pretty solid, but a lot of the things that are obviously supposed to be touching just aren't. But others are. What is it?

After a while, it became clear: I was having a really hard time empathizing with anyone that had a face. The facial animations are still so... weird and creepy. Ash and that white-hexagon-ass psychic lady especially. This was true in earlier ME games as well, but it bothered me less for some reason. The female faces are especially creepy: the men are okayish, but the women all strike me as zombies. And there are a lot of human-faced women in this game - around half the cast. The other half is filled out with all the men and the women who don't have human faces.

So I was running around the game world empathizing with the fat little environment-suit aliens and the walking tanks that say their emotions and the Turians, and along comes Tali's mission. The Quarians against the Geth, HUGE SURPRISE.

I pretty much already knew everything that was "revealed" about the Geth, because it was painfully obvious in the previous games. However, whoever animated the helpful Geth was a real master of conveying emotions without using the face.

Once again, I renewed my knowledge that the Quarians are total monsters who basically want to murder their own children. And me being all paragonny, I say "well, I'll go in and broker peace, somehow. Tali's my favorite character, so I'll try to get her people to be less horrible."

Unfortunately, the writers for ME3 made the Quarians incredibly aggressive, genocidal to a psychotic degree, and completely unwilling to listen to the two people who had literally saved their species twice already and were actively saving it again (me and Tali).

As a result, I basically said "fuck the Quarians, I'm siding with the gentle, intelligent, promising young Geth. They've been given a bad rap for defending themselves against psychopaths."

Unfortunately, the Quarians were ever more stupidly aggressive and unwilling to listen to the people who were actively resolving the situation, and in the end I had to murder an entire species in order to save the Quarians from the catastrophe they had created. For the fifth time this freaking game.

Every even vaguely intelligent NPC is like "oohhh, you should have found a way to save the Geth, you're awful!"

Here's the problem. There is no dialog option for "fuck off, you don't understand shit".

I'm sure there was a way to save both, hidden somewhere for people with a walkthrough or enough time to try every permutation of the event. That's what they are saying. Like when Eve died because I didn't, I dunno, talk to her enough or something, everyone kept harping on it. "If only Eve hadn't died!" Because Eve can live. That's what they're telling you.

But that's not the game I'm playing. In the game I'm playing, Eve died, and I can't rewind time and change that.

In these games, they give you a "paragon" option or a "renegade" option. Which is slightly better than "good" vs "evil" but still not very good at representing your character. There is no ability to tell people how you are feeling, or that they are being really horrible.

Because here's what I wanted to say to the Quarians after the mission: "you're all monsters. You forced me to murder a species in cold blood because you couldn't be assed to fucking listen to me. I've saved you from your own catastrophes at least half a dozen times, and you continue to cause them. You would all be dead if I didn't like Tali this much. Get off my fucking ship."

Hell, I didn't even want Tali on my ship after that, but I couldn't see any option to kick her off.


I sincerely doubt that the writers of ME3 intended for this to be a huge affair for the player. They don't seem to give a fuck about genocide between two species without human faces. To me, it's a big deal, and the Quarians are clearly the monsters.

So, ME3: poignant when it doesn't mean to be. Still has pop-in textures and very stiff facial animations. Pretty good, although the battles are definitely oriented towards heavy-gun Shepards.

Wednesday, March 07, 2012

Comics (Technical)

Most comic fans acknowledge that "webcomics" are changing the way comics are seen. Not just in how we read them, but also in what they can do.

For example, McCloud talks about an "infinite canvas". Comic "pages" exist because comics are thought of as books. On a computer screen, you can scroll easier than you can flip.

Unfortunately, this remained mostly a theory. The infinite canvases made by English speakers remained wobbly proof-of-concepts, interesting as art but not really as comics.

However, there is more in the world than English speakers. For example, there are Koreans. You may not know this, but Koreans have a very healthy webcomic presence:

Just in case you don't speak Korean, if you want to see the comic, here's how:

Click on the thumbnail of a comic. This will pull up the summary. To the right of the new thumbnail are four buttons. The first has a green star, the third has a green word that looks like "me". Click on the second button, between them.

(Naver is basically Korea's Google, but the differences are very pronounced.)

