Mount & Blade did this. They allowed players to buy the game while it was still in development. Early versions started off quite cheap, and they raised the price as they did releases (culminating with version 1.0 at the final retail price). Some info here:

http://www.taleworlds.com/mb_buy_en.html

Cortex Command is another game doing the same. The message on their website reads:

"Please note that Cortex Command is currently a work in progress and NOT a finished product! The campaign mode and missions are not yet present in the version available right now. However, you may buy a discounted license today, which will unlock the features of all future versions up to and including the final with the full campaign in it!

As we at Data Realms continue to release more complete builds of the game, we will also be reducing the discount gradually until we reach the final version and full price of the product. So the earlier you buy, the better value you get! Although we are very passionate about this project, we cannot guarantee that it will be completed - that's why we offer the discounted price instead of a pre-order deal at full price."

http://www.datarealms.com/games.php

Cortex Command is basically a one-man operation. Dan currently sells the game at $18, with identical bits at $38 labeled as a "support data realms" version. He is very clear that the versions are the same; only the price is different. A nontrivial amount of people buy the $38 version! Full disclosure: Dan is a good local friend of mine.

I've met Dan a few days ago in Sweden, and apparently he is making a decent living off of Cortex Command.

What happens if the project falls through/runs out of money? It would be cool if the developers signed some sort of contract requiring them to open source everything in such a case.

Game developers many times use third-party IP (for example game engines). In such a case, open sourcing is worthless.

That's definitely true in a lot of cases, but many of the more indie-friendly engines (like Source, I believe) are free or cheap for non-commercial use. I think that works well with the fact that this funding model is most likely to work with indie-scale projects.

This could be like a more extreme version of pre-orders: fully-paid pre-orders paid directly to the developers before development begins, with the potential for there to eventually be no product. Who wants to bet on Duke Nukem Forever? Maybe an agreed-upon fraction instead of the full amount. I'd pay $20 now for a chance for DNF to come into existence later so I can pay the other $30 NPV of it or however much (if I thought I would be a PC gamer by then).

I'm not sure on the legalities etc. but the idea of investing in a particular project with returns sounds great - it could help align the profit motive with making a great game.

An interesting extension would be a similar approach to additional features/content for existing products, although I'm not sure how you judge the returns of this stuff beyond combining it in expansion/content packs.

Would be a good model for some television series, too—if budgets could be reasonably estimated and controlled.

Nice idea, but in practice this could never work at scale.

A hundred, a thousand, a million stakeholders in a project? They'd have near-zero say in the direction of the project and essentially no recourse if the developer went off the rails.

It works for companies, doesn't it? I think you're putting too much stock (no pun intended) in the "democratic" aspect of the idea. I could see it working much better as a way of expressing support and vaguely influencing direction. The stakeholders don't need to make every decision for the design team. That's why there is a design team.

I'll probably be blasted for this, but I don't believe using democratic principles in a company is a good idea. You need someone with a view of what needs to be done and then you distribute out the work.

I downvoted, but for the "I'll probably be blasted for this" - don't presuppose reaction.

If it did work for companies then Newell's complaint would have no merit.

Simply: publicly-traded publishers would already be funding the games that gamers want. If you want a free preview of how fan-funded game development would go, buy a few hundred shares of Acti-lizzard.

The fact that this model has broken down so thoroughly in the entertainment markets -- where products not developed via market survey are the rare exception -- is precisely why Newell is complaining.

What he's -actually- saying is: "wouldn't it be great if I could have free reign; like with self-funding but without assuming 100% of the risk."

It has worked for the United States as well.

well, kinda

It probably works better for niche projects where the participants are well-informed and can agree on most of the decisions, rather than watered-down mass market projects that try to please everyone.

Same reason democracy works better in small countries with highly educated population.

thats why you hedge your bets by buying options

This is the entire business model of indie games in many cases.

The key difference is that people support indie developers by buying a completed product; they don't have economic interest in an ongoing development.

So there aren't any thorny legal and economic questions.

True that is the case with many, but in some cases online games like http://www.minecraft.net/ are solely supported by the fans, users of the game as it is being built.

This has already been done.

http://www.kickstarter.com/

What is Kickstarter?

Kickstarter is a new way to fund ideas and endeavors.

Project creators can offer products, services or other benefits ("rewards") to inspire people to support their project: A hot-air balloon ride to the first person to pledge $300, an invitation to the BBQ for anyone who pledges more than $5, exclusive daily video updates for anyone who pledges more than $1. It's up to each project creator to sculpt their own offers and there's lots of cool ways to do it.

I don't know if it's been used to fund a video game yet, though.