Topic: An Online Adventure Game Design Document App (The Plan)

[magintz]

Updates:
 - May 13th - Update 1
 - May 19th - Update 2
 - Jun 16th - Update 3

For those of you with a short attention span here's the skinny
I'm building an online game design document tool. I need your ideas, suggestions and criticisms. Reply below.

For those with five minutes to spare
I've got an idea. An idea I've had for a while that I want to run past all you wonderful adventure game developers. I want to build a tool, a web application to be specific, for designing adventure games. I'm so used to just jotting down notes in Google Docs or Word with no structure, and I was thinking, about six months ago, why not formalise my game development process and procedures into an online tool that not only I can use but the AGS community can benefit from.

What I want this app to be doing
I initially want to build a web app that will allow me to create game projects. These game projects, or stories, will have:
 - Characters
  - Characters will have bios, stats, likes, dislikes. Everything you may want to know about them.
 - Locations
  - These will be the rooms and places in the game and how they join on to the other rooms.
 - Inventory items
 - Dialogues
  - I'm planning an interactive tool for creating branch like dialogues probabaly similar to AJA's Dialog Designer
 - Puzzles
  - A formalised checklist of tasks and orders for things to be done to complete an objective
 - To-do list
  - Create tasks, assign them to people and items to get done
 - Polls
  - Vote or get feedback on a particular thought

Benefits
Well, apart from helping me to organise my thoughts and keep track of everything, I'm hoping this will help people and teams come together to design and build better games. It'll be structured, searchable and versioned. It'll also be a great reference point whenever you need a reminder over a character, place or item.

The benefit of it being online means that you can work from anywhere and with any one.

I plan on building story templates that will set out an outline for a story arc into acts, those acts into goals, those goals into puzzles which will involve inventory items, characters and dialogues in various locations.

For those of you who've read this far, an example:

Let's take an example, a really crude, brief example that will require a LOT of imagination for now!

Act 1 - Set sail:

Fade in to docks
Enter Hero from left
HERO: I need to find all the treasures.
HERO: Let's get me a crew!!1

- Goal - Gather a crew
-- Sub-Goal - Find a navigator
--- Task - Convince the navigator to join you
---- Dialogue (recruit navigator)
--- Puzzle - Find a compass
--- Task - Give compass to navigator

Characters:
Hero - Strapping wannabe pirate.
Navigator - The navigator of the Hero's ship. One eye, one leg. Goes by the name skippy.

Inventory Items:
Compass - Found in chest

Dialogues:
Recruit Navigator
 HERO - Hey, join my crew
 NAVIGATOR - Sure, find me a compass

Puzzles:
Find a compass
 Look at chest
 Open chest
 Take compass

... and your minds are blown!

This is a long way from being completed but I'm really enthsiastic about it. I've got some ideas and have been reading a lot on design documents. I've had a look at the Grim Fandango DD as well as a few other methodologies for game design. I had a look at a few apps for movie script writing and am building up a picture but really need you guys to help plan what you want or think would be useful. I can use this thread to update you guys and collect more info.

For lack of a better name for the tool I'm going with the lazy codename of 'dossier'.

My background in case people are skeptical about my skillz
Well. I'm a professional web-developer. My diet consists of Javascript and a healthy sized portion of HTML, CSS and other web-related curios. My professional career consists of building global brand management systems and online digital libraries. I'm also one of the least organised people I know and need lists for almost everything.

So, my plan is to build this with help from you. Tell me what you want, what you think would be good or bad. What workflow processes you go through to build your game. How you divide up your work to get things done. 

-magintz

[Ali]

Good idea. I think as much as possible, the description of puzzles should be in narrative order, the way the player experiences it. But that does make it a little tricky for us to follow as game-makers. Based on the Grim Fandango Design Doc, here's a screengrab of a flowchart I'm using at the moment:



Your app could make it much easier for me by allowing me to expand and collapse the sub-puzzles which make up the main puzzles. And if I could then click on each puzzle to go to the text-based description of it, that'd be a great benefit.

[Peder Johnsen+]

I like the sound of this! .

[Baron]

Oh man, this would be awesome for organizing a team, especially if it could be dynamically interacted with.  The actual design might need permissions so it couldn't be edited by just anyone, but to have different team members be able to leave comments or check off tasks would be hugely useful. 

