Playing Together: What Games Can Learn from Social Software

NLGD and Develop, 2008

There’s a strange piece of slow glass between the web industry and the games industry. It’s strange because whilst stuff passes through it in both directions, it seems to move faster towards the web, and slower towards games. In the past 2-3 years, there’s been a lot of talk in web software about making applications more playful and gamelike. Now that people have done that – and succeeded – I guess it’s an appropriate time to return the favour.

So who am I and why am I talking to you?

I’m a web software designer and developer. I’m interested in some of the more human ways we interact with one another and with our tools on the web. I’ve been involved in the social web, in one form or other for a long while now; I’ve been blogging for over seven years, and for about five at And I’ve also played a lot of games… and I’ve been playing for quite a while. So whilst I’m not directly inside the industry, I’m an informed observer and critic of it, and I think that’s a useful perspective to have.

So what do I mean by “Social Software”?

In the synopsis, I used the phrase “Web 2.0”, but I’m trying to shy away from that phrase, because it’s a loaded term now, and I don’t want you to think it’s necessarily anything new or a specific thing: it’s a name that has been given to an emerging trend.

That trend is of a web of tools where the user is just as important as the product owner; where creativity is valued and encouraged; where people are connected to one another through the things they share (be it blogposts, or photographs, or wiki-edits, or event-attendance, or book-collections) and empowered by the fact that data is public, and shareable, and re-usable. It’s the web becoming a platform – of “small pieces, loosely joined” to quote Dave Weinberger – and it’s very, very, very popular. I also want to make it clear that we’re not just talking about “social networking” because, really, pure “networking” is a long way from being the most interesting thing people do on the modern web. The key word, for this talk, is social.

In this talk, I’d like to look at some of the understandings that have emerged out of social software, and what they might mean for games. Where I provide examples: they’re almost certainly not the only available examples. Their sparsity is about keeping this to time, rather than turning it into a big list!

To start with, I’d like to consider map of the various social levels we interact at, with this diagram radiating out from the individual.

In the middle of this diagram is me.

Next to me is another: a single gaming buddy, a friend, my partner.

Around us is our close friends: a core group of friends, or a regular gaming crew; perhaps a small clan or guild.

Next out are slightly less close friends, of whom we have more – a larger clan/guild, our Live friendslist, our friends from a gaming forum.

Next out is “everyone you’ve ever known”. This is pretty much the Facebook model of the world for most of its users.

And finally, there’s the rest of the world; strangers.

These are all present in our social tools – from the direct messaging of email to the global reach of your average web page. And the social web actually makes most sense at the middle levels – targeting the smaller groups-scale, between alone and infinity. Because, when you think about it, everything we do happens at these levels. Take a single player gaming experience. Even that has a social life outside of it, with a variety of people radiating out from me:

  • my friends who see me playing it
  • everyone I tell about it the next day
  • the replay I save and show to the world online
  • and if you think of lone gaming experiences – playing multiplayer on your own – everyone else on other screens, watching it.

We should be facilitating activity for small groups, not for individuals.

So really, we should be facilitating activity for groups, not for individuals. And not rigidly, locked-down groups – guilds and clubs and societies – but the loose, flabby, overlapping groups that we really interact in. And most of these are those three middle layers: the smaller ones. Not the everybody I know level. That doesn’t work.

[At this point, I talked a little about Robin Dunbar and Dunbar’s number, talking about upper limits on feasible groups. This is well-worn ground on the web, and I’m not an expert, so it’s easiest for me to point to some Wikipedia pages. My main point was: tangible research exists to examine upper bounds for groups]

MMO servers are currently limited by technological sizes. But what does a server where you know everybody (even a little) look like? What if we limited group sizes for social, not technological reasons?

We wouldn’t have these servers that resemble sprawling cities; we’d build villages, and hamlets, and travelling camps. And as they grew too big, or small groups within the community grew, they’d break off and move to places where there’s more space for them. In one sense, that’s a more natural metaphor for our users, and it’s a scale that makes sense for them.

I like this idea that everything, fundamentally, happens at a group level.

“Single-player games are a historical aberration”

Raph Koster made this apt observation in his GDC talk of 2007, and he explained it thus:

“very soon, all single-player gaming will happen within a multiplayer context of connectivity, persistent and publicly visible profiles, and awareness of other users”

Context is the important word here. Taking the actions of an individual and giving them a wider context. The individual’s actions are pushed out into, to use a situationist term, the “spectacle”; an ongoing world of representations.

