Centralize design decisions

[olivier-grech]

I have a hard time finding information on the design of the game.

Beside this wiki page, I feel like the documentation on what the game actually is supposed to be (beside the very broad goal of being a Pokémon clone) is either non-existing or very well hidden.

In my opinion, the first step should be to have a small design document that can be easily found from the git repository. This design document should establish the following things:

• Commonalities and differences with mainline Pokémon games (already explained in the wiki page, tough I think it could be more detailed).
• Used technology and targeted platforms (C++ with SFML, etc.).
• Technical choices such as screen size, etc. This was already discussed, but I don't feel like the outcome of the discussion was recorded.
• Several mockups of the interface of the game, as Pokémon games are mostly UI-based.
• Broad outline of the progression of the player in the gamer (i.e. start in town A, go to town B, etc.).

In order to make our progress more tangible, I am also in favor of a very detailed description of, say, the first town of the game (layout, events, etc.). This way we could focus on implementing the necessary features to make this first town come alive. This would help motivate the people who contribute to the project, as this is a very concrete and (relatively) short-term goal.

[ghost]

I absolutely agree. The design document could be hosted on the wiki in a prominent place and progress could be tracked with projects.

At the moment, the code needs a lot of cleanup as well. For one, it won't compile with MSVC in the state it's in. The existing CMake file implies Linux, Mac, and Windows as the targeted platforms. That could probably be expanded to all the platforms that SFML supports (like Android). Checking that the code compiles at all could be done with actions.

I suggest coding standards be established and enforced. If compiling on other (non-Linux) platforms is a goal, writing portable code is a necessity and in general, cleaner code will make the project easier to develop and more likely to attract contributions.

[milkyroute]

For now, my top priority is to cleanup the code (which we've been doing for some months already). However, I'll try to detail this at the same time. To be honest, we haven't any precise plans for towns, NPCs and stuff like this, we only have a skeleton for the history, region and gameplay. I think we should discuss this on Discord since there is a lot of decisions to take, and some others that we have already taken but should be criticized.

I'll at least try to do a list of what we already have decided (without further detail) as soon as possible.