Just glossing over your proposals so far, I don't see a lot of thought put into organizing animations.  It'd be great to plan them out in advance to get a sense of the scale of a project. 

Also internal links would be awesome:  So you have a list of inventory items with descriptions, but it'd be great to be able to click on it and see where the inventory item is used in puzzle design, for example (or vice-versa), link puzzle aspects to To-Do list, etc.

[Eggie]

Something that worked a bit like a wiki would be awesome. I'd like to make a list of, say, inventory items and then have my notes about a puzzle automatically put a link to that item's page in when I type it's name.
Then a list at the bottom of that puzzles page would appear saying ITEMS USED: A CRUSTY BUM and the item's page would get a USED IN QUEST: A CRUSTY BUM FOR RUSTY MUM and yeeeeah. It would be a good way of keeping track of stuff

[Eric]

I've been using Scrivener for my design doc and find some of it's functions suited for that purpose. If you've never used it, you might want to poke around with it and see if there are some ideas you might borrow for your app.

[Stupot+]

ooh, Scrivener looks really cool.  I feel more creative and motivated for just having clicked on the link.

I think something like this aimed at designing adventures would be really handy.  I would use it.  I wonder if you could make it so that you can actually get the app to generate bits of code based on your input that can be copy and pasted into AGS*.  This could work well with dialogue trees, at least.

^{*or one of several other scripts/langauges of the user's choice.}

[magintz]

I guess the idea, in the long distant future, would be to have this export to xml which could then be imported to AGS to make your game for you. Or at least bootstrap it to a usable level. But that's a long way off. One step at a time I guess.

[tzachs]

This has great potential!

A few thoughts:

1. When I design a game, I usually work with a deadline and a limited amount of resources. I would like to be able to take all the tasks in the to-do list, assign both people and time estimation for completion. Then I'd like to be able to put an estimation of how many hours a person would put in each day (I usually do this in excel), and then have an estimation on how far behind I am, so that I could make changes as needed,. For example, I could assign myself to work 2 hours on each week day and 8 hours on the weekend, and then if I have tasks that take more than 18 hours (2*5 + 8) then I know that I won't make it in a week and I need to cut down on something (or ask for help).
It would be awesome to have it in the software.

2. There's a methodology for writing stories (I don't remember its name), where you start your story with one sentence that captures the main idea of the story. Then you expand that sentence to a paragraph (a sort of 'elevator pitch') and add more details, then you expand each sentence of that paragraph to its own paragraph, and so on until you have your book. If somebody knows what I'm talking about and have a link with more explanations, this method is a great way to create well structured stories (and obviously it's great for designing adventure games as well), so it would be great to be able to apply this workflow in the program.

3. A tool to analyze the game 'balance' would also be great. By 'balance' I mean: Have the ability to set difficulty and types to puzzles (where the types could be tags with no restriction). Then you could check that the game starts up with easier puzzles and continues to harder puzzles and not the other way around (maybe show a difficulty graph). The tool will also check that there are not too many puzzles of the same type, and that if you have some puzzles with the same type, that they are far apart.

[Stupot+]

2. There's a methodology for writing stories (I don't remember its name), where you start your story with one sentence that captures the main idea of the story. Then you expand that sentence to a paragraph (a sort of 'elevator pitch') and add more details, then you expand each sentence of that paragraph to its own paragraph, and so on until you have your book. If somebody knows what I'm talking about and have a link with more explanations, this method is a great way to create well structured stories (and obviously it's great for designing adventure games as well), so it would be great to be able to apply this workflow in the program.

I think this might be called the 'Snowball' method.

[tzachs]

Thanks!
You reminded me, it's actually called the snowflake method and here's the article.

[magintz]

Update

So, I've been busy thus-far over the weekend. I've written about nine pages (yes, a whole 9). I've done some card-sorting exercises and read through a few design documents - although my main guide will be the Grim Fandango one - purely because of a bias towards Tim Schaefer and LucasArts Games.

Take a look at this proposed structure:

Click to embiggenate

I've also divided the proposed work load into seven key stages. I plan to have each stage taking roughly two weeks worth of work (not necessarily two actual weeks, probably more like a month of actual time). There'll be testing and tweaks all throughout and the ability to change what's happening quite quickly. I'll probably need a few people to start using it and messing around with the system at the end of each stage starting from the end of stage two. With stage three and four I'm hoping to be more widely open to people who want early access to a beta.

