Saturday, June 13, 2009

Designing Ethical Dilemmas - Slides and Audio Available

Update: The audio should be all fixed up on the slides - sorry for any inconvenience. Please let me know if you have any issues with the presentation. Here is a download of the slides directly too if anyone is interested.

Well, I'm back from the conference (Actually, I never left since the conference was in Madison, but I'm back mentally). My talk at GLS 5.0 went very well, I thought. The talk was on "Designing Ethical Dilemmas" and I've embedded the slides below as well as included synced up audio so you can listen and get the full experience of the talk (The slides are idiotic out-of-context, and probably just plain stupid in-context).

The talk is about 6 minutes and 40 seconds (20 slides for 20 seconds each) and is fast paced. I've also uploaded my notes I used for the talk if you prefer to read those notes instead of listen to me speak. The actual talk starts with the second slide, as the first slide is just a title slide.

I have to apologize for the sub-standard audio quality - the only microphone I own is attached to a rather crappy Logitech headset I use for gaming, so the quality isn't all that fantastic. The conference will also be making their video and audio of me speaking available in the near future.

Next, I'm going to work on expanding this talk to be about 20 minutes for the Madison Chapter IGDA Meeting next month. So, if you have comments or criticisms concerning this talk, please let me know. I obviously want to improve and start expanding upon the ideas that I included and I welcome any input.

Thanks to everyone at GLS and everyone who showed up.



  1. Good stuff Manveer. Did you ever check out my own talk on the subject here

    Five year old now... wow. Anyway, similat ideas. One thing I am coming to disagree with though is the 'People' point... :)

    I think permanence is something we need to understand better. Games are not neceasarily temporally linear (though they can be). The abiity to save/load is not necessarily something we should reject. That's thinking in old author-centric terms, IMO.

    It's trying to put a narrative model decision-consequence structure into an incompatible medium. It's the converse of saying 'books would be better if you could go back and havethe protagonist make different decisions.'

    It is the nature of author centric linear media that we watch with empathy as heros march stubbornly or naively forward through the mistakes that arise from their characters.

    This is not the nature of games. We need to invent, embrace and perfect an ethical decision making model for games that is compatible with the ability to undo/redo... not try to impose an incompatible EDM model onto games.

  2. Great, I'm 5 years behind the times! I'll check out your talk shortly, I had not seen that specific one yet. Doh!

    So, I guess my thoughts aren't that we need to necessarily reject save/load. But rather, interject within that structure something that matters more. But you are absolutely correct that it is thinking in the author-centric terms.

    I don't roundly reject that model of game creation (yet), though I do agree it's probably not the ideal way for our medium to continue to grow or embrace our own unique features. Either way, it's important to understand that it's predominant now, so at the very least we can improve ethical dilemmas in these types of games while we also try to create games that don't rely on linear narratives.

    At the same time... if I can ever make a decision and undo it, I don't think I'll ever agonize over that decision. I think I'll always be a "try and see" player. And that seems fundamentally broken if we want players to be able to feel these other emotions. This may just be because no one has figured out how to make it compatible with undo/redo so I have no frame of reference on how to consider that, but at this time I can't even conceptualize that in my head.

    Maybe the answer is to have ways to develop ethical dilemmas better in author-driven narrative games that we see now, as well as start developing models for what the future of games will be. Maybe our ethics model needs to diverge there

  3. The biggest issue with any type of programmed ethical dilemma is that it is strictly boolean in nature. One reaction leads you down one path, the opposite leads you down the other. Games which claim to change depending on your choices really are just picking your path down a systematically designed course. The more attention paid to this "dynamic", the more obvious the programming is.

    The only way to provide a truly dynamic dilemma for gamers is to insert some sort of random choice within the choice. By this I mean that even if you decide to help the old lady cross the street, it may result in a smile, a cold refreshment, $100, or you get hit by a bus. There is less incentive to make the initial choice based on reward, because the reward is unknown.

    This, of course, puts more work on the programmer to think of many different results to the choice, but it makes the game's story richer and less predictable (and susceptible to ruin by FAQ).

