With all the buzz about the wisdom of crowds and mass collaboration someone had to be brave (crazy) enough to try mass creativity. This is a - case study of that project.

Objective: To write a novel.
Organization: The project was supported by Penguin Publishers and De Montfort University.
Personnel: 1500 relatively anonymous writers from around the globe.
Technology:public wiki - Mediawiki.

Lessons Learned:

• too many cooks will spoil the soup
• the result of unstructured mass creativity is chaotic zanyiness.
• in a collaborative project people need to spend more time reading than writing.

A Sample of A Million Penquins

All You Need is some Mysterious and Ineffable Thing

"So a community can write a novel?" Jim looked up with a skeptical expression.
"Yes, but only a humorous one."
"And only if they put all the new stuff at the start, and the old at the rear"
"It is true. Although we all know the gravity of the human condition, I promise you it is humor that is shared by a community. We respect serious authors who write great globs of sad and dramatic prose because the natural inclination of us all is to laugh."
"Oh come off it." Jim objected, but he was starting to wonder.
"This will ONLY work if we move it into satire and humor," George insisted with great emphasis. Oddly, he said this as though someone actually cared. "Can anyone explain why there is a bias against humour as opposed to melodrama or suspense genres? Life is full of all these elements, but it seems that 'tears' trump 'laughter' every time. But don't worry, 'Romeo and Juliet' is more popular than 'As You like It' or 'Summer Night's Dream' so even Shakespeare felt the pointy end of this debate. No wait, I correct myself: The critics always love 'Romeo and Juliet' better than 'Summer Nights Dream' but I reckon the audience might feel differently. In fact, isn't 'Much Ado About Nothing' actually 'Romeo and Juliet' with a happy ending??...." George was almost frothing by this time.
"Yes" Jim screamed angrily, "but Laughter is the panacea for the pain of the human experience, just like apathy is the icing on the tractor and pepperoni is the topping on the pizza when someone orders pepperoni and they get the order right. Dude, laughter is, like, one of the only true opiates that makes us recognize our humanity and take bold pleasure in it, or some junk. That and a few other things, like singing to trees and picking daisies in the sunshine."
"Still," George dully continued because he was too much of an idiot to end this boring argument "some serious things have to be formulated seriously, man. Seriously, they seriously do. I'm serious!!! Otherwise, they will seriously lose their sobriety and not be serious any more. Some things should reach our minds unadulterated. Yeah, I'm a big shot who knows words like that one. I seriously am."
At least, that was George's take on things. Jeez, he was sure fond of his shiny little glass eye that he periodically took out to scare strangers. It wasn't so much the ghastly wasted void in his skull that scared strangers. Rather, it was the eye that they were drawn to. The shiny little glass eye, doubled as a crystal ball, one that displayed the strangers fate!
"Marbles is for looking through!", George exclaimed to his pet hermit crab, who glimpsed his fate of being crushed by a big fat buttery banana cake dropped straight from the oven this coming Christmas day.

What happend?

This relatively well publicized project ran for two months and included 1500 separate editors, contributing over 11,000 edits while 75000 people visited the site and there were more than 300,000 page views. Like most online communities, however the long tail distribution was in effect and it would appear the majority of edits were made by the most frequent 20 or 40 editors, with hundreds apiece.

Projects like Wikipedia, grounded in reality and history, have successfully used collaborative editing to create a knowledge base in a wiki. There is only so much debate about who shot Abraham Lincoln, for instance, so eventually content emerges that most people accept as the "truth" or fact.

Such grounding does not exist in a creative world. No facts, plots, characters or universes were fixed in amillionpenguins. Characters popped into and out of existence like subatomic particles. This was no small frustration to the projects would be editors. Jon Elek, the projects “chief editor” wrotes in his editors blog:

“I see that you, the 6 billion writers of amillionpenguins, have decided to gamble. I, your miserable and long suffering editor, admit to feeling completely at odds with the novel as it stands. In Stalinist Russia they would have considered this a good thing: that the familiar had somehow become very strange indeed. But in this, our stupider age, I find I just can’t get very far. . . . The main problem I have is that every time I go back to the website it’s changed, a bit like my girlfriend’s mind. And perhaps like that, it resists rational enquiry.”

