Electric Dreams, Part 5: Waking Up

Welcome to the last part of Electric Dreams, a series about the many possibilities for tomorrow’s games, and the technology that might make it happen. Over the course of the series we’ve talked about a lot of different futures for the games industry: an endless graphics race; an exciting world of research; promising experiments in the industry; and a demographic of dreamers. These futures aren’t exclusive from one another. One of my favourite bits of games writing, by George Buckenham, is a list of Rules for Making Games. Rule number 5 simply says “Which future of games is correct? All of them.” Let’s see if we can squeeze in two more futures before we come to a close on this series: my own, and yours.

Writing this series has been an interesting opportunity for me. While I’ve been giving my view of the world of research, and the ways the games industry could change, it’s also come at a time when I’m examining my own reasons for staying in it. As we’ve discussed in previous parts, the power of research funding also comes paired with a lot of baggage and other responsibilities, and while games researchers might be more free than big developers to explore new ideas, we’re still constrained by funding agencies and government visions. If I want to pursue my own ideas about games, if I want to focus on whether my work actually benefits games rather than some abstract notion of ‘the economy’ or ‘science’, academia may not be the best place to do it. But this raises a more difficult question: where else is there?

There’s a gap in games as it stands today: a space of questions waiting to be answered that neither the games industry nor academia can easily reach. We’re familiar with spaces like this – we celebrate indie game development for being able to try out ideas and explore areas of games that are not possible for other kinds of game development. But ‘indie’ has become a blurry term, and within that blur there are areas that are well-explored, and areas that are less so. There are swathes of wonderful artists out there exploring certain aspects of the medium and stretching its definition and its role in culture. Without wishing to partition the world into art and science, I feel like we are missing the complement here: indie scientists, pushing the limits of technology, with an understanding of cutting edge research ideas but without the strings attached that come with working in industry. In theory, academia provides this space. In practice, I have little faith that this missing space will be created by either academia or the games industry as they stand today.

As I wrote this last piece, by complete chance I was reminded of David Hayward’s ‘Videogames are more than an industry’ talk. Hayward is one of the organisational wizards behind events like Game Camp, Feral Vector, Rezzed and Game City. He’s full of positivity and friendliness, and just the right amount of sass about the world. In the video he’s walking across lush green countryside, calmly talking about games and the changes he hopes to see. Towards the end he settles down on a beach and peers into the camera. “I’m convinced that there are more interesting creators and new audiences out there to be found… but we can’t look to the industry as it exists to do that.” Similarly, I don’t think we can look to academia any more to provide all the answers to gaming’s future. This is not necessarily the fault of academic research: it has a lot of problems which weigh it down, but it also has different aims in many cases.

Those aims are often things like ‘economic impact’ or whatever euphemism for ‘help the mainstream games industry’ your local funding agency likes best these days. Which is not to say that I think research being useful to big game developers is a bad thing. One of my favourite pieces of academic work is an account of how Nathan Sturtevant, a researcher at the University of Denver, worked with Bioware to improve Dragon Age’s pathfinding. I think this work is valuable, and is a great outlet for academic research. What you have to understand, though, is that there is no-one to advocate for other outlets. The industry has the money, influence and named faces to come and stand at academic events and proselytise. Other outlets – like having scientists making games themselves, or helping disenfranchised developers, or being relevant to society, or simply doing their research for its own sake – they don’t have companies, governments and entrepreneurs to stand up and shout for it.

Society has a strange relationship with scientists right now. Everyone ‘Fucking Loves Science’ according to Facebook and Reddit, particularly when it offers an opportunity to get prissy about religion or appear intellectually superior to people. But our conception about what scientists do, or who they are, is pretty warped. Shows like The Big Bang Theory reinforce unhelpful stereotypes: scientists are socially confused, detached from reality, and of course male. The end result of these two extremes of love and hate is a weird blend: people want to support scientists but they don’t really want to know what they do. This can be a good situation for people who want to knuckle down in peace, but it’s potentially a worrying situation if we want to turn to this same public and ask them to fund games research. Would you be willing to pay me to explore hypotheticals about the future of games? Or would you rather throw the money into a Kickstarter for another Broken Age?

I wouldn’t blame you if the answer was the latter – after all, that’s why we fund research with government money. Scientific research as a whole is normally very risky work that doesn’t have outputs that society would be directly interested in. I’m not sure that’s always true here, though. In fact, I think we’re lucky as games researchers that often people are very interested in what we do. When scientists talk about impact, we too often tunnel vision on the idea that academics should make big, complex games with budgets and artists and polish, as if people are too stupid or ignorant to actually be interested in the work we do. Yet whenever I talk to people about my work, or write up the research of colleagues, or just show a silly game I made, it’s only ever met with enthusiasm and support. Like David Hayward, I get the sense of an audience out there waiting to be found. Perhaps with them we can find a way to fund a balance between research and game development that lets us answer interesting questions, and release tools, toys and games along the way.

