An Interview with Annie Vandermeer Mitsoda – Part II

Design Documentation

What do you think is important having in a design documentation?

That’s tricky. I’ve dealt with multiple different methods of documentation, and while a couple years ago I would say I favored big aggregated Word documents, I’ve found consistently they’re tricky to update piecemeal, and as they grow in size, they’re more and more daunting to have to deal with. On Alpha Protocol, we used an amalgamated story document that contained story background, themes and goals, influences, both a general and a detailed overview of the action, a character list, and secondary themes and ideas… and the frickin’ thing got to be over 80 pages!

For Storm of Zehir, major documents that were key to the core of the game were kept far smaller, and weren’t updated as frequently: I combined research and jotted-down thoughts with first-draft pre-production concepts to create a master story doc that was my sole responsibility to update, but was shared with the team. Beyond that, there was a master document for each of the major systems introduced or revised, and for each area of the game, the designer responsible for it would write up a Module Document covering the basics of the area – from story to gameplay to art – and those would grow from a basic overview (5-8 pages) to a large master document (20-30 pages).

Nowadays, I work with a wiki for both Guild Wars 2 and Dead State, and have found it to be the absolute best thing for documentation: it tracks revision history, links easily to other relevant topics, and is really easy to update. The world of Guild Wars in particular has a vast amount of history to it, and having each major topic have a dedicated page that links to minor ones is amazingly helpful for becoming swiftly familiar with previously unknown topics. And at home, working on the Dead State wiki makes it really easy for me to catch up on things Brian has worked on during the day, and keep myself up to date on any changes.

I hope the tl;dr version of “put it on a wiki” doesn’t seem like a cop-out, but seriously, if you’re working without a source control solution (i.e. a program that saves various iterations of files and allows you to revert changes to previous versions), do not risk possible deletion or corruption of major documents if you can help it. If you have to deal with that, or you really don’t like using a wiki, make sure you’ve got lots of dated backups (and include a section at the end of the document dating and detailing each change that was made to it) in case of a disaster. If you do have a wiki, just start with the big topics and branch out from there. If you can think of more than a paragraph or 2 subsections dealing with a subtopic, summarize it on the main page and create a new page dedicated to the topic, then link it back up to the main one. Don’t go too crazy with the individual pages, but honestly, when it comes to a wiki page, it’s often best to go many shorter pages than several really long ones.

For Storm of Zehir, major documents that were key to the core of the game were kept far smaller, and weren’t updated as frequently: I combined research and jotted-down thoughts with first-draft pre-production concepts to create a master story doc that was my sole responsibility to update, but was shared with the team.

How are influences planned? I’m talking about Mask of the Betrayer’s and NWN2 influence system. I don’t know if there is any of that in Storm of Zehir, or even if Alpha Protocol has something like that (perhaps the way you influence factions is comparable).

This is tricky for me, as I didn’t get involved in the influence system for the NWN2 games. For AP, it waseasier, as the dialogues were more condensed, and usually just followed small iterations (either Boolean yes/no settings, or a small sequence of checks). So if a character liked being treated Professionally, a counter would note how many times you reacted to them that way, and have the NPC react more favorably to them. If major decisions happened within a dialogue, that key “either-or” decision was a variable that was tracked throughout the game.

For the purposes of organizing influence checks, I’d say the first step is to set up a simple timeline (or tree, as the case may be) representing the action of the game, and put down major moments for an influence change as key points on that line. What levels of influence should characters have a reaction to – simple like/dislike, or a more nuanced kind of scale? If it’s more nuanced, then numerically, what are a character’s “breaking points” – when they either attack/leave because they hate the player, or they confess their eternal friendship/love? Where is the earliest point on the game’s timeline for the player to hit each of these breaking points with each influenced character? Furthermore, are there points in the game’s timeline that force decisions on the player that could really drastically alter a character’s perception of the player character? Make-or-break moments like these up the tension in the game, but if they happened everywhere, they’d just end up being frustrating. Taking a very careful long view of the flow of the game helps to ensure that these moments are seeded throughout so they feel satisfying and natural, not random and confusing.

How was the innovative multicharacter Storm of Zehir dialog system designed?

The Party Conversation system was one of the ideas that took shape very early in development – pre-production, in fact, and prior to my involvement in the project – and one that pretty much cemented the intent of the game to be a title that didn’t focus on continuing the story of the Knight-Captain from the Original Campaign and instead branched off in an entirely new direction. I recall very early on in NWN2’s development that the dev team wanted to make the game feel more like an actual D&D campaign, something that focused on a party instead of just a single character. NWN2’s original campaign emphasized the mechanics of needing a party in combat, but didn’t shape a sort of situation where a player felt like they “owned” the non-player characters in the same way as they did the one they created. Mask of the Betrayer veered further from this original path by making the plot a very personal story to the player character, and emphasizing the other characters as side players in this journey. Storm of Zehir was always intended to be a departure from the other two games, and that was a key feature of it.

The design of the Party Conversation system was supposed to allow the player to develop the personalities within a full adventuring party in single-player, and allow people in a multiplayer format greater freedom to participate in joint conversations. One of the factors that we knew this would affect early on was that we had to cut out the cutscene-style conversations of NWN2 and MotB, while another was that it required a great deal more writing to properly implement, so that players saw enough variation in each conversation to not miss the aforementioned cutscene-style format. In practice, this led to me having to meet a kind of “diversity quota” with the amount of responses I had to write to each NPC line – how many lines were needed that would be reactive to stats, skills, classes, etc. The “by the numbers” approach made me a bit nervous at times that we were dictating personality by elements of a character sheet rather than letting it evolve in a more organic way (through alignment-changing decisions rather than alignment-reactive ones), and I did my best to make sure that there were appealing responses for people who didn’t just want to make either a trait-based choice or a totally vanilla option.

That’s it for the second part, check back tomorrow for the last part, on Inspiration & Craftsmanship!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s