Participating, not collaborating

Early in the experiment, the narrative within the novel's many sections quickly became entirely chaotic.

If only people would use bananas in traps, rather than cheese! They were getting tired of cheese - besides, Larry thought he might be lactose intolerant, and the continual diet of cheese was making him feel very lethargic and run-down. Which wasn't a good state to be in, for a mouse. And which state he wouldn't have been in, had he eaten more spinach than cheese.

Artie, the whale, was also feeling lethargic. He had eaten far too much shrimp! He groaned, holding his full belly with his flippers. He felt like he might throw up - which was always a messy sight, but he managed to hold himself together. Finally though, he succumbed to the unpleasant rumblings in his stomach. It was, thought Artie, a very unpleasant fruit bowl that he eventually threw up in. "I think I'll take pottery lessons and make something better."

The tendency of the narrative strongly tending to "zany" reflects that without substantial communication or coordination between participants, the best strategy for writers was to create humor by exacerbating the discontinuity of the "novel". Creating order, on the other hand would have been a much more difficult and probably futile task.

The community of participants did not necessarily taking the chaos lying down, however, and numerous proposals were floating:

• that once a chapter is written it should be “locked” and this will hopefully encourage other editors to continue to weave threads established in the earlier chapters.
• That the quest for universal narrative should be abandoned and a “parallel universes” approach should be taken whereby from each page in the story a reader can depart to numerous other pages – reflecting alternate narratives and endings.
• The Chief editor seems to have unilaterally imposed “reading breaks” when nothing is allowed to change – which might give editors enough time to develop a sense of the overarching narrative.
• A fully, partially or not at all democratic editorial board which manages all contributions and imposes order on the mob.

As the project wore on, the core group of editors seemed to gain back some ground - probably as the "drive-by" traffic decreased, the chaos receded. Many sections (like the opening quotation above) recovered to achieve some sense of narrative (although a storyline never emerged.)

The question remains to be examined though, how much actual collaboration occurred, or whether authors participated mostly in developing sections they tried to take ownership over, or engaging in some quasi competitive "riffing" to produce artful phrases or paragraphs that were generally unconnected.

This experiment was closed after two months.

Can mass creativity work?

It is in the truly creative enterprises that the intelligence (or lack thereof) of a crowd meets it most formidable opponent in the “one great genius.” In most areas of creativity, the crowd will probably win, but creating a novel takes a lot more effort and cordination than random crowds will be able to manage.

In the long run, the contest may be much like the one which occurred when computers battled humans for supremacy in chess, and won. “Computers” in this contest were not sentient robots, of course, they were mobs of human programmers, who collaboratively, eventually came up with the robust algorithms that could respond to any challenge initiated by their sneaky and creative little human opponent. Computers don’t think on the fly, and neither can mobs – they have processes and sub processes and sub-sub-processes.

In creative endeavors, writers are often loath to examine or formalize their processes, and hope for great bursts of inspiration to see them through. Collaborative writing cannot work this way. Collaborative editing will require editors to see the process the way a computer would see it – as processes. Govering processes, sub-ordinate processes, structure, structure structure.

Suggestions?

The wiki literati might start with writing sitcoms. And might save the world from bad TV in the process. The discipline of the sitcom is perfect for wiki because it eliminates most of the zaniness that overwhelms the Penguin experiment.

1. Develop a short list of archetypal characters.
2. All plots must resolve to “normal” – everything back the way it was before – this implies that “normality conditions” must be explicitly stated.
3.Objectify the plot devices – experienced writers have been saying for years that there are only so many stories. The wiki literati must embrace this and use plot templates.

This article was first posted by mlpilling on his site at highproductivity.ca