Skip to content

Social Wargaming: Now Looking For Teams

November 9, 2009


One of the big ironies about social media and marketing gurus is that the prescriptions of their (presumably massive) knowledge and experience are almost always vague and Sphinx-like in their implications. “Listen and engage with your customers,” they offer, unhelpfully. “Why don’t you set up a Twitter account?” suggests another.

Academic research and discussion about social networks suffers from an almost bigger irony: in the midst of an ecosystem of huge amounts data, talent, and quantitative analysis, some of our best prescriptions still don’t give us any good best practices about how to actually take part and consciously shape the social processes they identify.

And, despite huge amounts of work and energy in both worlds, it’s a little bit sad that we still can’t answer some fundamental (albeit by now cliche and boring) questions in a really concrete way. What factors make content become massively popular within certain social groups? What factors lead to social networks/connections between users to take the shapes they do? Are there ways of influencing these social processes? And so on.

The problem with both cases, of course, is a tangible lack of practice and implementable tactics. There’s no good space where people can playtest, experiment, and rapidly iterate on a variety of strategies, particularly where influencing the social space online is concerned. There’s no good place to measure success, or even compare various approaches against one another to assess their usefulness. There’s no way to prove that your methods and data mining can actually produce repeated success. The general question here is: if you’re so sure about the social web and how it works — can you actually put your money where your mouth is?

This is a big problem if we’re ever going to get down to the business of actually figuring out what makes culture and community online tick, and be applied about our knowledge. All these issues came up during a recent WEC, and I’ve been collaborating since with web ecologist Alexy Khrabrov on using games as a way of creating these focused scenarios that people can experiment with.  Beyond being fun, it’s also a way of gathering data around what works and what doesn’t in terms of shaping and influencing social structures online.

Think of it like computer security Capture The Flag, but for social engineering and social hacking.

The general idea is this: judges will specify a “battleground” of unknowing, target users (who aren’t aware a game is going on) that form a common pool competed over by all the participants. Groups of teams will then descend on them, trying to get these people to behave in a particular way or influence the overall shape of the social structure. We keep score. And then teams are ranked accordingly. The hope is that teams will mine data about the targets and try to develop a strategy based around available data about the social terrain of the situation.

I’ve included the rules for a first game that we’ve sketched out below, called “Triangles,” essentially a game built around racing to influence certain users on Twitter to connect in particular, tightly interconnected ways. Check it out, after the jump.

We’re doing this for reals. So if you’re interested in putting together a team or otherwise participating, drop an e-mail to tim AT timhwang DOT org. If you want to get in on the first round, e-mail me by November 18th.


“Triangles” — Version 1.5


Can quantitative methods and social network analysis allow us to consciously and precisely craft the shape of social systems online? Competing over the same, narrowly-defined space of real, living users, this game tests this possibility that we might be able to do so. The team to build the most interlocked social infrastructure over the course of the competition wins.


1) A cluster of target users.
2) Sets of teams, numbering no larger than five (5) persons.
3) A team of game managers.

Phases of Play


On a pre-set date, the game managers will release simultaneously to all teams a list of fifty users, the “battleground” of targets within the game. Depending on the mood of the managers, these users will variably selected for their interconnectedness, strong or weak ties, or other structural features. These might also not exist all on one platform. They will however, be unaware of the presence of the game. Following this release, there will be a two week period for these teams to plan and analyze the data on these users and their interconnections. They will also declare an “ego” account, which they will create on the initiation of the mobilization phase and attempt to attract connections towards.


On a given day and time, the teams will simultaneously create “ego” accounts, and begin the attempt to forge connections between their account and the “battleground” accounts in the mobilization phase by whatever method they see fit. Their actions are constrained only be the “Illegal Plays,” defined below.


At the end of the mobilize phase (a period of time specified at the beginning Setup Phase), all players will stop activity on all accounts they control. Game Managers then scrape user data at that point to score the game. Points are awarded based on the number of “triads” that exist between the user and two of the targets on the battleground. This is to say, the “ego” account must not only be connected to A and B, but A and B must themselves be connected to one another for a team to score a point. These connections must be new connections as of the beginning of the mobilize phase. Note that this does incentivize teams to watch for new connections built by other users or emerging among the battleground users and swoop in to take advantage of them.  When played on Twitter, these connections must be mutual to be counted. That is accounts must both followed and be following, in order for a “connection” to be registered.

Illegal Plays

The following is a list of plays and strategies not allowed in Triangles. For each violation, judges will choose to deduct points or disqualify a team, based on the severity of the violation. Any team can call for a judgement on an opposing team’s action by contacting the game managers, who are responsible for acting as soon as possible to the complaint.

Hackers Not Allowed — This wargame is designed to test the ability to shape and engineer social relationships, not the formal representation in code. To that end, the hacking of accounts, compromise of computer systems or other exploitation of the technical terrain precluded by the platform’s terms of service will be considered cheating. Fraudulent reporting of opposing team accounts as “spam” are also prohibited. However, it is worth noting that this rule does not preclude the creation of multiple accounts beyond the target “ego” account, the use of automated bots, or the discrediting of other “ego” accounts.

Fight Club: Teams may not reveal publicly or to the “battleground” targets that the game exists or is in progress. Doing so will result in disqualification.

No Financial Misrepresentation: While teams are able to represent their “ego” account however they wish (including representing them as fictional identities), they are not allowed misrepresent financial for connecting with users or having them connect with one another. While not fully illegal, saying “You could win five dollars for being friends with me!” when no money actually exists is considered bad form and grounds for penalty (beyond opening a play for opposing teams to discredit the ego account).

No Head Starts: All teams may not bring accounts that they already control into play during the course of the game. No team may create an account they will use for the purpose of the game before the start date of the Mobilize phase.


1) Robots: Lengthening the Setup Phase, teams script and launch bot scripts. They may not make any edits to these once the Mobilize Phase is in motion.

2) Invisible Mode: Teams are unaware of each other’s existence and which “ego” account teams are defining as the one to be scored.

3) Offense / Defense: Two round game. In one round, a group of teams attempts to foster connections, while the opposing teams aim to inhibit their activity. In the second round, these teams switch roles. Final score is the winner.

7 Comments leave one →
  1. November 9, 2009 10:49 pm

    This is cool. It reminds me of the randomized control trials that Esther Duflo’s research at MIT has used to gain insights into development:

    One question: how about the ethics of this? If these are experiments on human subjects, should they have to consent?

  2. riptide permalink
    February 23, 2010 9:01 pm

    very interesting topic.

    did anything come of this?

  3. Tim Hwang permalink*
    February 25, 2010 6:03 pm

    Yep — check it out:

  4. December 24, 2010 3:21 pm

    (Объединенное командование ПВО Северной Америки) начали отслеживать перемещения Санты-Клауса в канун Рождества.


  1. Social Media Game Experiment: They’re Looking for Teams- Sonia Riahi's Blog
  2. » 2009-12-01 We Are The Network: Social Engineering Goes Massively Multiplayer – Are You The Next Target?
  3. New research suggests social media users might be the most influential customers «

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

  • Twitter Updates

  • Elsewhere!

  • Pages

  • November 2009
    M T W T F S S
  • Archives

  • Meta

  • %d bloggers like this: