Ok, so, about this game thing…

If you look up into the night sky, just to the right of the bit that looks like a giant shopping cart, you’ll see a small blue star, called Sulis. Around it floats a stormy orange gas giant, and around that in turn swims a small moon, called Selene (until I come up with a nicer name).

selene2Selene is gravitationally challenged by all that whirling mass and hence is warm, comparatively wet and volcanic. It’s a craggy, canyon-filled landscape, by sheer coincidence remarkably similar to northern Arizona. The thin atmosphere contains oxygen, but sadly also much SO2 and H2S, making it hostile to earthly life without a spacesuit. But life it does contain! Spectroscopic analysis and photography from two orbiters have confirmed this (never mind how the orbiters got there – work with me, guys!)

There are hints of many species, some sessile, some motile. And just a little circumstantial evidence that one of these species may be moderately intelligent and perhaps even has a social structure. Your mission, should you wish to pay me a few dollars for the privilege, is to mount an expedition to Selene and study its biology and ecosystems. If at all possible I’d also like you to attempt contact with this shadowy sentient life-form.

Nothing is known (well, ok, I know it because I’m God, but I’m not telling you) about Selene’s ecosystems, geology, climate or, in particular, its biology. What is the food web? How do these creatures behave? What’s their anatomy? What niches do they occupy? How does their biochemistry work? How do they reproduce? Do they have something similar to DNA or does a different principle hold sway? What’s the likely evolutionary history? For the more intelligent creatures, what can be learned of their psychology, neurology and social behavior? Do they have language? Can we communicate with them? Are they dangerous? How smart are they? Do they have a culture? Do they have myths; religion? What does it all tell us?

You need to work together to build an encyclopedia – like Wikipedia – containing the results of your experiments, your observations and conclusions, stories, tips for exploration and research, maps, drawings, photos and all the rest. It will be a massive (I hope!), collaborative, Open Science experiment in exobiology…

So that’s the gist of what I’m working on. I was going to open a pet store and sell imported aliens but I decided it would be much more fun to build a virtual world you can actually step into, instead of watching through the bars of a cage. I’ll try to develop a whole new, self-consistent but non-earthlike biology, building on some of the things I learned from Creatures and my Lucy robot. I’ll discuss some of the technical issues on this blog but I’ll try not to give the game away – the point of the exercise is to challenge people to do real science on these creatures and deduce/infer this stuff for themselves. They/you did it admirably for Creatures but in those days I couldn’t give you anything as complex and comprehensive as I can now, and this time I don’t have marketing people breathing down my neck telling me that nobody’s interested in science.

I have no idea what the actual features will be, or to what extent it’ll be networked, etc. I’m just starting work on the terrain system and I have an awful long way to go. Because I’m working unfunded and have only a limited amount of money to live on, I’m going to work the other way round to most people, so instead of working to a spec I’ll squeeze in as many features as I can before the cash runs out. I know it’s absurd to hope to do all this in the space of a year to 18 months – after all, how many programmers and artists worked on Spore? Something like a hundred? But I think I’m as well equipped for the job as anyone, I work far more efficiently on my own, and it’s worth the attempt.

Whaddaya think?


About stevegrand
I'm an independent AI and artificial life researcher, interested in oodles and oodles of things but especially the brain. And chocolate. I like chocolate too.

62 Responses to Ok, so, about this game thing…

  1. Matt Griffith says:

    That is exactly what I was hoping Spore would have been about. I’d definitely settle for paying for such a simulation/game/exploration/scientific experience. 🙂
    Please count me in on testing/paying/playing! 😉
    I actually have a variety of games under my belt as a beta tester, and I am a hobbyist game developer myself, so you can expect a familiarity in my comments as far as what you might expect in criticism/critiques.

    Also, down the road you might want to consider how payment will work — it could be a service for exploring the world, or just a one-time fee, which comes down to how you consider running it. But obviously that’s something to worry about when you actually have us running around. 🙂

    Anyway, this is very off topic, and I shouldn’t post it here, but I don’t know your email address!
    But my team’s Scientific Balloon reached a new record among Amateur Scientific Balloonists around universities. The press release has been posted into AOL, NBC, Yahoo, Forbes — yet, none of those would be as cool as a post here! 😉
    Sorry if I shouldn’t do that! I don’t win or gain any money by posting it. It’s just exciting! 🙂
    (Our software/hardware beat a team of 19 Cornell/Lockheed Martin engineering students! Woo!)

    • stevegrand says:

      Wow! That was quick. Were you sitting just waiting for me to click PUBLISH or something???

      Yeah, this will be very different from Spore – no building your own creatures, no cutesy stuff, just real biology and emergent behavior. I want it to be a big open-ended puzzle, but with no concessions to pre-ordained gameplay – just a world that exists and it’s up to you to discover things about it.

      What I liked about the pet store idea was that I oculd charge a small monthly fee and it had a long shelf-life, whereas this might be a bit of a one-off event. I’m not sure. But I certainly need to get a proper income out of it this time.

      Thanks for the beta test offer. I’ll certainly rely on you lot to help there (under strict conditions of secrecy, of course).

      Anyway, more to the point, CONGRATULATIONS! 128,000 feet is a hell of an altitude! Was the flight profile interesting? Did you get any data other than position and altitude?

      • Matt Griffith says:

        Actually I had just checked your blog when I got home from work. Good timing! 😉

      • Matt Griffith says:

        Sorry for not properly replying last time. This is possibly my busiest (and last) week of this internship.

        When I said it reminded me of Spore — what I meant was, this is exactly what I was hoping Spore would have been, and also, what it seemed to have been at GDC 2007, before EA told them to “make it fun, not science”. There was a time when it wasn’t cutesy! 🙂

        But yes, emergent behavior, evolutionary transitions, all of this would be amazing. I definitely want to explore a new world that you would create.

        Thanks again about the altitude record 🙂
        The flight profile was very interesting. The balloon flight predictions I made were deadly accurate. We launched it in College Park, Maryland — near Washington DC — which is a very bad idea now that I think about it. It crossed past 5 different military bases and airspaces. We were in constant communication with the FAA and airports though. The balloon sailed straight due south past DC as predicted, then, the amazing part is, the winds shifted due wast almost instantly — what we thought was a fluke in our predictions actually came true. The predictions I came up with based on NOAA sounding data dropped it right next to Fredericksburg, Virginia — and guess what? I landed due west of Fredericksburg! 🙂

        Technology is neat!

        We were able to capture temperature over altitude, but sadly no pressure data. This was a record making flight rather than purely a data analysis launch. We had very minimal hardware parameters to work with. 🙂

        Of which I can’t share for a while, so we can keep this record at least for a little while!

