For example, in Dogz and Catz (1995-1998), particularly the earlier versions, the characters “knew” they were computer pets, looked out at the screen at the “real you”, the person sitting at the computer — versus an “avatar you”, some virtual representation / role-playing version of you. By reducing the distance between you and the characters in this way, I believe the sense of presence achieved is stronger than when characters are presented as part of an elaborate fictional world of their own (like 99% of games do). Ever since working on those projects, I now find it awkward to role-play in a computer game, particularly with semi-autonomous avatars, which I feel are conceptually messy and confusing. They make little sense to me — who is this character that I partially control and the computer partially controls? — and my guess is this conceptual confusion is one of the barriers for some of the non-gamers masses getting into games. (First-person experiences, such as Myst or Doom have little or none of this problem.) Design-wise, we want fusion, not confusion.

Note that two extremely successful virtual characters, Tamagotchi and Furby, have very high degrees of this kind of fusion. This is no accident.

Before Petz I worked on an Philips CD-I title (talk about a long lost format!) called Max Magic (1994), my first industry project. Max Magic had even more explicit fusion than Petz. You, the player, bring your family and friends into the living room to gather around the television, and then stand next to the television itself, side-by-side with your performance partner, the on-screen character Max Magic, a mechanical magician who kind of looks like “Zoltan” from the movie Big. You use a mixture of rehearsed performance and real props (e.g., a deck regular playing cards) to perform up to 14 magic tricks for the audience, replete with music and witty banter. Max calls you by your real name (a feature we’re also implementing in Facade). You send the occasional cue and signal to Max via a remote control joystick, the standard interface for the CD-I player. The experience includes a rehearsal mode, in which with no one in the room but you and Max (a magician never reveals their secrets!) you rehearse the tricks together. There was no printed instruction manual or computer-esque menus; all rehearsals and performance were done “naturally”.

By the way, both Max Magic and Petz were originally conceived by the creative director of the company, PF.Magic, I worked for — Rob Fulop. This “fusion” design principle I initially learned from Rob, and got further developed and reinforced for me working with fellow designer Adam Frank and others at PF.Magic.

Regarding experiences like Online Caroline, or Majestic, I like the parts of the experience that do achieve this fusion, e.g., the fact that you use your real email, fax, cell phone, what have you, to participate in the experience. But this fusion becomes greatly weakened if it’s not consistent throughout. I tried both Online Caroline and Majestic a while back, and I had trouble even willingly suspending my disbelief when (if memory serves) I had to fill out various menu-y web forms to perform some of the experience. When some of the experience has fusion and some does not, it doesn’t work well for me. Frankly I’m kind of unforgiving when it comes to consistency — a reason I often have trouble with IF. I’m not saying I necessarily have a design solution in mind to fix these fusion inconsistencies, but they’re a problem for me nonetheless. To me it’s a sign that much of these forms are still maturing (obviously).

Wanted to mention, though, that I think there is a difference between certain types of fusion. You aren’t literally feeding your Tamogatchi in the real world, by getting food and giving it to a creature – you’re pressing buttons on a little device. In the fictional world, you’re feeding a creature. But you are really attending to the Tamogatchi round the clock in both worlds.

So I think it’s more nuanced than some parts of an experience having fusion and some not. And maybe it works well at times to be surprised by ontological fusion (or even a lack thereof!) – in the right context.

The “real name” feature is an interesting one, used in Racter, Seastalker, and loads of other computer arts and amusements. It’s interesting because it forces a choice on the interactor right away: are you going to be Andrew by typing in your real name, or are you going to be Cornelius or Rupert or something and decide from the start that you won’t “fuse”?

I found the discussion of these pieces to be somewhat out of place in the context of games. Neither the AI campaign nor the LARP can be re-played (the Vampire game drew upon current events and was highly improvisational), which to me makes them more performance than game. But they do blur the line between acting in the real world and acting for a fiction.

Nick, if you like Hacker and have a few hours to spare you might want to look at Uplink.

As the lead writer on Online Caroline this thread piqued my interest. At the time of developing that project I was VERY interested in this idea of fusion and indeed tried to take it further (in a more satiric, comic vein) in the next project we did called Mount Kristos. But the more we tried to blur the lines between the real world and the digital fiction, or at least ‘play’ with this idea, the more confused and disturbed I became by the audience’s reaction. Displace people in terms of their involvement in (and responsibility for) the drama and a lot of them quickly don’t really know where to draw the line, or at least it *appears* to me that they don’t – but then they could be pulling the author’s chain, right?

On Andrew’s point that filling in data forms and the like rather breaks the illusion of fusion (to coin a phrase), I’d just like to say that we did think long and hard about what might seem to be a ‘natural’ way for Caroline to gather data about people. In then end, we had to consider what it was realistic to expect Caroline to have built herself.

One of the issues about creating seamless ‘natural’ interfaces for interacting with digital personalities is that you also have to consider what kind of interface ‘fits’ the personality and the world in which that personality lives. It’s amazing how many characters in digital fictions turn out to be amazing interface designers or whizzy HTML programmers in order to create the right kind of ‘fusion’. Kinda limits the range of stories us writers can tell, tho ;-)

Thanks for getting me back to thinking about this. It’s been a while…

Your point about your distrubed reaction is interesting… I think I’ll continue the thread in a new top level post…