“Social action takes the form of the action of objects”

Karl Marx said this. It’s important to consider that once separate from us, these objects have a life of their own; they become what Jyri Engeström calls “social objects“.

“The fallacy is to think that social networks are just made up of people. They’re not; social networks consist of people who are connected by a shared object.”

Jyri explained the importance of these social objects thus. His point is: successful social software pivots around objects. Hence things like Flickr’s focus on photographs, Youtube‘s focus on video, Dopplr‘s focus on trips, Upcoming‘s focus on events.

So what are the social objects of gaming? They’re not the games themselves, but rather playings of games; our experiences of them. And they’re not even the vague notion of a “playing”; the social objects of games are great speedruns, skillful kills, remarkable goals, stylish tricks, and sometimes just entertaining memories that have value to the players involved in them.

Sensible Soccer is just one great-big engine for generating stories. People regale each other with tales of great goals, incredible saves, and their stories are detailed down to the drops of sweat on players’ brows, tugging onto shirts during dirty tackles. Of course, the animation in Sensi is far, far cruder than that – it’s a few frames at most – but the game, and the memories of it, send the game players’ imaginations into overdrive. And it’s that potential to create stories from memories of a great game that gives Sensible Soccer a vast social life outside itself.

Compare that to what EA have done with Euro 2008, the latest installment in their FIFA franchise. In Euro 2008, it’s possible to export replays from your game console to the web. They appear on the Euro 2008 game site as an embedded flash player; you can then send the link to anyone, whether or not they have the game, and show them your great saves or screaming goals. The story is still preserved outside the game, in the social realm – but now we have the actual imagery, not just a fond re-imagining.

EA are doing lots of great stuff in this arena. See, for instance, Skate, which has similar replay-export functionality. That kind of feature makes even more sense in a skating game; skating is a culture built around community, and the sharing of amateur videos is a core part of real-world skating. EA’s replay-sharing feature isn’t just a neat idea; it’s actually a re-appropriation of real skating culture, and it feels very appropriate in the world of that game.

As we perform in this spectacle, our actions become part of a collective history.

So as we perform in the spectacle, and start creating all these social objects, we also start to generate history. This is one of the simplest activities in that simplest of Web 2.0 applications, the blog: as you start posting blogspots, old blogposts recede into history, reverse chronologically down the page, and so – even though you’re making this single object very much in the present – you’re actually contributing to a kind of history, and it’s a history that exist in a public space.

An example of this I like a lot is 360Voice, which is the blog my XBox keeps about me. No, really! 360Voice scrapes data from my Gamertag… and then transliterates it into the first person. Which is cute, and all, but the interesting thing is: it’s taking an activity I performed on my own and turning it into a social object; making it public, making it history. Nice!


And, of course, by being placed into history, these social objects are given a kind of permanence. Permanence gives things import. And on the web, we have a special kind of permanence: the permalink; an addressable form of permanence. A permanent, unique address for a picture, or a story, or a person. Or a playful moment. And it’s this address that lets me share my history with people not present at the time. I can show people who weren’t at an event a picture of it; I can tell people about something I’ve done. This history is social and communal because I can give things life outside themselves.

Some games are beginning to find ways to give permanence to the social objects that emerge from them. Here’s a series of in game virtual photographs I’ve taken from PGR3 and Forza. You can upload photographs from the in-game camera to the web. Unfortuantely, it’s a locked-down corner of the web belonging to the software firm, which means I have to copy and paste the images into Flickr, where they become part of my photostream, mixed in with the other photographs from my life.

And here’s a group on Flickr that’s formed around this; an online community of people uploading pictures of them, driving virtual cars in virtual cities.

EA are doing some really neat stuff around this, with features like their Euro 2008 replays. Here’s a video of me skateboarding, from Skate. Recorded in game, pushed out to the web, where it has a unique URL. It’s become a social object. And whilst there are a few issues to be smoothed out – mainly to do with the conflict between EA’s portal-like approach to their site, and a more open, web-like approach – this is exciting proof that these features are possible to create, and that they add value to our experiences of games.

Go where people are

But there’s still some progress to make.