  4. Programmed ethical dilemma's are not boolean by nature at all. They are boolean by ease of implementation. But that's part of the problem. We too often map the choices to "good" and "bad" and that's just not interesting or even thorough. Yes, even if you have 5 choices, a designer has to choose what happens as you go down each of those 5 paths, but that's ok if those 5 paths are all interesting and have their own set of outcomes that change the game.

    Randomness would instead take away agency from the player, making them feel like the hand of god caused something to happen they could never realize. This would be counter-productive to having the player feel new emotions, as the player is going to feel like they got cheated or screwed over by the game (Unless that is the emotion you want to evoke).

    Well thought out writing, game world, and consequences can get away from the feeling of "obviousness" you point out that exist in many games with boolean choices. There isn't specifically a need for dynamic dilemmas.

  5. I agree with manveer there. Its like saying the problem with stealth in games is that it is inherently boolean (I think the word your looking for is binary though). Obviously in many games there is stealth which is binary. But there are a whole range of games where it was made analog. The same could in theory be done with ethical decisions by doing as manveer suggests in his talk - making a rich and compelling consequence space instead of a binary good/evil flag.

    Of course, this is orders of magnitude more difficult than what mostgames currently do in the same what making analog light/darkness stealth models was orders of magnitife more difficult than 'cloak of invisibility ON/OFF' was in the late 90's.

    Give us ten years and we'll have it nailed for sure. The only question is whether we can do it in 5, or 3, or sooner....

  6. Clint was nice enough to go into more details on his thoughts on his own blog which has spawned a number of comments as well, so if you are reading these comments and are interested in other people's thoughts go visit his post and check the comments there as well.

  7. Good discussion! Some more thoughts on Permanence:

    It seems like the goal with Manveer's proposal is to give a real weight to your actions - to make them feel less flippant. To me it seems like a natural way to do this is to make the act of carrying out your choices not be instantaneous.

    Black and White was one example of this. The choice to be good or evil was more or less binary, but it took several hours of play time to carry out the choice to be good or evil - so in a sense, it got around the save/reload problem because reloading the game from an earlier point wouldn't let you instantly choose the other direction.

    I also think that it's worth exploring the idea of *not* showing you the results of your decisions instantly. Let the player know that this decision has consequences, give them some kind of indication of what they might be, but don't spell it all out the moment they make the choice.

    Lastly, I think it's possible (and perhaps important) to have consequences of moral choices, as well as emotional weight, without necessarily presenting everything as a dilemma. Many choices in life are made almost unconsciously. I think it'd be interesting if some NPC wandering the streets in a game like GTA actually had some story element - perhaps he was the cousin of the guy who runs the car paint shop, and by thoughtlessly running him over you tick off the shop owner off and can't use his services unless you change your outfit. It might make you think more about the little things you do.

  8. Hi Manveer,
    I got here from Clint’s blog – I’m working on a game-for-change that is trying to affect moral/ethical decision-making in the real world, so these discussions are useful for our whole team. You made some nice points in your presentation, but I want to offer up a critique (perhaps unsurprising?) - I’m disappointed at some of the choices for visual content. Humorous contrast to a serious subject is great, but the sophomoric flavor of some humor here detracts from your interesting ideas. For the 20 minute version I’d suggest taking the choice of illustrations (and acronyms?) to a more professional level. That will help you reach more people and also be better for your career in the long run.
    - Heather

  9. Heather,

    There's always a fine line when it comes to humor. I am someone who is juvenile in my sense of humor, admittedly, so that's where that humor comes from. Also, the Pecha Kucha format is rather restricting in this way. I didn't want just headers (I did that in my last talk and got BLASTED by a number of people for it) and I didn't want bullet points as people would read that and miss my talking while they read. The quick paced format of the talk means I need people to concentrate on the words coming out of my mouth and not the slides on screen. Admittedly, I'm very uncomfortable with PowerPoint slides still - I much prefer just talking with no visual aides at all, but most people cannot handle that I've found.

    As a result the slides are meant to be a humorous, light juxtaposition to the serious nature of the talk. In a 20 minute version, that restriction of the Pecha Kucha format is removed. I will, undoubtedly, have plenty of jokes and humor in that talk as well and I can guarantee some of the slides will probably remain. At the same time, I would expect the density of these images to dramatically drop.

