This felt very one-sided so I decided to go do my own thing. In case a Google search dropped you on this page, the bulk of the content and currently all updates now populate my gamedev blog here.
---My Blog---
http://www.gamedev.net/blog/1780-theoretical-games-that-evolve-from-player-input/
You're in the right place if you want to see the first public post on the topic.
First impression.
If you're playing a game, and feel like you could have wrote a better dialogue, why not? Because it's hard coded and there is currently no system to accept new additions other than what the developer intentionally implemented.
Asynchronous multiplayer, crowd sourcing
Skip to "Cause and effect" to see the first generation of this idea.
In its very nature the game will require constant input and have no end because of constant review and revision.
How I concluded this would be fun.
First of all, the text based genre doesn't seem to connect well recently. But Wikipedia ranks 6th in the world's global popularity for visited web sites*.
Users are either going there to edit or find out something new. If they didn't need to, or they thought it wasn't fun, they wouldn't, they'd go read a dictionary.
General Logic
Crowd sourcing solves problems. Games require players to solve problems. Video games can be solved by crowd sourcing.
Crowd sourcing makes problems ... so games can be made by crowd sourcing.
We can observe in meme or genetic evolution. Only the strongest survive.
So who's up for a debate?
More to the point, any concept this logic results in will follow:
Cause and effect "pairs"
Similar to regular games, this first generation is text based. It is also comparable word of mouth story telling.
This concept is intended to result in a database of cause and effect that expands so large there is no one place in the database you can access without encountering a new puzzle, riddle, or joke. A single browse may result in deeply complicated games one person cannot properly solve without "cheating", but an open ended solution input system (text for now) can both accept and reveal all past user's input.
It utilizes one of the weakest and most primitive forms of puzzle gaming to its own advantage, direct user text input. It throws away the worst part of intelligent gaming (not to be mistaken for mindless), the grind. Any solution is allowed to be entered, it then reveals past accepted solutions if the player wants to see them, and any other outcomes to that new solution can later be polled.
Possible use case:
A highschool student enters a "new problem" into a database. They press send and two hours later (when some undetermined process is completed) the system has concluded this is not a new problem, it found 281 past instances of similar problems with over 12,000 uniquely input solutions that have been reduced to 3 similar accepted ones. The problem was used in 23 stories and there are 112 other user identified entries that applied the problem directly, remotely, or metaphorically. 2 word games were submitted that utilize this problem. 1 game is registered to utilize this problem. 18 external user submitted links are identified to contain this problem.
When the student returns home 6 hours later they find out how common their new idea really was. It's too interesting to ignore because they thought of it all on their own.
specific logic (refer to general logic above)
Different problem solutions can have different outcomes. It is possible for 100 players to think of 100 unique solutions with its own new outcome, the most common solution and its outcome pairing is usually the most applicable (due to probability). It is possible for 100 players to think of 100 different pairs, the most common pair is the most applicable.
Reverse gaming
from our point of view this is science fiction
the concept
I came up with the second generation pretty much the same way I came up with the first one. If any previous generation is incomplete there's no way for this to function.
Now I'm writing science fiction. We do not have the technology.
When something happens while telling a story, people may become intrigued and make revisions. There is an understood flexibility in the rules of the story, and we'll make our own adaptations. After the first generation is matured there is a already a system that understands these things, but it needs to be taken a step further.
This is not the same as virtual reality, but it may be related. When a user interacts with a game, it is possible to record metrics. Metrics include physical locations players reached, and what options are picked, and possibly any other in-game stats that may be broken down into statistics. It is really hard to think of a good use for all of these metrics when the game is already done.
Similar to drawing an animation. The game creators will provide the key frames, they define the game and link specific events to cause and effect, the end users will simply enjoy the game and then possibly log their "what if" back into the cause and effect. This makes it a better game.
Still drawing from the crowd sourcing concept, specialized user made content can be created that defines the in-between frames (visuals and programming).
Combine the metrics and content creation. When a player simply acts out a new solution, rather than suggesting it, the new metric is recorded. Now here's the beauty of this. The game developer's assistants, a special user, a programmer, an artist, one of the end users, whatever they may be referred to later can learn someone tried something completely new, and ludicrous, and wrong, and then they will create content to make it right.
Currently science fiction at best, a fully integrated system could "tween" the animator's and previous content. Specialized users, or the "animator's assistant" will be more suited. For the same reason virtual reality is just not done, this isn't happening any time soon. But if you refer to generation one in this series, maybe you'll understand how it works a little better now.
In its final stages, it would be very possible to play a game "backwards" like a video tape, because the cause and effect are modular. This solves the problem of why we can't have true time travel in video games. This also solves the complexity of developing open world games. This doesn't quite solve our lack of content required for virtual reality. Well, chances are the third generation will involve all those things.
specific logic
The cause effect database of the past generation in gaming has matured, I can go play just about any text game online and feel like I'm interacting with an incredibly creative game master who knows every rule. I can even suggest things and sometimes the GM can tell me it won't work, and the reason why. Now the specialized users who make content for video games have been using these ideas more and more. We need a better program to quickly bridge the two.
references:
* http://www.alexa.com/siteinfo/wikipedia.org for URL metrics
about the author:
This is how I wake up in the morning.
recent edit:
Wrote second generation game concept.
Added science fiction, content tween, and game reversal tags to reflect the changes. Note: multiplayer tag still seems out of place, it's coming in the next generation.