      • Matt Griffith says:

        This week, we have 150 or so high school students (18 of which are ours to work with) who are part of the NFB (National Federation of the Blind). They are all blind students of various levels of blindness, and we are giving them activities to show them that they too, in fact, can change the world for the better. This time around, they are putting the balloon payloads together (Same basic hardware), and we are launching 6 of these balloons together. Which we have never done before. The communication protocol has been a nightmare. If I had time, I would have been able to make it work better. Fingers crossed though. We’re launching the balloons on Wednesday. 🙂

        (Sorry to be off topic once again)

      • Matt Griffith says:

        Oh, I guess you might be wondering how they will experience these balloons? The hardware transmits GPS and temperature data every so often to the ground station, and the ground station has a text-2-speech synthesizer to speak out this data. 🙂

        “Payload A, altitude 10020.4 meters, temperature -10.2 Celcius”

        And so forth. 🙂

  2. Ian says:

    I think this would certainly fall into the category of ‘games I would be interested in’ (sorry, that sounds like damning with faint praise there: perhaps I should also point out I’d be happy to pay for it, too!)

    With regards to financing yourself through development, perhaps you could consider doing what the Wolfire chaps have done, and set up some form of pre-order system with access to alpha builds or such like (http://www.wolfire.com/overgrowth)?

    Anyway, looking forward to hearing/seeing more about this in the months to come!

    • stevegrand says:

      Hi Ian,

      Don’t worry about damning with faint praise – lukewarm encouragement is all I ask, really, because nobody can see what I can see in my head, so you don’t learn much from the scenario. Take this game for instance:

      “Well you have a little 2D world and some furry creature things that can learn a little bit and eat carrots. No, there’s no point to it. You can tickle them or slap them, which may or may not have any effect. Oh, and you can type words to them, but they don’t understand English. Um, that’s about it, really”

      That game went on to sell two million copies, but nobody could see the point until it was close to finished, so it’s not what you do but how you do it that matters.

      A pre-order system may well be needed if things get a bit dicey near the end. I hate to take money from people for vapourware, and especially if they’re helping me out, but I may need to lean on friends and supporters a bit, once I can show them that it’s worth it. There are various ways I could do that, so thanks for showing it’s a possibility. Long way to go yet, though.

  3. Steve Cas says:

    Sounds fantastic!

    I’ve no doubt you have the skills, determination and drive to develop, market and distribute this game yourself. My only concern is if you’re doing this as a one man band you need to look at the VERY long term (especially if you’re going down the subscription route). Julian Gollop (the creator of the original Laser Squad) formed a company called Codo Games and wrote a great strategy game called Laser Squad Nemesis. The has been going for a number of years but (inevitably?) the number of subscribers has slowly dwindled and Julian hasn’t provided any updates for a couple of years… which has led to even more people quitting.

    Julian hasn’t been seen on the (once very active) forum, support emails aren’t getting replies and the game is suffering a slow and painful death – which is a truly sorry thing to see.

    As a one man show, you need to think about the costs of hosting the game – which will push you towards the subscription route (and it sounds like you certainly could do this – with new features being added on a regular basis), but think long and hard about:
    When you plan to stop adding features (and whether or not you’re going to do something about the subscription price to reflect this),
    When you plan to halt support,
    And when you plan to pull the plug

    Darn! I was hoping to show enthusiasm but rereading this post, I seem to have come over a little negative (sorry about that). I’m sure you’ve already thought about these things and hopefully none of them will happen 🙂

    On a slightly different subject, have you seen this?

    • stevegrand says:

      Hi Steve,

      Yes, I can see how that happens. I spent some time looking at game engines, thinking I should take some shortcuts for once in my life, but for so many of them it was clear that the developers were spending all their time fielding support requests and had no time left for fixing the bugs or writing any documentation. I gave up.

      But there are awful dangers at the other end of the scale too. I’ve been there. I was tech director of a well-funded company of 80 people and it was a nightmare. I’m not going down that road again – I’d rather become a tramp. It’s a slippery slope – as SOON as you start involving other people or investors, things ramp up very quickly and before you know it you’re burning a million dollars a month, the game is ten times over schedule, the world is screaming at you to release something and at the end of it you have to sell a lot of units to see any profit. I was in the games business for ten years and saw it happen a lot. So I’m aiming to develop this by stealth – I’m only telling the people who read this blog what I’m up to because they keep asking me when I’m going to write another game…

      Hosting shouldn’t be a worry – this is definitely not going to be a MMOG. That would be way too ambitious. I’m hoping to have limited networking – explore with a friend – but Alife is so fundamentally different from a character-based game that there are “interesting” technical problems involved. So I’ll do what I can but essentially I’m thinking of it as a standalone simulation to begin with. I’ll just put as many multiuser hooks in as I can. The collaboration will primarily be at the “community” level, though.

      As for Blue Brain: Hmm… I think it’s an important project but either there’s a lot of hype or a lot of stupidity involved. They claim to have simulated a single cortical column, but for one thing that presupposes columns are largely independent entities, which I doubt, despite the limited interwiring. And secondly it’s a bit like inferring the behaviour of a waterfall by making a model of a single water molecule. Plus they only seem to be modelling cortex, which is like building a loudspeaker and thinking you’ve made a radio. Cortex just provides a layer of operations that supervene over the already complex (and in lower animals, quite self-sufficient) architecture of the thalamic nuclei, brainstem, etc. A brain made from cortex alone is not a brain and can’t work. But despite the hype and unrealistic expectations (after all, it’s really just IBM trying to sell supercomputers) it’ll still be a useful project. Thanks for the link!

  4. Terren says:

    Hey Steve,

    Great idea! Sign me up for beta tester too! :-]

    As someone who’s been programming for the last 25 years, I do have to say that what you’re describing sounds nearly impossible in that timeframe as a one-man show. Why not hire some (very lucky) kid out of college to help with the coding? Someone junior enough that you won’t have any qualms with sticking with your vision. There are always tons of menial tasks that need to be done – why not delegate to someone whose time is less valuable (you know what I mean) than yours? Even better if they know who you are and would work for next to nothing for the experience.

    I love the idea for the game but I do have one concern. That is, if so much of the game is collaborative, with disparate gamers contributing to some public wiki, then you’re effectively limiting your gaming audience to a small window, temporally speaking. It won’t make too much sense for someone to buy the game a year after it’s released, in other words.

    That said, if that’s the model, then you need to drum up some huge hype for it before it’s released, so you get as many people playing right from the start.