Stage One: The Core - end of June
Underlying framework and engine
Database architecture

Stage Two: Simple stuffs creation - end of July
Registration
Create Project
Create Characters
Create Locations
Create Inventory Items
Create Objects

Stage Three: Story - end of August
Story creation
Acts
Puzzles

Stage Four: Teams - end of September
Adding team members
Member roles and permissions
Task lists
Task creation and assignment

Stage Five: Release v0.1 - end of October
Get feedback from users about the UI, functionality, user journeys etc...
Changes, fixes

Stage Six: Files - end of November
Graphics and audio file uploads (I've left this till last as it doesn't really provide much core functionality. It's more of an aesthetic thing. It also uses up a lot of space and bandwidth that I don't want to commit to early in the project and runs a lot of risks such as malicious files being uploaded to the system).

Stage Seven: Release v0.2 - end of December

The dates laid out above are just rough guidelines. I'm sure some will come sooner and some will come later, but that's the joy of an agile-like development. I'll take things as and when they come. But I feel that the schedule is neither ambitious nor generous.

Any thoughts about the architecture? Any thoughts about the schedule?

[Ali]

Good start, but I have a few thoughts:

I think locations and props ought to be a subset of acts, as a shift from location to location is often the main factor in a shift from act to act. Or at a pinch, story. I think it's odd to make dialogues a subset of props, surely they belong to the story or characters?

I think the Audio & Visual (would "Assets" be better) ought to be automatically filled with locations, characters and props as they are created in the game area.

I also think it's wrong to have sprites and animations split up like that. Partly because it presupposes that the game uses sprites, which doesn't take into account 3D, but also because I think animations (or actions which need to be animated) should be things which can be added to characters. I'm not actually sure what I'd put into a sprites section...

So you could create universal actions like "walking", and give lots of characters a walking action. But you'd also want to be able to add a unique action to individual characters.

Is there much of a difference between videos and cutscenes? You could simply tag cutscenes as being in-game or video.

[Peder Johnsen+]

I'll happily help test this as you work through the stages!

[magintz]

Let's see if I can chip in my thought process and see what you think I hope I've answered these things clearly.

I think locations and props ought to be a subset of acts, as a shift from location to location is often the main factor in a shift from act to act. Or at a pinch, story. I think it's odd to make dialogues a subset of props, surely they belong to the story or characters?

The idea behind this is that all resources and locations can be defined externally from the story and the purpose of the story and the acts is to bring these unrelated entities together into a linked prose. A good analogy or example might be making a scrap book. You take all your photos and collect all your memorabilia (this would be your characters and locations). You'd then make your scrapbook (or story) by piecing together all your locations and characters.

Don't get me wrong, the characters and stories as well as dialogue are the driving mechanisms behind the story but my thought process was to decouple them slightly. So that someone can work on editing a room or a character which can then be called upon at will.

I'm not sure if you understand what I mean but it's more so that when using the system you don't have to delve in through the story when all you need is to work on a character but at the same time, because the story will use characters and locations, you'll have references in-line to take you to view the characters in the character section etc...

Have my thoughts assuaged you, if not I'm still a bit sceptical about not having a central repository of characters. Remember this may not necessarily reflect the interface i.e. you would still be able to create a character from the puzzle area of an act. But it would still reside in catalogue of characters on a higher level. I'd see it something like Facebook. Add all your friends first (characters). Create a new party event (act, puzzle, story) and choose the friends you want to invite to the party (add to the act).

I think the Audio & Visual (would "Assets" be better) ought to be automatically filled with locations, characters and props as they are created in the game area.

Hehe, this was my other thought. I 'ummed' and 'ahhed' over which way this would work. I didn't know whether or not to keep this separate - thinking fro a team perspective where an artist might not care about anything but uploading pretty pictures and wouldn't want to be bogged down with following links. Similar to above, everything would link together, you'd pull images from your 'asset' library to define a location. But yes, I can agree with you that it makes sense to incorporate into each individual game area.

I also think it's wrong to have sprites and animations split up like that. Partly because it presupposes that the game uses sprites, which doesn't take into account 3D, but also because I think animations (or actions which need to be animated) should be things which can be added to characters. I'm not actually sure what I'd put into a sprites section...

So you could create universal actions like "walking", and give lots of characters a walking action. But you'd also want to be able to add a unique action to individual characters.