It’s difficult to wholeheartedly wish for these changes sometimes. I spent last weekend at Different Games, a conference about diversity and inclusion in games hosted by New York’s NYU Game Center. There was so much discussion about so many different, yet related, struggles both within and without the games industry. People trying hard to find or make a place within a medium they love, either as creators, players, critics or something else. By contrast, the concern over whether we ever manage to fully generate game narratives, or ever find a way to talk to the monsters, seems somewhat trivial. For all the troubles of researchers the world over, most of us are going to make rent this month, and continue to live without much worry. So it’s hard for me to sit here and tell you that there’s a need to create a new space to work in. I am safer than a lot of people who want to change games and are working hard to do so.

Ultimately, though, I think creating a new space where games research can be done, distributed and explained benefits everyone. In Anna Anthropy’s wonderful book Rise Of The Videogame Zinesters she speaks about the power of technology and how its barriers to entry affect and control how people can express themselves through games. “Where is the printing press for videogames?” she asks. I think part of the answer is to be found by enabling research to happen outside of both the industry and academia. While gaming is an artform, it is also a science, and we must value independence and experimentation on both sides of this coin. I truly believe that in doing so, we can make complex techniques and ideas available to more people than ever before – not locked away in proprietary software, nor obfuscated behind academic nonsense words, but shared and spread with the same community spirit that we see in smaller developer communities.

A lot of this is idealistic, but thinking about the future often is. I hope that despite the cynicism and skepticism in some parts of Electric Dreams, I’ve left you with the feeling that things can change, and that there are people out there who want to do something new, who want to change games and games research for the better. I said at the start that this part was about both my future and yours, and that’s because any hope of enabling researchers to work comfortably outside of academia will depend on whether your own vision of the future has a place for it. We’ll both need to be idealists – and, probably, risk-takers – in order to make it happen. If you want to help out, let’s keep in touch. If you happen to have a few million sitting around and want to start a research institute, then definitely get in touch. If you think I’m totally wrong about games and research and the future, then you can tell me that, too.

I can’t promise anything will happen right now, because I’m not sure what the next steps are just yet, but it’s something researchers I know are talking about with increasing frequency. Could we Kickstart a research program? No-one really knows. But we’re beginning to think about these questions. A few weeks ago, just after the Dagstuhl meeting I talked about in part two, the Virtual Institute for Computational Expression was started up by some colleagues of mine, and I feel very fortunate to be a part of it. We’re hoping that as a group we can find the answers about our future that we couldn’t do alone. Whatever your visions for the future of games are, I hope you’ll keep dreaming with us, and always be looking out for new things. The future, and games, will be better tomorrow. I hope to see you there.

Electric Dreams was made possible by your generous funding through the RPS Supporter Program.


  1. cybernomad says:

    I really liked this series and wouldn’t mind to read more of this game research stuff. Maybe even something more specific or technical like some current AI research or some really cool implementation of a technique. It could help getting more people interested in becoming indie (game) scientists.

    • Mike says:

      I meant to mention in the article: in a few weeks’ time I’ll be restarting a series called The Saturday Papers, where I talk through a research paper and try to relate it to game development or otherwise. You might like it! Some back issues are here: link to gamesbyangelina.org

      I’m actually happy with anyone syndicating the series, as long as they let me know, for free. RPS are happy to use it if they want!

    • Premium User Badge

      Ninja Dodo says:

      >Maybe even something more specific or technical like some current AI research or some really cool implementation of a technique.

      As someone previously unfamiliar with AI, I found this site educational: aigamedev.com

      Particularly this article: link to aigamedev.com

      • Mike says:

        AIGameDev is really great. It does a lot of good work spreading the word about cool games and cool research. I love the balance.

  2. LogicalDash says:

    Games research can thrive without being particularly profitable as long as it is cheap.

    An especially lively area of innovation in computational expression is in the area of data analysis. I don’t mean statistical analysis, not really; the questions are the same, but in data analysis you’re not asking for any kind of proof, you just need an argument based on data, and you need it to be convincing enough to inform a question of public policy, or a business decision, or your budget. Data analysis has existed for way longer than computers and you almost certainly practice it in some form or other.

    Cheap, ubiquitous smartphones now allow people to apply data analysis techniques to everything everywhere all the time. There are a zillion little software tools you can use to keep a “lifelog” and treat your own entire life as a research project if you’re so inclined. There’s an obvious analog in games research, where developers build analytics into their games to track the way players play them and make adjustments to things like level design, weapon balance etc. That kind of analytics is not likely to produce fundamental innovations, but what if instead of analyzing the player, we analyzed the designers?