    • stevegrand says:

      Hey Terren,

      I think I’m going to have more beta testers than customers! 😉

      Yes, that “one-off” character worries me too, and I’m not sure whether to embrace it or avoid it. On the one hand I have to base a lot of my expectations on Creatures, and although it’s kind of “out of print” now, due to its corporate history, it still has an active community today – 13 years on! That’s a hell of a shelf-life for a game. So although people may work out all the bioloy and stuff over the space of a few months, they’ll still have these creatures to interact with as individuals (and with all that information about them to draw on they’ll be fully rounded, complex individuals), so it might still have legs. On the other hand I could approach it like… what was that storybook years ago where a real treasure hunt was involved, and the clues were in the book? Can’t remember, but although it was a one-off event it sold a hell of a lot of copies.

      I’ll keep this all in my mind as it develops – I like to let my creations (books, software or robots) make their own decisions and show me what they want to become. Come to think of it, mybe that’s why I’m not rich… But hell, Art for Art’s Sake, eh? 🙂

      I’m really reluctant to take on coders – it sounds so plausible but it really doesn’t work like that in practice. I’ve tried. Maybe nearer the end, when it’s clear to me what is ACTUALLY a menial task and what just looks menial but in truth has implications for the design. I may need help with website development, UI stuff, graphics, etc. But you can’t take an evolutionary approach to software when you have more than one person – you need specs and interfaces. But I’ll keep that in mind too.

      I’ve been programming 30 years myself and I realize this is absurd. But it’s not quite so absurd that nothing will ever happen. Sometimes you’ve just got to start something and then serendipity will find a way. I’ve had a series of failed projects over the past few years so I’m probably not wise to say that. But they all failed for the reasons I’m trying to avoid – lack of funding, lack of communication or commitment from team members, disparate visions and motivations, 99% specification and 1% code…

      Thanks for the feedback – I’ll keep it all in mind as things develop.

      • James Brooks says:

        I wouldn’t be too discouraged about the one-man show thing, toady over at http://www.bay12games.com/dwarves/dev_now.html has written the most detailed and complex game I have ever seen (by a few orders of magnitude) and done it all on his own. He is paid fully by paypal donations and updates and add to his game constantly.

        Oh and I love the idea and would be happy to donate, test, play or be involved in some way.

    • stevegrand says:

      Thanks James. I’m not discouraged in principle – I’ve ALWAYS worked alone. I wrote nine tenths of Creatures by myself and other people just came in near the end to convert my UI code (which I’d written in VB because I basically only needed a script language for that) into C++. So I’m not scared of writing big projects by myself. It’s whether I can do it in the time available that worries me. And, increasingly, whether I want to – I’ve spent so much of my life working long hours in front of a computer, never having a social life or being able to think about other things. I can see that happening again.

      I’m keeping an eye on D, but there’s no point using that yet. C++ would be the obvious choice for 3D code but it’s a shitty language in other respects and would radically increase the number of bugs in my Alife code. I’ve been programming in C# ever since it was invented, and it’s well-supported for 3D via XNA, so that’s the obvious choice. But thanks.

  5. Steve,

    I think it’s a wonderful idea… it produces that same sense of mystery and adventure that C1 did. I always loved the C1 intro… it was very captivating.

    I like the cooperative creation of an encyclopedia as well. It seems a good way to build a mass following.

    What fun it could be if group explorations could be carried out online. I have a feeling this project is going to send your imagination into overdrive. 🙂


    • stevegrand says:

      Thanks, Richard, very encouraging!

      It would be great to make it massively multiuser but I don’t think I dare. Maybe version 2. But I’d like to make it multi-user on the smaller scale – P2P. It’s quite hard to synchronize something like this between servers because of all the sensory data, but P2P in the same environment should work. Although it’ll be heavily CPU-bound, because unlike a FPS game, all the characters are processing massive amounts of data, even when you can’t see them.

      Anyway, I’m glad you like it – you’re my target audience! 🙂

  6. Norm says:

    Saving the massively multiuser business for later might not be such a bad idea. After the excitement of the first round, you’ll have some money to work with and can then focus on expanding (or deepening) your original premise.

    Your synopsis gave me a glimmer of the excitement I would feel if humans really *did* discover and study life on a distant planet. Serious players of your game might actually take this to heart. Knowing that you are the game’s creator, the life forms available for study might possibly be as close to being “real” as players are likely to encounter in any other game. What will motivate users to continue will be the depth of the experience; the ever-increasing realization of how complex and life-like these creatures are; how alien compared to humans, and how similar.

    Perhaps, as time goes by and funding permits, you can enhance the “depth” of these life forms further. Like getting to know any complex living thing, this process can take years.

    • stevegrand says:

      Hey Norm! Nice to see you here.

      That excitement is exactly what I’m hoping for. There’s no question that I want these creatures (or at least the more complex ones) to be the closest thing to real artificial life ever – WAY more than has ever existed in a game. And by being self-consistent but alien, I’m hoping they’ll take real skill and diligence to understand. Real biology is such a demanding subject – so much inference from circumstantial evidence – and I’m hoping to emulate that. All the way through the development of Creatures the marketing types kept telling me to hide the science, because people aren’t interested, but tens of thousands of expert Creatures users proved them wrong in a staggering way, so this time I want to give those people something they can really get their teeth into. F**k the marketing people.

      As it happens, I’m living in the perfect place, because there are a lot of planetary geologists in Flagstaff and maybe a few exobiologistis too, so if I get stuck for ideas or need validation I know where to ask.

  7. Daniel Mewes says:

    Hi Steve,

    I think the idea for the game sounds very promising. I hardly can wait to see it 🙂
    I really hope that it will also be a commercial success. Think it may be important that the marketing and placement is done right.
    Perhaps you may also consider selling it to schools for example? They could use it for educational purposes in biology classes for example.

    Regarding the long lifespan of the Creatures community, I think that this at least partly is because Creatures is vastly user-expandable. Don’t know if the architecture of the game will allow for this, but perhaps you could also consider selling additional development tools for “power users”, just as CyberLife Technology / Creature Labs did with the Genetics Kit? It may be important to kind of “trick” potential customers into thinking that these tools are powerful scientific utilities (they may of course also actually be such). I can well recall how I saw the Creatures 2 Genetics Kit the first time in some marketing/demo video and saw how it said “Thinking…” while loading a genome and did take quite some time for this. This suggested that something complex was going on there, although the loading of a genome file probably could have been implemented much faster and may have been a more or less trivial task. But it certainly worked as a psychological trick. What I want to stress with this is the fact that the exact way of presenting the game and possible tools may play an important role for sales, or at least I think so.