The idea of 3D never really crossed my mind to start with. I think perhaps the terminology is wrong. My thought was to have a separate library for all the pretty things to later be incorporated into a game. I think what would be better is to go with your earlier suggestion of having assets added directly to characters etc... however keep a 'team upload area' available should an artist be uploading a background or a frame for something and not know where it should go. They could then create a new to-do list task for the project lead to assign the asset to something.

Is there much of a difference between videos and cutscenes? You could simply tag cutscenes as being in-game or video.

Perhaps this one is more terminology than anything else. The idea of a cutscene was meant to be a written block of story that was non-interactable. The key thing here is written, it would be descriptive such as an intro dialogue or sequence. This could be a video or simply a blocking script of a character moving about the screen in a soliloquy style manour. A video would be a graphical representation of a cutscene similar to a sprite being a graphical representation of a frame of a character. Similarly to what you discussed earlier I'd assume a video would be uploaded to a cutscene as the cutscenes asset.

Thanks for the quick feedback.

Sometime tonight or tomorrow I'll try to upload a reviseddiagram.

[magintz]

So. I've made a few changes.

Click to largify

A project will have a team made up of users; teams haven't changed.
 - A user will have a to-do list and belong to a role group (roles can be permissioned or just be titular)

A story will be divided into acts.
 - The story and acts will have written descriptions. Small and/or large.
 - Acts will defined characters and locations or pull characters and locations forward from a previous act
 - Acts will have goals (not shown, whoops)
   - A goal is made up of puzzles (think Grim Fandango puzzle layout as shown above in Ali's post)

Assets will be created and linked to the following:
 - Inventory Items
 - Objects
 - Dialogue
 - Locations
 - Characters

Assets will be divided into two categories graphical and audible, assets added under each section can be tagged as sprites, animations, sfx, voice etc... these tags will be user defined and will really only be used for searching and helping organise things. Assets in general will be very open and loosely defined.

Let me know updated thoughts, remember things can be changed as we go based on feedback. This is more to get a general 'North Star' to aim for, but things will hopefully evolve down the line.

[magintz]

Just a quick update to let everyone know I have been working on this in my spare time. I've been playing with some designs and been thinking long and hard about code / technologies.

If you want to have a sneak-peak at the design evolution so far head on over to my site - omgaz.co.uk

[Frito Master]

Looking cool.

[magintz]

Another update, right on schedule and - somewhat - to plan. Development is picking up, but I don't have any free weekends for another month. So, any further work in July will be evenings however I've got a few other side projects that will be taking my priority for a week or so.

Stage One: The Core - end of June
Underlying framework and engine
Database architecture

Stage Two: Simple stuffs creation - end of July
Registration
Create Project
Create Characters
Create Locations
Create Inventory Items
Create Objects

Stage Three: Story - end of August
Story creation
Acts
Puzzles

Stage Four: Teams - end of September
Adding team members
Member roles and permissions
Task lists
Task creation and assignment

Stage Five: Release v0.1 - end of October
Get feedback from users about the UI, functionality, user journeys etc...
Changes, fixes

Stage Six: Files - end of November
Graphics and audio file uploads (I've left this till last as it doesn't really provide much core functionality. It's more of an aesthetic thing. It also uses up a lot of space and bandwidth that I don't want to commit to early in the project and runs a lot of risks such as malicious files being uploaded to the system).

Stage Seven: Release v0.2 - end of December

I'm hoping that, with my current and predicted progress, to have some friends look over this towards the end of July. Then hopefully get some public feedback mid-August; I'm not even going to call that an Alpha release, I think prototype is more fitting.

Hopefully the December release will be Alpha/Beta stage.

[magintz]

Two questions people.
Question 1:
Does anyone care about oauth or similar logins i.e. being able to use their google, facebook or twitter account to log-in rather than registering for yet ANOTHER service.

I think it'd be plausible, in the future, to link to your AGS account but I'd have to talk to AGA about that.

Question 2:
How important would a task/to-do list and messaging (internal system chat/message inbox)? I'm tempted to scrap it entirely... at least for now. We can revist any features later.

Answers on the back of a postcard, if nobody replies I assume nobody cares either way and thus it will not be included.

As always, if anyone else has any comments/suggestions give me a shout. I'm still aiming for something to show you guys by end of August.