Why should I have to work around your site to take my memories to Youtube or Flickr? Why can’t you put my data where I want it to be?

When you start doing that – working with existing relationships, going where people are, making history permanent and social – your game stands a better chance of becoming a part of their personal history.

Some games have started to understand this. Spore Creature Creator exports animations straight to Youtube. That said, it’s best not to box yourself in by being tied to a single vendor; it would make more sense if your game exports to your own corner of the web – much like PGR3 and Forza do – but then allow me to export from your website anywhere else.

You shouldn’t box yourself into individual options, though: I’d rather I could export from the Bizarre website to anywhere) than to Flickr in game. That means you’re not tied in to individual solutions; you’re future-proofing the content from your game. This basically comes down to building a way to export to other, defined, public APIs.

And that isn’t actually as hard as you’d think. Most social software plays together well with other tools. Specifically, most social software products now have some kind of public API that allows amateur – and professional – developers to connect services together, or to their own utilities and tools.

So this means that whilst content might originate in a single place – your game – it has the potential to end up anywhere on the web.

“Design for the next largest context”

That reminds me of this quotation from the Finnish architect Eliel Saarinen that I’m fond of. So you design a chair knowing that one day it will be inside a house, and house knowing that it has to be part of a city. It’s the same with social software, and perhaps it’s the same with games: games are becoming citizens of the wider world outside their own borders, and we need to work out where the affordances that allow them to overlap with other products lie.

“Things that happens in groups are really interesting”

I want to back up a bit and talk a little about Groups again. We’ve diverged a bit. Specifically, I want to talk about the potential Groups offer for “metagaming”. Robin Hunicke put it very well in her talk at LIFT in Geneva this year (which is available on Google Video, and if you haven’t seen it, you should) when she said this quotation.

So the thing is: when we form up into little clusters or groups, we start messing around. We develop private in-jokes. We do things that are fun. I mean, some games are only fun in groups, right? Pub Cricket is an example. So’s I-Spy. Objectively, those games are no fun at all. But when you’re on a long boring car journey with friends or family, it’s a laugh because you’re playing together. This is why, as Robin explains, the stupid Facebook Zombies application is really popular. It’s not because it’s good, or because people like pretending to be zombies. It’s because they like pretending to bite their friends.

This happens on a lot of the Flickr groups. Here’s a group called 3-2-1. It’s a group where people post their best pictures and try to get better at being photographers. There are a few rules: you can only post one picture a day; for every picture you share with the group, you must comment on two others; and for every picture you share with the group, you must look at three.

Here’s the interesting thing: some of these rules can be enforced. So, the one-a-day thing is enforced by software. But it wasn’t to begin with. To begin with, the groups had no limits like that. But people started playing games like 3-2-1 with them, the group admins realised they couldn’t delete all these photos by hand – their groups were just too popular – so they wrote little scripts to do it for them, and then Flickr realised that loads of people want to do stuff like this, so they built the limitations in as an option. Of course, they still can’t force you to comment on two or look at three, but there’s an important thing to be learned from this:

Rules that began as social agreements can become codified over time

Look at cat-and-mouse, the hugely popular “unofficial” race mode in Project Gotham Racing 2; back then, it was word-of-mouth-agreement between participants. In PGR3, it’s a built-in game mode. And that’s about as rapid as feedback gets in the retail-product AAA industry. But think about how much faster you could turn things around… especially if your product isn’t packaged, isn’t compiled, is never finished!

What’s fun is that Flickr isn’t an engine for playing 3-2-1. It’s an engine for sharing photographs. But when people started clustering together, they started playing little games within the framework that was available to them. You don’t need to encourage this kind of behaviour, if many of the Xbox Live games I’ve been are anything to go by; it’s not a software problem. But it is a people problem: you need to understand that what might look like messing around to you is another kind of play for the people involved, and you need to build an understanding of how people play that encourages, supports, and (if possible) reacts to the kinds of play that will emerge.

“Social complexity increases based on economic participation. You need people to be economic participants even when not actively present”

More Raph. Raph Koster gave a talk on some ideas from Game Design at O’Reilly’s Web 2.0 Expo last year, and one of them was entitled “Play when not there”, which he explained thus. Right! Just because somebody’s not about doesn’t mean they don’t bring value to their playing-group – guild, clan, crew, mates, whatever. Indeed, even if their avatar or character isn’t there, their influence will be felt depending on the social capital they have – “man, Kim would have loved this bit.”