    Perhaps you may also consider the use of platform independent libraries like GTK (or GTKMM for C++), OpenAL or OpenGL in development if this does not lead to an increase in your development time/effort? I would loosely offer to do a (perhaps unsupported) Linux port for free in this case.

    Hope this does not sound if I would claim demands for you game. Please just do as you think it’s right. Just wanted to tell my personal opinion about it…

    Kind regards,

    • stevegrand says:

      Hey Daniel,

      Yes, I’m sure you’re right about the extensibility. That was what I liked about the pet store idea – we could all keep adding stuff to that forever. But then I decided I liked the idea of a first-person immersive viewpoint better. But you can rest assured that I’ll make it extensible and I know all the tricks with add-on scientific equipment, etc – don’t forget, I was the one who WROTE Creatures, and I specifically designed it that way! Up to a point, anyway – it was much harder in those early days of Windows – I had to use DDE and then OLE1. But now I program in C#/.NET and it’s SO much easier to design extensible software in this. In the Simbiosis game that I decided not to finish, you could extend almost everything.

      GTK wouldn’t be much help (especially in C#) but that’s fine because I already have a UI widget toolkit I wrote for Simbiosis that just needs porting to XNA. I did consider using MOGRE for the 3D but because I have unusual requirements I decided it was more trouble than it’s worth (and MOGRE development seems to have ground to a halt). So I’m using XNA, and a Linux port would therefore be difficult (there just aren’t enough Linux gamers around for me to care too much. And even a simple FPS 3D game (where the baddies just appear as needed, run a couple of IF/THEN statements and then fall over dead) needs all the clock cycles it can get; for a complex A-life game that would be too big a compromise). An XBox version would be easy enough, though, if I wanted it (although this is definitely a PC-style game).

      I’m not too much worried about marketing. I know a fair number of journalists and they often come to me asking for interviews – they’ll be pleased when I actually have something to show them, for once! It would be wrong to hype it or market it forcibly (look at the Spore marketing disaster) – I’ll take a much more viral approach. And it’s not like I need to sell half a million units, because I get 100% profit. I’d rather have a small, dedicated user base than try to aim at the mass market.

      I would definitely like to see it used in schools. However, I spent ten years in the educational software business and I know that the education market is a strange place. It’ll be better if it happens because some passionate teachers sieze on it and promote it themselves, rather than me try to market it as a curriculum resource.

      Thanks for the advice!

      • Daniel Mewes says:

        Hmm ok so it’s going to be C# and XNA…
        Looks as there is an Open Source implementation called Mono.Xna (http://code.google.com/p/monoxna/) but it seems to be very incomplete and the Status page says “The project has been dead for a while […]”. Doesn’t sound very promising.

        So I should better forget about the native Linux port. Maybe Wine will do the thing then. We’ll see…


      • stevegrand says:

        I hadn’t come across a Mono version of XNA. It’s shut down now, but who knows, in a year or two? So maybe a Linux port will be possible after all. Thanks!

      • James Brooks says:

        You could consider using the D programming language, it’s very similar to C# and the developer (mainly one very smart guy Walter Bright) has added lots of tools to help programmers. It works on Windows, Linux and Mac OSX.

        The tool chain does need work and obviously you would have to learn another programming language but you might find it interesting.

  8. Nicholas Lee says:

    Dear Steve,
    I like your game concept. It will attract a more thoughtful category of player than the games that currently saturate the market, which are all about killing things. (Unless you allow your exobiologists to perform disections or alien autopsies?!)

    In terms of the game-to-player dynamic, many gamers play games becuase they provide rewards in terms of points ,’leveling up’ or the gathering of a complete set of a type of object. Such games can be highly addictive, sell in huge numbers and often have a monthly subscription.
    For example ‘World of Warcraft’ is the number one selling game and its players spend huge amounts of time trying to level-up to level 80, collect a complete set of spells or kill particularly rare monsters. Sometimes collaborative play is essential in order to kill particularly tough monsters.

    Your game could potentially incorporate non-destructive versions of these addictive (and successful) elements.

    For example, you could have ranks of exobiologist (apprentice to grand-master) which you can advance through based on the number of species you have cataloged, or there could be medals for saving a species from extinction or for identifying a new species that nobody has catalogued before.
    Completing quests (radioed in from earth) to discover plants with particular vital medicinal properties might make the plant hunting more exciting.

    PS: as you are liley to need to need huge computational power to play this game, might I suggest you look at NVIDIA’s “CUDA” system. This uses the graphics processor on the graphics card as a massivley parallel co-processor for the PC.
    We have successfully used this technique at my work to speed up image processing software which otherwise required hand-optimised MMX machine code to enable it to run in real-time!

    Nicholas Lee

    • stevegrand says:

      Thanks Nick.

      > Your game could potentially incorporate non-destructive versions of these addictive (and successful) elements.

      Yeah, I understand the principle but that’s not really what I want to do. It destroys belief. When you play WoW or any other traditional game you know that you’re playing a game. You’re already into the psychological domain of cartoons and board games. But I don’t write games – never have, except under duress. I write simulations, so this is more like Microsoft Flight Simulator. Now that has challenges in it, I admit, but they’re completely optional and carefully crafted not to disturb the player’s suspension of disbelief (and even then it’s really just a feeble attempt to lure the mass market – no serious FS fans are particularly interested – they want to create their own rules and scenarios. There are a bunch of psychological strategies you have to employ when writing an immersive sim. One is to be very careful about what happens either side of the “glass”. 2D UI widgets like menus and stuff should only ever be used when setting things up, because they remind the user that this is just a computer. If you want to keep them immersed you must make any controls that don’t have anything to do with file management on the other side of the glass – in 3D and in context. A second big no-no is Yerhafters, and “Yerhafter rise up through the ranks” is one of these. It works under some circumstances but not others. When you write sims, your job as a developer is not to wreck the user’s willing suspension of disbelief – you owe it to them. So although I understand how power-ups, ranking and all these other things work for a game, this isn’t a game. I shouldn’t even call it that, except there’s no other category.

      CUDA sounds interesting. It’s too early to tell, but I suspect the game (sorry, simulation) will be CPU-bound (most 3D games are GPU-bound), so I might well be able to offload some of the neural net computations (as well as physics) onto the GPU. Thanks for that.

