I have been following up on the hypothesis that it is fun for a group of people to create a story together. It could be an enjoyable event, a game you might play at a dinner party (or a virtual one). I had some friends come over for a socially distanced evening, and we tried it out while hanging out on our front porch.
One of the questions I had for myself is how incoherent a story becomes where the characters are all played by different people. To this effect, I wrote a short dialog between two people having an argument over the kitchen sink. Notably, whether dirty dishes belong inside the sink, or next to the sink. There are two characters: RED and BLUE. But the way we set up the game, any one of four people could play any of the characters. The identity of the speaker was coded by the color of the clothes.
It turns out that there were two “interesting” versions of the night. The one full of out takes as we laugh at ourselves because we are forced into ridiculous context “Why am I wearing this red bathrobe and acting angry on my front porch?” In this version we can look back at the night and remember the experience of doing something silly together. A video of us playing charades would be similar. We laugh at our antics.
However there is another one: where we stitch together all the best takes to produce a 45 second skit. The one where the production actually tried to respect the conventions of fiction: where a fourth wall was placed between the audience and the characters on the screen. Could we make a believable scene where the audience was ignorant of our process, and just was able to tap into the story? Could we actually produce something that other people wanted to see? I think for this first of this first skit, the answer is NO. The video is pretty rough, and the story does not hold together. We can improve the quality by various means, minor editing, and a modest change in our behavior. But the transition from person to person it TOO jarring to create continuity in character, especially for the rate that we were moving the story forward.
When a new person appears (even though they are acting out REDs role, and RED was already introduced), we can’t help but interpret a new person AS A NEW PERSON. So we think, “who is this? And the conversation written as a dialog between two people becomes a collection of 4 people talking. I am still interested in what kind of extreme post production or virtual production would rescue the coherence of the character. (e.g is a vocoder and digital mask enough? This is like the opposite of jibjab, allowing you to act the body and apply the virtual face. Deepfakes might not even be needed, as memoji might be enough.)
There were too many edges to tackle all at once. I think next time it makes sense to keep one character == one actor. And then just film in two locations.
UPDATE 01/21/2021
I decided to map “Red” and “Blue” onto two separate avatars. I like the aesthetic. The incoherent voices are still distracting, but its a bit easier to bind a character with a similar face. It become more interesting, but mostly because its unusual, uncanny.
Why am I doing this?
I believe there will be future types of media that parallelize across participants. Wikipedia curates a very impressive encyclopedia. Hit record is a fascinating community of creators working together. There are some examples of crowd sourced film production, which flips the notion of audience participation, by having backers act as the investors up front. This is hard, especially with the cost, complexity and delay for current methods of production. I am looking for things that are faster, cheaper and more scalable. That is, where more people can add in a few minutes or hours and drive up the quality of a production.
I don’t think that crowds will displace hollywood, or their business model. But something about OTT distribution in the world of Youtube, TikTok, Quibi RIP, will stick. And the key question I am looking at is how can you drive quality in production, despite fragmentation of key aspects: distributed production and distributed editing. We can actually look at these three moments of coherence: writing, directing and editing. It helps to have a strong creative voice to stand for a consistent narrative at each point along the journey. Is this creative expertise locked up in hollywood, or is there a way that talent in each aspect can be tapped to work alongside large groups and cohering AIs? Another big aspect is, lol, acting. I don’t know how to solve that.