This is Nike+. It’s really simple: a little wireless pedometer in your shoe, sync’d to your iPod. You listen to music and run. When you sync your iPod, it lets you upload your run to the web, and your stats go online.

You can set up small groups and challenges, so you can run “against” your friends; you can see who’ll be the first to run a distance in a limited amount of time, and so-on. It’s time-attack for your feet! But this is brilliant: you can run with friends around the world without being together, and whilst you don’t get the social aspect… you do get the social challenge.

And what do you discover about Nike+? You discover there’s a metagame to it. People start syncing late – filling up their run data and then only syncing at the last minute – to disguise how much they’re doing. They mess around!

Nike+ is ticking so many of our boxes: it’s asynchronous; it’s designed perhaps best for small groups; it turns running into a social object, putting it online. It’s a really great example of future for social play.

And it goes where I am: it’s a game that I don’t have to learn how to play. I already know how to run. And that leads me into my next topic, you see.

Hard Fun and Easy Fun

These are two terms outlined by Nicole Lazarro in her paper “Why we play games”. I like these two the most. So look at Nike+: the Nike+ service is all “easy fun” – it’s little bits of statistic game and metagame that you can grok very easily. Where’s the hard fun in Nike+? In the running. Running is hard.

Look at all the Web 2.0 properties, and what you find is that the hard fun exists outside the product. Flickr itself supports easy fun; the hard fun is in taking better photographs. There’s still hard fun involved, but players can learn the hard fun at their own speed, where they already are, and in their own time – there’s no pressure to learn eg. twin-thumbstick look at the same time as being yelled at by fifteen eleven-year-olds. The hard fun in Halo is the game – or, for many people, the interface. It’s not that these more casual, more social experience don’t have a “hard fun” component. It’s just that they encourage that hard-fun component without demanding too much of it. Flickr supports both the new photographer and the hardcore amateur. Hardcore, to paraphrase Koster, isn’t about skill-level but committment. Lots of Flickr users are hardcore at the easy game, without necessarily being the best photographers. By encouraging easy fun, we tease out the hard fun from the users.

Richard Bartle’s seminal paper “Hearts, Clubs, Diamonds, Spades: Players Who Suit Muds” also reveals something about why social software places so much emphasis on easy fun. Specifically, this point when Bartle maps his “four kinds of players” who suit MUDs onto this graph.

The hard game of most games exists in that top left quadrant: acting (ie, performing within the rules of the game) on other players. Everyone else is either acting upon the infrastructure – the world – or simply interacting – poking, playing – with one another. And the thing is: way too many games focus on “Killers” – people who are really really good at learning the rules and acting on it. There’s a finite number of Killers in the world. By contrast, we really need to think about designing products – games, social software sites, whatever – to focus on those Socialisers, Achievers, and Explorers. And as we move towards the bottom of that graph, we move towards easy fun.

And easy fun often goes hand in hand with creativity; socialising, and exploring come naturally in the sandbox. The social web exists as a platform for creation, and encourages creation in all its forms. By doing so, it encourages the kind of easy fun that goes hand in hand with casual play.

The best thing about easy fun is that it’s easy: people don’t notice it happening, they just enjoy it. Too many games – especially at the massively multiplayer end of the scale – focus very much on hard fun, and compound this by making it difficult to have “easy fun” in those worlds. “Easy fun” is something all designers should be targeting more.

One of the greatest misnomers in gaming is that products end when they’re released. Of course they don’t! They’re only beginning! If you look at every successful Web 2.0 property at the time it began, you’ll see something markedly different to how it is now. You might even see something that’s truly awful. Most successful social tools have become successful by observing how real people use them, and adapting to their users. That’s easy on the web, where our release cycle can be brought down to a matter of minutes. But it’s something that games need to consider, and work out what ongoing improvement might look like for them. Downloadable content is too-often used as a way to offer vast chunks of new functionality; why isn’t it considered a steady stream of minor improvements, as, to be fair, it is in MMOs?

Valve have done some interesting stuff around this: their patch for Half Life 2 Episode 1, to ease the difficulty of a particular section, came out of their stat-tracking on Steam. They’d tested the game and believed it to offer fair challenge; when they saw that a drop-off in play correspond to a high level of deaths on a particular map, they rebalanced the game in a patch. That’s very weblike thinking, but that’s a good thing. It improved the quality of the game for the end-user.