  9. Really happy to hear that you’re working on something new and cool. I heard about Simbiosis somewhere and was a little disappointed that it didn’t eventuate because, as a lifelong follower of your work, I really enjoy the simulation-centric design philosophy that you often follow. It’s nice to see a focus on science in games — the thoughtful manipulation of complex systems is often as entertaining to me as any strategy game. I’ll definitely be picking this up on release (and I certainly wouldn’t be averse to a little beta action :D).

    At any rate, with regards to the extensibility of the critters in the world, do you think the player will looking mostly to manipulate the morphology and sensor/effector loops of species, or will there also be scope for real changes to the physiology and biochemistry? I know there was some capability for this in creatures, but it all it amounted to was norns that could process detritus as food. It’d be cool to see different niches in the ecosystem filled by extremophiles with radically different homeostatic processes or whatever.

    Anyway, Reaaaaally looking forward to this! All the best!

    • stevegrand says:

      Thanks Anton!
      Yes, Simbiosis was heartbreaking for me. It died in the end, partly because it was associated in my mind with a traumatic time in my life, and I just lost enthusiasm. But also the fact that Spore was being developed at the same time made it hard for me to keep my spirits up. Spore and Simbiosis are completely different, but I could see journalists failing to understand this, so I was always working in the shadow of a project that was being marketed by millions of dollars.

      The initial premise of the new game is just the same as it would be if you could REALLY visit a distant planet and study aliens. You’re not there to make them; you’re there to study them. But they will be built genetically, from complex networks of cell signalling, metabolic and neural processes, so in principle you can manipulate a huge amount. Creatures was very limited by comparison (PCs are hundreds of times faster now!). So you could build whole new lifeforms from scratch, just as I will build the initial population of species. But that’s not a fundamental part of the scenario, and how to fit that “do it yourself biology” aspect into it without spoiling the believability of the initial scenario worries me a little. Still, that’s why I asked you all for your opinions!

      Thanks for the encouragement.

  10. Nicholas Lee says:

    Dear Steve,
    I see where you are coming from with the immersive simulation vs. game distinction, and I think it will be a better product with greater market differentiation as a result.

    I am not clear on how multiple users can collaborate within the simulated world. If each simulation is stand-alone then its virtual ecosystem will evolve independently and will rapidly diverge from the ones being run on other user’s PCs. In this case, the creatures being catalogued by one player would be different from the ones that have evolved on another user’s PC and the users can’t compare notes or collaborate to develop a consistent ‘Wikipedia’ bestiary.

    If the ecosystem simulation was running on a central server-farm on the internet and each user’s PC just did the graphical user interface for that shared simulation data then synchronisation would be possible and users could meet in the simulation world and (more importantly) their creature information will be consistent. It also means that time in the simulation would continue even when the user was not logged-in, which could add to the sense of realism of the simulation.

    Another possibility is that rather than having a huge central server, the task of processing the ecosystem simulation could be distributed amongst the users. This scales better than a single server solution in which your server would get bigger and more expensive in proportion to the number of users.

    The grid computing model used by SETI is the open source software from BOINC (http://boinc.berkeley.edu/) which spreads the processing task across a huge number of home user’s PCs.
    If every person who is running the game is also be running a background task that receives packets from your BOINC server, processes that chunk of the simulation problem and returns it then the more users you have, the more processing power your simulation will have available.

    The advantage of the BOINC model is that the ecosystem simulation can be updated at any time by you (and you can push updates to user’s PCs). Also, if you have a few PCs yourself that you leave running to process chunks of the problem then you can guarantee that the simulation will keep going and it can be left to run autonomously even if no actual users are currently logged in.

    An alternative grid computing model would be to use BitTorrent style P2P where a server just maintains and shares a list of user’s IP addresses. Each user’s PC computes a piece of the ecosystem simulation and shares its results with all the other online users, thereby giving all the users access to a consistent fully-computed simulation result. This system will also scale fairly well as the numbers of users increases but it is harder to update the simulation code as none of it is under central control. Also, if all users logged-out then time would stop in the simulation.

    The disadvantage of grid computing in general that the chunks of the calculation that are parcelled out may not come back finished in a timely fashion, come back in the sequence they were sent out or even come back at all. The data chunks needs to be multiply redundant (i.e. sent to several different users) to have a chance that one of them will actually process the data. Also, the necessity to send the chunks across the internet can be very slow if the data size is large compared to the time required to perform the calculations on it. Grid computing is therefore sometimes not as good at real-time tasks as a big central server doing all the calculations.

    Whatever route you take I suspect it is something that you will need to settle upon as a strategy up-front rather than trying to add it as a feature further down the development road.

    Nicholas Lee

    • stevegrand says:

      > If each simulation is stand-alone then its virtual ecosystem will evolve independently and will rapidly diverge from the ones being run on other user’s PCs.

      Who said anything about evolution, though? This is a realtime simulation. Evolution takes thousands of generations. The rate of evolution is controlled by the turnaround time between generations, and that in turn is determined by the complexity of the creatures, and therefore how long they need to interact with their environment and each other in order to grow, develop, find a mate and breed. The kind of creatures I develop are many HUNDREDS of times more complex than anything anyone else uses in artificial life. It’s neither possible nor desirable to execute a simple fitness function on them – this is real natural selection. So the creatures WILL evolve – their entire morphology and physiology will be defined by genes and they have to survive and mate, so natural selection will occur. But it would take thousands of realtime years for anyone to notice. The most that will be seen is variation – unique mutants and crosses.

      I clearly haven’t explained my thinking well enough. Just because I’m an A-life researcher, doesn’t mean I merely apply GAs to trivial gene strings or make little stick creatures that vaguely wobble about, like other people. My real interest is an artificial life approach to AI, so I’m interested in vastly more complex creatures.

      > Another possibility is that rather than having a huge central server, the task of processing the ecosystem simulation could be distributed amongst the users.

      Yeah, I’m not as naive about these things as it seems. In fact I’m the author of a patent (sadly the rights belong to my former employers) that attacked some of these issues for distributed games. But grid computing, as you point out, is not designed for realtime use. It’s more like batch processing on a mainframe. And although there are many successful server-based and P2P games out there, they’re traditional games. You have to remember that a game character in a normal game is just an animated model and a few if/then statements. My creatures are complex beasts with thousands of interacting parts, drawing realtime, complex sensory data – including vision – from their environment. Allowing a creature to (almost literally) see over the edge of a server boundary is not a trivial problem, because of the amount of sensory data being pushed around. Keeping everything in sync whilst retaining 60fps is no picnic! If it was all on a single server that gets easier, but you still have a lot of avatars trying to interact with the same creatures at the same time, and the poor creatures’ senses have to cope with this, as well as the server being able to scale. It can all be done, but it’s a nontrivial problem and I just don’t have the resources for it. And I’m simply not interested in setting up a company, employing programmers and then spending my life trailing around Wall Street in a suit, begging for money while other people write my game – it’s not where my skills lie.

      Ok, that’s good feedback. Thanks. I’m starting to feel some unpleasant deja vu here. I’ll go away and think about it for a bit.

  11. Matt Griffith says:

    I know exactly what you mean, Steve. I am of the philosophy that, to know how to design anything, you must design every part. Anything and everything I design, I must do myself. Not because of ego, but for the pure ability to know every state, and every interaction possible in whatever I am designing.
    As is, I’d like to be able to support you in any way I can, and if its just as a tester, I shall certainly take you up on the offer. In the mean time, I hope these discussions continue as you work on it. 🙂
    We’re all ears!

    • stevegrand says:

      Thanks Matt! I’m glad I’m not the only control freak around here! 😉

      I’m feeling a bit down about it at the moment – like it’s not such a good idea to hide away and work 14 hours a day writing complex code that might not succeed, when I’m supposed to be in Flagstaff getting my head together and exploring my more artistic side. Maybe I should write a new book first and enjoy the scenery. But I’ll keep thinking about it and see what the gods tell me to do.

      • Terren says:

        Hey Steve,

        As much as I’d like to see whatever you’d come up with, hunched over a computer for thousands of hours, I think if you’re one of the lucky souls on this planet with the leeway to explore, then to hell with expectations, other folks’ or your own.

        As much as I love coding, it is isolating. Particularly if you’re the sort of lone wolf coder you describe yourself as. And at the end of the day, isolation is unhealthy. Strictly speaking for myself, the isolation that coding brings has sometimes served as a dysfunctional way of escaping, so the efforts I’ve made to join with my fellow man have brought me tremendous gains in happiness and creativity.

        So I applaud your urge to get out from behind the computer and find alternative ways to express yourself. I’m sure that whatever you undertake in that regard will reap benefits in all sorts of unexpected ways. It sounds like you are off to a great start.


      • stevegrand says:

        Thanks Terren. That’s very helpful. I’m pretty much convinced that this is what I should do now and what you said helps seal my resolve. The chances of getting a game finished before the money runs out aren’t great, so it would make sense to write a new book first and buy some time. And that means I can get the most out of my life here in Arizona. I didn’t want to be in this situation but it does give me a precious opportunity to do things for myself and get the creative juices going in a way that “normal” life makes difficult. Of course I’ve got to come up with a book idea that I feel inspired by, but that’ll happen. I’m thinking about a book on feedback. I just bought myself a new camera, so that’s kind of a de facto decision to use it to illustrate a new science book, I guess! I’m writing a novel anyway (for my own benefit more than the intention to publish it) so it makes sense to focus on literary stuff for a while. Programming can come later.

        Thanks for the encouragement!

  12. Matt Griffith says:

    Yeah, I’d definitely read a book too. 😉
    *runs off!*

    Write books to give us youngin’s motivation and ideas for executing your dream simulations!

    • stevegrand says:

      🙂 What? And give away all my secrets? Never! I’ll carry them with me to my grave! (Which won’t take long if I don’t earn some money…)

      • Matt Griffith says:

        Full credits to the master! Promise! 😉
        (in the form of trusts and grants! Haha)

  13. Alon says:

    Steve, good thinking! You never stop being creative, and we’re lucky to have someone like you on this planet.

    You haven’t finished many of your past projects, but I think when you realize that you’re going broke, you’ll pull through. The end product will be fantastic, since this is all about your survival, and we know how well things turn out when that’s a factor.

    Oh btw, You should let people pay to beta test your game in advance of its release. I’d do it, sounds fun to see the buds of your creativity before they blossom for the general public.

    Anyhow, good luck. Post often!

    • stevegrand says:

      Thanks, that’s sweet of you!

      Yeah, the thought of bankruptcy concentrates the mind wonderfully! But actually it’s already quite concentrated as it is – I have less than 18 months. I’m considering writing another book first, to give myself a bit more leeway. And also I really don’t want to be stuck indoors slogging my guts out coding all day – I came here to explore my artistic side. So I’m still musing on what’s best to do. But thanks for the encouragement!

  14. Ibad says:

    Hey Steve! I communicated with you a few years earlier when I was in UT Austin and you were working on Lucy! I hope that you are well. I could not help but think what would happen if you and Will Wright teamed up on something! haha.

    Anyway.. I do hope that your latest project succeeds. I spoke to you about a brute force way of reverse engineering the brain, just simulating it part by part, before understanding the high level engineering principles and cybernetics of cognition etc…and then learning from the simulation, like that chap who simulated a whole human heart! Well… it turns out that someone out there is crazy enough to be trying this, and they fit nicely into your category of “bottom up” engineering. I was wondering what you thought of the Blue Brain project? Do you think they might succeed?

    I was also curious about what you learned from the Lucy project, and if you think the Blue Brain guys will find the same things? have you had a chance to compare notes with these guys?

    anyway…best of luck Steve!

    • stevegrand says:

      Hey Ibad, I think I remember. Yep, I’m well thanks. Life’s been a bit “interesting” lately but I’m still hanging on in there!

      I’m not sure what to think about Blue Brain. To be honest I will be surprised if they achieve their larger goals – for one thing they’re only modeling cortex, and cortex doesn’t work without the thalamic nuclei and many other parts of the brain. I don’t see how they’re going to interpret their results beyond an improved understanding of memory without a model of the thalamus too, which is as big a mystery as cortex. On the other hand I’m sure something useful and insightful will come out of it. I can’t help thinking it has more to do with selling supercomputers than solving neuroscience puzzles, but since IBM is happy to support it, it’s definitely worth a try. I wish them luck.

      I learned a lot from Lucy but not the big insights I was hoping for. I ran out of money too soon. I was moving on to do some work in holography next, to follow some of this up, but without an income I just can’t afford to do research any more. Book or game first; research later.


  15. Brian Pangburn says:

    Hi Steve. Big fan of your books and actually keep extra copies of Creation around to give to those that would appreciate it.

    Did AI research in the late 1990’s and early 2000’s developing software to “learn” nouns/verbs based on text descriptions of short animations & videos. May have even sent you a link at that time (http://www.aclweb.org/anthology-new/W/W03/W03-0607.pdf).

    Anyway, decided last night to look online to see what you’ve been working on and it appears that you spent a stint about 20 minutes from me in rural Louisiana (I’m in New Roads). Would have certainly offered to buy you a beer had I know you were across the river. Oh well. Hope your latest journeys are successful and will stay tuned for your new software and/or book.