    At the end of the day, I view my talks as an extension of myself and therefore my personality will naturally pervade those talks. So, there will also be some amount of sophomoric humor. I understand that not everyone appreciates that and will work on making sure that it doesn't overshadow the larger point of the talk.


  10. Thanks for sharing your thoughts and going to the trouble of slidecasting, Manveer. You've provided some good food for thought.

    It'd be interesting to try and deal with players' moral actions as an analog interaction rather than a discrete one (in Randy Smith's terms). That is, a player's actions aren't hard choices (killed neighbour's cat/saved cat from fire), but cumulative (saved cat, but destroyed the house and swore a lot while doing so).

    This is much like stealth: the guards may attack when they see the player in the open (NPC's consider you an evil jerk after you kill your neighbour), but you can also get in trouble by being a little too noisy, not hiding quickly enough (they're suspicious after you were caught sneaking through your neighbour's house, reading her mail). The game world's reactions to these actions are more like a thermometer than an on/off switch.

    This goes not only to greying out morality in games (placing the player along a selfish –selfless continuum rather than in the "Jedi" or "Sith" bin), but also goes to what @Aaron says about permanence: you're not suddenly, after one decision, more good or evil, but it's something you build up over time. It may even shift or fade over time (when you move to a new neighbourhood).

    It would also be interesting to create a game that represented player's moral actions in ways that do not classify them in terms of good/bad. Why not consider the selflessness or loyalty of the player instead? Further, have mixed consequences: create NPCs that act more favourably toward the player when she is self-sacrificing and others that are dismissive of her for not being sufficiently self-interested.

  11. Can some body tell me what language should we do for game designing?

  12. Hey Manveer,

    I just wanted to comment that I found your short presentation very interesting – especially the idea of permanence – and I quite enjoy the slightly immature humour. It's unfortunate, however, that you had to go and put in Michael Jackson for the black and white comparison slide, given that he's just recently died. I imagine you'd want to take that out for when you actually do the presentation.

    On the subject of permanence, as a player I made a conscious choice a while ago that I would not go through the "make decision -> quick load -> choose alternative" process that I used to as, unsuprisingly, it robbed those decisions of any meaning. I think the only time I disobeyed the rule was in Fallout 3, when I killed so many NPCs that I actually limited my enjoyment because those quests were no longer open to me. In essence, you might say I "broke" and ruined certain parts of the game, but without having the impact being told to me until I looked up an FAQ.

  13. Taking away agency by not giving the player enough information as to what the consequences of the players action will be is kind of like everyday life, no? Less authorial power in some cases simply because the game director won't be able to acount for all the possibilities and won't have tie-ins for all possible choices.

  14. And also, does a choice need consequences in order to be included in a game? Of course, the possibility of dynamic yet long term change is often overlooked and towns in RPG:s become a really boring place once "cleared", but is the trip and moral decision in itself not something imporant? Otherwise it could be another case of moral dilemma being connected to gameplay mechanics, and the player starts thinking; which result of these two choices would I rather see? Do I want to extinct the race in mass effect, or do I want to get help from it in mass effect two?

    Being closed in a box instead, with only you as the judge, actually puts a lot of pressure on you, and the character you're playing. It would be nice if the moral choice is thematic to the rest of the context, of course, but this is a different matter.

  15. Thanks for your informative post.

  16. Your post was quite good than other blogger, so simple yet imformational.

    anyway I'm william
    mind if I put a link back to you?

    (clickable) ------> Formal Wear

  17. This blog is so interesting. I must say that at first I thought it would be a waste of time, but after reading your post was impressed with its quality. I just hope that over time you continue maintaining the same quality, and most of the same passion at the time of writing.

  18. [][/url]
    Nehru jacket
    your raise my mind to read your blog and now i thanks you also for signing me here and i hope you do visit my web thanks..

  19. Try cricket is actually dull and additionally prosaic. Some concrete saw faq proposals to make sure you revamp all the test cricket had been suggested just by former American native indians cricket musician Ravi Sasthri.