The communities in successful social software applications start at the developers and radiate outwards; there are few Web 2.0 properties which aren’t consistently cared for. Part of this is cutting down the time to live – single-command deployments, automated test suites, placing ownership for feature into dev’s hands. Part of this is learning to relax a little and forget the boxed product, the release date, in favour of the long-term relationship that you seek to establish with the user or player.

Of course, one has to be careful. The rules of a game are a contract between the player and the game, and changing them without the player’s knowledge could be considered unfair, or even harmful. It’s also worth considering at what point the game might have changed so much as to be unrecognisable. Breaking that contract with the player is a very, very bad thing to do, and you shouldn’t be using patches to do that. Also: consider that almost nobody reads release notes, or EULAs. Given that, what are appropriate changes, and what aren’t? What’s logical growth, and what’s radical rebalancing?

I’ve been thinking a lot about this section of the presentation since I first delivered it at NLGD, and I must admit, my perspective has changed somewhat, especially when it comes to contracts. This is perhaps the section where I convince myself the least – but I think there’s something in it, so I’m going to keep thinking on it and pursue it.

Time to wrap up with some conclusions. I could go on – and probably will if you corner me later. I think these are all interesting questions to ask – and I can’t wait to see the answers to some of them.

Time to try answering some of these questions, too. What would a MMO based on some of these ideas look like? A game where servers are small, where everybody knows everybody else; where guilds not only play together but earn points together, and where you can earn points from your guild without even being logged into the game; and a game where genuine social captial – based on communication, grace, wit, and the kinds of interactions we see on the social web all the time – was valued above all else?

Well, first we need to consider what kind of world might support that kind of play.

The answer I came up with was an MMO set in the 19th-century Victorian Novel, perhaps based on the works of Jane Austen.

Guilds are now families: groups of players who play together whilst fulfilling different roles. Most importantly: families earn XP as a group.

I was wondering why you might want to play a female character in this world – women being more social, mentally active, characters, rather than characters who do activities that translate well to games (such as farming, warfare, and property management) – and the answer is: you want to play as a social player. And the key social manoeuvre for a woman (in this era) could be argued to be marrying well. So: players earn (and lose) XP, points, score, however we define it, at a 1:1 ratio to their spouse.

Like so. And then it’s exponential from there on out.

Children and parents earn 1/2 XP; grand-children and grand-parents earn 1/4. What does that do to the gameplay? Well, consider divorce. Divorce carries a massive points penalty; it’ll ruin a husband and wife, and have serious consequences for their children and parents – but their great-grand-children probably won’t be affected too badly. Similarly, a husband victorious in war can return knowing his whole family – particularly his wife – will feel the benefit. And for those who choose to play as single characters, outside the family system – well, it’s harder to score vast positive points (due to not having the support network in place) but less risky.

This means that players who are less active – playing passively, more interested in the social side, or just logging in less – can still do well at the game, by playing together with people they trust. It’s in everyone’s interests to play well together, but there are also canny ways to group power-play.

What else changes? Servers, restricted in size for social reasons, rather than technical ones, become villages. Players are free to move between villages – and even lives – for the purpose of good narrative, rather than because of gameplay mcehanics. Communication and genuine social skills are valued highly. Characters are forced to live many lives, as they age rapidly: they may play an old patriarch, and later be reborn into his grandchild’s shoes. The players play many histories, and those histories are persisted in the gameworld and the web.

It’s just an idea, of course, and not necessarily a practical one – but it’s an interesting design exercise: an exercise that’s more about considering the thinking necessary to get there: to look at games as simply another social interaction that has to fight for position in our daily lives.

The social tools and toys of Web 2.0 have all taken inspiration, somewhere along the line, from other playful experiences. And now they’re consuming the playtime of both young people and adults – time that might otherwise be spent playing videogames. What do they see in these experiences that they don’t see in modern, mainstream titles? Perhaps some of the things I’ve mentioned today.

There are kids growing up today who won’t know a world without Facebook, a world without the ability to write and publish and create in a global space just because they want to. What do the games we design for them look like? Perhaps, by answering that question now, we might make better games not only in their future, but in their here-and-now.

Thanks very much.