There have been several threads over the years arguing these points. My considered opinion is that story design belongs here in Phase 2 of the game design process because:
1) In any entertainment, whether it is a game, a movie, stand-up comedy, or what, story is the backbone of that entertainment, organizing all the other elements. The art illustrates the story, the music creates the mood called for by the story, the gameplay tasks gain significance within the context of the story, and the GUI and programming allow the player to interact with the story.
2) Artists and musicians need some instruction for what you want them to create - giving them the story as inspiration is a good starting point.
3) Everyone likes stories. Not particular stories, but stories in general, you could even call storytelling one of the fundamental behaviors that makes us humans rather than animals. As such, an appealing story is an excellent means by which to persuade your team members to commit to and contribute to your project.
So hopefully you can see why it's essential that every game design project have a good writer or two involved at the earliest stages of the process. And IMO game stories can be meaningful, philosophical, and literary if you write them that way, just like any other form of entertainment. They can also be funny, endearing, heart-rending, horrifying, educational, and anything else you have the skill to make them; there's almost nothing you can do in a novel that you can't do in an RPG.
That, however, is an important 'almost'. There are a few types of stories that games as a medium are not suited to, just as there are a few types of stories that don't work as movie scripts. The unique challenge of game story writing is making room for the player. In a novel you decide every thought, feeling, action, and bit of dialogue your main character has. In a game you can't do this, because your main character is being played by a real live human being with his/her own thoughts and feelings. And you know very little about the player - gender, sexuality, age, moral philosophy, IQ, interests... you don't know any of that. Plus, each copy of the game is gong to be played by someone different.
However, this actually makes things easier instead of harder. We can start making statistical assumptions about our audience, such as there will be some players of each gender, but probably more male than female, some of every sexuality but more attracted to women than men, etc. Continuing in this vein, there are some things we can assume about our audience:
1) They are interested enough in the concept of your game that they start playing it. Maybe they even bought it, instead of borrowing it from a friend. But the point is, whatever story you decide to write, the people playing your game have decided it looks cool, so at least you don't have to worry about trying to win over a hostile audience.
2) The player will get annoyed if you put opinions that they disagree with in the mouth of a character they are supposed to be controlling and identifying with. Ditto with making the player's character take an action the player thinks is stupid or morally reprehensible or just plain yucky, like being in a romance with a romantic object character the player doesn't like. The way to work around this is to have opinions stated by NPCs and then let the player choose how to react to them, let the player choose whether to carry out any action that isn't completely essential to the plot, in general envision yourself, the writer, as the god of the game world and NPCs, but not the player; present things to the player, and let him/her choose how to react. Like your game is a big psychology experiment, and the player is the rat, you amuse yourself by poking them and seeing what you can get them to do. ;)
3) The result of having many players and letting them make choices is that you have to anticipate every possible choice and how (if at all) the story should change in response to these choices. This can result in plot branches and multiple endings, which may a bit difficult to write for a writer who is used to creating stories that have only one ending, which is shown to be the 'right' ending by the fact that a problem which harried the main character throughout the story has finally been resolved, usually accompanied by some narratorial moral pronouncement about the right way to be a person and go about living one's life. But it's really tough to write a game story to show the main character learning a lesson of some sort, because you don't know whether the player 'needs' to learn this lesson, or has already learned it and is now bored by it, or thinks your moral is wrong, or thinks your issue is irrelevant to his/her life.
As such, stories of personal growth don't work so well for a game; external problems (like evil villains and ticking bombs) work better, particularly since these lend themselves more to giving the player missions and gameplay tasks, and doing these is, after all, why the player is playing a game rather than reading a book. But personal growth stories are generally considered deeper or more literary, so you may really want to write one. There are two ways to do it: either you can devote the whole plot tree to testing how the player actually feels about this issue, then challenging their beliefs, or you can displace the issue onto a major NPC and have the player watch as this other character 'learns their lesson'.
So, alright, we still haven't started designing our story. Well, let's revisit the story portion of my outline of game design areas:
- - - Archetype/Personality/Motivation/Backstory
- - - Character Dynamic/Roles
- - - Dialogue
- - Beginning
- - - Initial Incident
- - Rising Action
- - - Resolve to Action
- - - Failed Attempt(s)
- - - Complication
- - - Further Attempt(s)
- - Climax
- - - Crisis
- - - Resolution
- - Ending
- - - Denouement (Closure)
- - Physics
- - Geography
- - Ecology
- - Culture
You can really start anywhere: by creating a character, a scene, a culture, a location, whatever you have some inspiration for after answering (or looking at the game designer's answers to) the questions in my previous journal entry. If you are a writer working with a designer's answers, you should first go back over the questions and add your own details and ideas to them, and make sure the designer is happy with your additions/changes.
VERY IMPORTANT: Keep records! Write everything down! Preferably on the computer rather than paper, so you will later be able to copy and paste things into the design doc or email them to other team members to get their opinions/approval. If you discuss story ideas in chat, save your chat scripts. Trust me, I know from first hand experience that you will really regret it if you don't do this, good ideas can be accidently lost forever, time can be wasted in retyping, you can get confused about what you discussed with who, and many other headaches if you don't make the effort to keep your work organized.
So in conclusion, I recommend that you start designing your game story by looking over that features list and the answers to yesterday's questions, then taking the outline above, pasting it into a new document in your word-processing programming program, and then going through the outline and jotting down all the inspirations you have for each area.
That's enough for today. Next I guess I'll give advice on how to design each of the different story elements, as usual I would like a comment or two letting me know what you all think of what I've written so far (pwetty pwease?), and also if you have a particular story element you want me to cover first, feel free to request that. :)