You're free to click around as you like, but here's a specific example. Depending on your screen resolution, you may want to zoom out quite a bit: that's an infinite canvas, and it's more suitable for phones than monitors in this case.

Properly scaled, reading it is pretty easy - just a matter of panning. On many devices, there is a "continuous pan", rather than the scroll-scroll-scroll (pant pant pant) scroll-scroll-scroll that you might get using it like a normal web page. I find it is better with a continuous scroll, adjusted to your preferred speed.

Some people will consider this kind of infinite canvas a "cheat", since it is more or less just pages stacked vertically at first glance. But this is more than that. The layout of the panels is completely rearranged to serve a infinite vertical canvas: there is very little horizontal read, so essentially it has become a long line of vertical panels. But the panels are not simple, packed-in panels like in some American experiments: use of white space, buffer panels, offsets, and other simple techniques give the column a flow and rhythm that is pleasant.

Other things you will notice: the colors and tones are screen-friendly, not paper-friendly. The image is extremely resize-friendly, with straightforward linework and large lettering that won't vanish into artifacts when shrunk.

I bring this up because this is much closer to what a webcomic really should be. English-speaking webcomics tend to be "comic books on the web" or "newspaper comics on the web". Their flow is rotten. They succeed in spite of the format, not because of the format.

I can see a few other things happening for webcomics, as well. A few are already happening: comments sections, forums, contests, fanart - all are ways of using the connectivity of the web to enhance your comic (and your readership).

More innovations are coming. Webcomics with "open" and "closed" chapters for tiered readership. Webcomics with tie-in games, multiple comics that share the same world, comics that are auto-generated based on player scripts, comics which are also virtual worlds.

But here's the thing: trying too hard is bad. A big part of why we couldn't make infinite canvases work is because we put way too much thought into the possibilities. These things only work when they are effortless. When you focus too much on a new possibility, you end up making art that happens to look like comics, not comics.

Anyway, look through the Naver comics section, you'll get what I'm talking about. These are properly webcomics. It's a very energetic community: you're going to see a lot more Korean comics and cartoons in the near future, because they have a leg up on how to actually make webcomics "webcomics".

Monday, March 05, 2012

Scut work science

I've been thinking about scientific progress.

There's a lot of scut work in science. That's always been true - just look at how many freaking beetles and finches Darwin collected.

Ideally, the scut work serves to clear the path to a final "product" - whatever that may be. If you do the scut work wrong, some scientist's project fails because of the bad data you submitted, and they angrily punch you right in your reputation.

However, recently we've been seeing the scut work get politicized and profitable. At some point, people began to think "well, the important thing is what is true, and that's what science can tell us".

Ehhhhh-ehhhh. That's the scut work. "True or false" is scut work. "How does this fucking thing actually work" is what science is about.

So we've got all these scientists getting paid to find out whether something is true or false. Most famously, drug trials: every new drug is apparently effective.

Of course, that's bullshit. The truth isn't that every new drug works best, and the truth isn't that every new drug works worse. The problem is that whether a drug works better or worse is politicized. Everyone who finds the drug works gets published, everyone who finds it doesn't, doesn't.

And who's gonna call you on it? It's not the kind of scut work where someone else will try to develop from that data and find that it is clearly wrong. There's no further development built on that data. It's just marketing data masquerading as science. "Cover our ass" science.

To me, that's a big problem. Obviously, due diligence is required when creating products, especially medicines. But paying scientists to do scut work while at the same time removing the layer that checks their work? That's a recipe for disaster.

That's like telling a kid you'll pay him a dime for every question he gets right on his tests at school, but instead of looking at the tests, you just ask "how many questions did you get right?"

He'll probably do it earnestly for a while, but when he realizes you never check how many he actually got right, he's going to start getting a surprising number right. Maybe it'll eventually bite him in the ass when grades come out, but even then you'll just yell at him a little.

... That's my opinion of "cover our ass" science. I'd really like to see us pay a lot more attention to the scut work, and whether it is actually worthwhile, useful, build-uponnable... or whether it was the one trial out of fifty that just happened to get good results and was therefore the only one published.

As ever, the goal with science is to figure out how things work. Trials and tests serve that end. Any trial or test for another reason isn't science, it's just due diligence and/or marketing.