    In a sense we already do: game code is often kept under version control, which makes it more practical to track down the exact line of code that introduced one bug or another; and there are best practices for design documents to help understand the higher level decisions. Both are fairly laborious. Neither are especially good at capturing anything unplanned. If you want to try out a strange mechanic that is likely to break your game you’ve got to find the time to give it its own game, for a jam perhaps. Or, if you do implement it in your “real” game, you keep that in a branch of its own and never release it.

    Why don’t developers release the strange, failed experimental versions of their games? Probably because there aren’t many who would play them, and releasing anything at all requires some effort, and your publisher might get ornery about it, and holy shit deadlines! The latter two problems are social — indie developers have worked out ways to get around them enough of the time that weird experimental games get released all the time, although you usually only get the one release. But I think the former two might admit technical solutions.

    Imagine a “Netflix for game jams” that gives you some games for you to rate, does some Google bullshit on your ratings, and serves up not only other games that you might like, but the specific releases of those games that match your interests. Some users might prefer games that have the least comprehensible user interfaces. Others, bizarro movement systems like in the alphas of Under the Ocean.

    Or, hell, let’s look at Angelina. It randomly pulls from its bag of rules to make a working game. What if you’re bored of your game and be like, “yo Angelina, mix this up a bit will you,” and it mods your game to do something stupid and random? How cool would that be?

    (psst — modding a game that way is a lot easier if you have a rules engine, you know, like in LiSE)

  3. Premium User Badge

    Ninja Dodo says:

    Was a bit late reading the earlier entries and comments to participate in discussion but just want to say I really enjoyed this series.

  4. rebb says:

    Now for something completely different :
    Is anyone else seeing this huge and jarring “Share this – OR ELSE!” Text block at the top of every article ?

    • Premium User Badge

      Ninja Dodo says:

      I was for a while, but now they’re back to being small icons in the corner. I think there is/was something wrong with the site’s stylesheet.

  5. FriendlyFire says:

    I really do think that the most practical and likely way for this to happen would be through an independent research institute, funded either by a gaming organization or by magical rich patrons looking to broaden gaming as a medium without expecting much, if anything, in return. Unfortunately those people seem pretty rare, and I don’t see the ESA making such an institute. Research also isn’t impactful enough to the average gamer that a Patreon would work out, I’m afraid.

    I heard that some countries have much more flexible funding agencies than others though, perhaps that’d be a possible middle ground? I know this is all a bit useless but I’m reflecting upon this myself a fair bit as a graphics researcher. I don’t want to stay in academia because of the pressure towards just racking up papers at the cost of doing actual exploratory work, as well as all of the paperwork and waste of time that academia suffers from. Yet, graphics research opportunities in the industry are fairly rare and tend to be limited to the narrow scope that the AAA games industry is looking at: flashy, realistic, most of them shooters. There’s very little space to explore procedural generation (which I find is as much of a graphics problem as it is a generation/AI problem), non-realistic rendering, stylized rendering, etc.

    I’ll definitely keep track of your work in case you find that perfect middle ground!

  6. wendy says:

    Earning money online was never been easy as it has become for me now. I freelance over the internet and earn about 81 bucks an hour. Get more time with your family by doing jobs that only require for you to have a computer and an internet access and you can have that at your home. A little effort and handsome earning dream is just a click away…………………….

    • LogicalDash says:

      Huh. That would be a solution to the problem Mike’s written about…

  7. Josh W says:

    I’ve been wondering for a bit about this, specifically on the paradox that big companies have the institutional capacity to take long bets, and do 3+ year research projects, but the inclination to focus on tried and tested ideas, going for the most reliable profit they can. Indie providers, in contrast, have more of a desire to do their own innovative things, but are more likely to be hit by technical limitations because they’re usually in a position to have to make new games every 2-3 years or so to keep cashflow.

    So what if it was possible to refocus the “economic growth” academic funding on things multiple groups of independent game developers are actually interested in? People making games create retrospectives about things they wanted to do – but couldn’t, due to technical limitations – all the time, so as a start you could focus on those things, and start exploring them. Then you could start looking ahead, where people start asking for certain features.

    It might be that you’d need to focus this, creating some sort of money element so that people don’t just ask for things they didn’t really want, but rather than thinking of this as money matching, it could be a kickstarter or assembly style process where wide numbers of developers put small amounts of money behind categories of research problems, and the government then fills in a substantial quantity more. The money bit is just there to show “demand” to funders, as much in terms of numbers as amounts.

    The trick would be that people would work to point out things that are potentially useful to them in making games, and then research and development could fire off, confident that these things could be useful if the research pans out. Developers could effectively use their “real economy” credentials to free up games researchers to get on with looking at these problems in more detail, on more of a 3-5 year rolling program, with different people picking stuff up and working on it, feeding their results back into the general pool of knowledge that developers can work from.