    If you ever make it back this way, the offer for the beer stands…

    • stevegrand says:

      Hey Brian,

      Hah! I wish I’d known that two years ago! I was starved of intellectual company outside of my family. But somehow I don’t think I’m ever going to feel welcome in Saint Francisville again…

      I sometimes used to take the kids over on the ferry – lots of fun for a dollar. Bet you’ll be glad when the bridge is built.

      Thanks for plugging my book! The paper looks interesting – so rare to see anything so grounded. I’ll read it more carefully later.

      – Steve

      • Brian Pangburn says:

        Sorry to hear about St. Francisville.

        Certainly looking forward to the bridge. Think it will do a lot for New Roads.

        As an FYI, all of the code from my research is on SourceForge (http://sourceforge.net/projects/ebla/). Had a database online for a while to facilitate demos, but took it offline due to lack of use. Would like to toy with it again one day, but life tends to get in the way of research.

        Take care.

  16. Jon says:

    Hi Steve,

    Have you considered doing an iPhone / iPod Touch application as a warmup? The turnaround time on those apps can be very quick. Crank out a few prototypes, sell them for a few bucks to see if anything sticks.

    I think a straight port of Creatures (of if don’t have the rights to the name, something similar) could be successful too.

    I know many professional devs who love working on iPhone games during the evenings. The fixed hardware keeps things simple, and they do many little projects so finish each in a couple months so they don’t get bored.

    • stevegrand says:

      Hi Jon,

      No, I haven’t. Not recently anyway. Maybe the cpus are up to it now. The reason I didn’t do it for phones in the past was that they were way too puny. I could have written something that looked vaguely like creatures but it would have been a sham. I’m just not interested in pet games – I’m a biologist and it was the biology that interested me. An iPhone must be a lot faster and more powerful, although I’m still dubious that it would have enough power to do anything that would interest me much. It would be a potential money-spinner though. I couldn’t port Creatures at all – I don’t own any of the rights, not just the name. I could invent a new NN and stuff, although I’m sort of past that level of AI now and would be loathe to revisit it. I’ll have a think about it.

      Thanks for the thought.

  17. Jon says:

    Things really have changed since in the cell phone space — definitely enough power for something interesting now. The hardware is at least equal to machines that Creatures ran on; better in practice since the iPhone has dedicated 3D hardware and PCs of that era often did not. Rome AD92 or Robin Hood would be a cakewalk of course.

    Original iPhone:
    412Mhz, 128MB RAM, dedicated GPU.

    The 3GS, the newest:
    600Mhz, 256MB RAM, much better GPU. Absolutely not puny.

    40 Million install base: http://gizmodo.com/5287426/the-iphone-is-a-pretty-damn-big-platform-to-develop-games-for

    We’ve seen a lot of successful one man projects on the platform — it’s the first platform in awhile where the top sellers have been tiny brand new studios rather than huge gaming companies. The kind of garage-based development that was common in 80s. Big companies are moving in now, but a tiny studio still looks the same to the user as Electronic Arts in the App store. I know more than a few people who worked on things purely for fun in their spare time and ended up quitting their day job.

    In any case, you may want to take another look at that space. I was thinking it might be nice break from your main project once in awhile: low risk, high potential reward, and most importantly fun.

    • stevegrand says:

      Well I’ve shelved the main project anyway – I just don’t have time to finish it before I run out of money, so I was going to write a book first to buy some time. But I’ll definitely look into iPhone development and see what’s involved. Thanks!

  18. Theodore says:

    Well I was in basic training when this was published and I missed it! Was doing my bi-monthly check to see if the creatures community had done anything and got sent to this blog and then saw rumors of a new game….

    I won’t be faint with the praise – I am MORE than happy to pay for something of this caliber. Creatures 1 is still way beyond anything like Black and White –

    ….yea just read the last comment…

    Well I wish you all the best on the book. Any thoughts on a subject matter? I was pondering buying your first book but I simply don’t have the free time to read it freshman year (at a service academy), so next year – which would be nice if the new book was out too.

    No one does AI like you Steve – and to be honest NO ONE does Artificial Life anymore – so keep up the hard work and all the best.

    • stevegrand says:

      Hey Theodore,

      Thanks for the support! Much appreciated! 🙂

      Yeah, I still plan to do the game but I do need some income to keep me going in the meantime, so there’ll be a book just in time for when you finish your freshman year. I’m considering the topic at the moment. Either a book about androids and the human machine, or one about cybernetics and emergence. Not sure yet.

      Thanks again for the kind comments and good luck with school.

  19. Jeremy Heighway says:

    Hi Steve,

    With the help of a well known search engine I’ve just been doing my occassional search of things I haven’t updated or heard about in a while. Thankfully, the bots there know who you are ;).

    One of the issues I’ve been thinking about is just how fast computers might now be able to simulate/handle evolution and emergent intelligence, and which is still more of a bottleneck: designing a system with enough ‘freedom’ per genetic iteration (AI lifetime), or the number of complex iterations which can be run in a certain amount of time.

    It is especially the ‘zooming in’ or ‘slowing down’ which makes this problem clear – this event of mine (writing in your blog) may not be of evolutionary importance, but then again, who knows? However, to take it as a base, it would be nice if I achieved 1.3 million periods of half an hour in my life. How many ‘episodes’ is each generation of life likely to have in an artificial environment and how will outcomes be affected by these periods?
    Following on from this, how many beings can inhabit an artificial world at the same time whilst maintaining a very much speeded up time base of at least, say, 1 million to one?
    It appears to me that the creator of any nicely complicated world would not have to worry about it being short-lived. The headlines would more likely be ‘Original Grand-World finally enters the Ordovician 10 years after creator’s death! – Grand World II (quad-speed) steams into the Silurian!’.
    Maybe that’s a headline for New Scientist in 2050 – I don’t know.

    Still, if you’re writing a book first, then I wish you all the best with that, although I have the niggling feeling that evolution has been waiting for you to put the next episode into practice, not into print!

    • stevegrand says:

      > Thankfully, the bots there know who you are ;).

      Ah yes, my minions! They’re out there waiting for the revolution to start, you know. Then I’ll be emperor.

