There are two options to organize and structure play in “8”. The first option is remove most traditional notions of gameplay as a challenge-based or goal-oriented activity and offer the player a completely open game world and a lot of freedom as to how to play. The second option is to design a very tight gameplay structure that guides players through the gameplay more or less transparently.
Any decision regarding which option to implement in the game will be made together with the main publisher of “8”.

Open play world
In this structure, access to all areas in the palace would be available from the start. The main game goal of returning the eight stolen objects to acquire spell powers would still be included. But finding them would not require solving any puzzles. It would be a simple matter of exploration. As a result, the player will probably acquire all spells relatively early in the game, which opens the potential of playful interaction even more. There would still be puzzles. But solving one would be its own reward (in the form of a spectacle or a narrative revelation). And there would be many objects that serve no purpose other than improvised play.
This kind of game is one that many players have expressed as being desirable, if not ideal. It is a form of play that many people long for, the designers of “8” included. As such, it holds commercial potential (at least towards the more or less hardcore game-playing part of the audience). The fact, however, that this type of "game" has never been made before succesfully, makes it hard to judge just how strong this potential is (what gamers wish for and what they actually like and buy may be two different things). In the end, any decision to make this type of game will be motivated mostly by artistic ambitions.
Note that since this is an untested form of game design, the production will be longer (and more expensive) due to more extensive user testing and more iterations of corrections and redesigns. And also because the game world will probably need to be richer and more detailed.

Structured game
For this approach, we have worked together closely with the International Hobo team of game designers. They have prepared a consultation document with their recommendations. There's a link to this document at the end of this page. Here follows an introduction and summary.

The goal of the structure proposed by iHobo is to achieve the original design goals of “8”. The proposal is based on the opinion that total freedom in a game world makes it very difficult for people to orient and motivate themselves. And on the analysis of “8” as having most commercial potential in a target market that consists of casual non-gamers and the opinion that the only way to reach this market is through a smaller group of hardcore aficionados of simulation and adventure games. As such, the proposal contains a combination of playful freeform "toyplay" with partly optional layer of goal-oriented "gameplay".

The game world would be divided up into 3 major zones, one of 60% of the complete gameworld, the others of 10%, with the remaining 20% serving as connections between these zones. These zones would each be structured in a socalled "logical loop", guiding the navigation of the player while still giving him or her the feeling of total freedom of exploration. In the beginning of the game, only the first zone of 60% would be available. The player needs to solve puzzles to gain access to other regions.
For finding the eight stolen objects, the player needs to solve similar puzzles. These puzzles are all to be solved through the use of a small number of tools, available to the player from the start of the game. As such, it would be completely clear to the player which objects can be used to solve the puzzles. All other objects that are present in the game world, are there only for non-competitive "toyplay".
There will be more or less hidden ways for the (hardcore) player to measure his or her performance in terms of puzzle solving and spell casting. But only 70% of the (15 or so) puzzles need to be completed to finish the game.
In the end, the complete palace will be available to the player, allowing for the same type of play as in the open structure model desicribed above. The advantage of the gameplay layer would be that, by the time full access is possible, the player knows his or her way around in the palace and knows very well what to do and how to do it.

The full consultation document can be downloaded here (PDF).