      It’s a real issue, I agree. The other day I was at the Grand Canyon, where the top is at something around the Triassic, then you go down a bit and hit the Precambrian quite soon. Then there’s a HECK of a lot of rock stretching back another 1.5 billion years, during which not much really happened. Evolution is SLOW! The big problem from a game perspective is that if you can speed life up a lot to make evolution happen, nobody cares about the creatures any more – who can get emotionally bonded to a creature that lives for a millisecond? Equally, to get realistic natural selection the creatures do have to have a lot of episodes in their life story and that makes things slow down. I have arguments about all this with researchers who think artificial evolution can be used to design complex neural networks – they seem to have many orders of magnitude too much wishful thinking! So basically I gave up, even in Creatures. Evolution was possible but it wasn’t the point. And much of what users saw as evolution was really just variation, which of course is not the same thing. Selective transmission of genotypes across the community sped things up a lot, so it could be said that real evolution happened, but not much. So for this game, I want genetics and variation, but I’d not be focusing on evolution even though it could eventually happen – in a few tens of thousands of years…

      Interesting subject. Thanks for the observations.

  20. Cathal says:

    Hi Steve,
    I’ve been eagerly awaiting your next software sim/game for years, so I’ll be glad to fork out for this if it ever hits the ‘net!

    As far as continuing revenue goes, I have a suggestion: if life is to be shared online a-la Docking Station or a “Life Store”, for people to inject wholly alien developments from other games into their own homebrew ecosystems, perhaps a €2 fee could be charged to the *publisher* for uploading, and download of the critters might cost €1, with €0.10 going to the uploader and the remainder going to store upkeep and as profit to you.

    That way, the meagre upload fee will keep people from spamming terrible critters just for the sake of it, while the tiny revenue stream they might earn for their efforts would still keep serious gamers interested. In other words, I’ll only upload a critter I think might interest at least 20 people, and if I’m confident of my design I might even earn a cup of coffee for it! Meanwhile, the designer gets a continued stream of income to support development and daily life.

    What do you think?

    Oh and Off-Topic: As a roboticist, I’m sure you’re familiar with this phenomenon already, but I thought I’d make sure: Reprap.org and makerbot.com are two websites supporting open-sourced 3d printers that can make parts for more 3D printers. It’s digital Intelligently Guided Evolution in action, and the lead person behind Reprap likes to think of Reprap as a symbiotic organism; requiring human aid to reproduce, while offering a fantastic payoff for the investment of time.

    Take a look. It’s incredible technology for roboticists. Check out the model-sharing site Thingiverse while you’re at it. And if you want one? I’ll print you the parts for a Mini-Mendel for free. 😉

    • stevegrand says:


      Thanks for the ideas! I’d gone off the thought of writing a new sim because I realized I’d run out of money before I finished writing it, but your comment is good timing, because I’m revisiting the idea again now. I’m heading back towards an earlier concept in which transactions like that would make sense, especially using virtual currency, so thanks – and I hadn’t thought about the other benefits, like reducing spamming.

      Thanks for the reprap link too – that’s Adrian Bowyer, who I know from from Bath University. I haven’t seen him for a few years, so I had no idea his plans had gone so far already! I would love one, although it sounds like it’s yet another of those things that might suck me in and distract me horribly from making a living – I’m a real sucker like that…

      • Cathal says:

        Far from distracting from your career as a Roboticist/AI developer, I think having access to a 3-D printer would help accelerate your work.

        How much time goes into building the frame of a robot, that could be better spent developing its brain? If you detect a bug in the Robot’s structure, is it too late to replace it because the tools and materials are filed away?

        Having a rapid prototyper (A makerbot) has enabled me to build a lab centrifuge for an hour’s time and a euro’s worth of plastic. I would really love to see what you could make with one of these things. 😉

        There’s a prototype, unsupported version of Reprap called the Mini-Mendel, which if it takes off could be very quick to make and share. Count yourself on my list of people to send parts to; building the bot after that takes a little investment, but the payoff is worth it.

        More on this when I get enough replacement plastic to commence work. 🙂

        Keep up the awesome work! Really hoping to see Simbiosis someday soon!

      • Cathal says:

        Oh yea, forgot to add: If your robots were made of printed parts, you’d have the satisfaction of seeing people around the world replicating them within hours of your first working prototype. Finally, Robots can share the same sort of self-accelerating development that made software explode.

        The next life Sim you design might be physical, with playpieces and critters made on home prototypers. 😉

      • stevegrand says:

        No, you’re right – it would be useful. I know this because I already own one and use it, but mine’s a subtractive micro-milling machine and hence it can’t produce things as quickly as a 3D printer. It’s been a great asset for making PCBs (or was until short-run PCB suppliers showed up on the web) and it’s been useful for complex plastic and small aluminium parts, but for the majority of parts in my Grace robot I had to make them with a conventional(manual) milling machine. A 3D printer would be great for little robots.

        That’s a really good thought about spreading my robot designs around. Plenty of people have asked for plans in the past but it’s been pointless, because so much was made by eye or using scrap, but a RPM would make copies possible.

        If you’re serious about including me on the viral distribution list then I’d really appreciate it, thanks! But I just KNOW it’s going to suck up my time, because I’ll want to build a zillion things with it!

  21. Trevor says:

    Hi, Steve. are you in contact with Philip Rosedale, the founder of Second Life? Can’t make out what he is currently up to, but take a look at this: http://www.lovemachineinc.com/

    Second Life is kind of interesting, but Mr Rosedale seems to be moving on to new pastures. If he is serious about AI, he really ought to be talking to you!

    • stevegrand says:

      Hi Trevor, No, surprisingly I’ve never crossed paths with Philip, but that looks like a really neat application – so simple and yet potentially so effective. I doubt my kind of AI really fits the bill, because I’m interested in life, not knowledge engineering, but I’m glad you brought it to my attention. It’s a fascinating concept. I may even blog about it. Thanks for the heads-up!

    • stevegrand says:

      P.S. Oh, I probably missed your point. I thought you were talking about the tools he’ll need for analyzing all this “love” that’s flitting around his application, but I guess you’re referring to the mention of brains on the front page of his site. No idea what he has in mind there. I’ll try to get in touch with him and ask.

  22. Trevor says:

    Dear Steve,

    Yes, it was the AI angle that I thought would interest you.

    Also, Philip has a ready-made virtual workd in the shape of Second Life, and all the code used to develop it.

    It looks like you guys should definitely make contact with each other!

  23. Andrei says:

    I suspect it is very hard to develop a game with tight funding. Please consider following the model that these independent developers have done: http://www.wolfire.com.
    Their new title is Overgrowth. They pre-sell their games in exchange for giving buyers an opportunity to: get access to weekly alphas; and a special forum.

    They have also just finished an interesting marketing campaign called the “humble bundle” – see it on youtube. It generated $1.3 million for them and other developers, but most importantly – allowed them to make more pre-sales.

    Your games are great.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: