GDC 2010 Notes

Beyond Scrum: Agile Project Management for Games

Speaker/s: Clinton Keith

Clinton Keith is pretty much the agile game development guru, and he definitely makes a good case for the process.  The talk focused on applying project management practices where needed, using common sense with the transparency that Scrum provides, and filling out the Agile framework.

He described the main challenges in game development as managing fixed ship dates, maintaining minimum feature sets and avoiding crunch.  Scrum, as a tool for transparency and in providing velocity, helps show where you’re coming up short in time for you to do something about the problem.  Clinton talked about some scoping tools to help deal with the challenges outlined above aside from deadline shifting or velocity forcing (crunch).

1.      Organize features using MuSCoW Analysis: Must have -->Should have-->Could have-->Won’t have.  The Must-haves are standard features while the Should-haves and Could-haves are novel features.

2.      Backlog prioritization: reveals the value, cost, and risk of features and provides knowledge.

3.      Critical path: Identify dependencies that must occur in sequence.

4.      Critical Chain Staffing: Identify stuff doen in the past, how many people/how long did that feature take?

My favorite part of this talk was its focus on common sense.  Clinton encouraged attendees to find what fits their team’s needs and use the proper tools instead of following labels.  Scrum is not the goal (magical remedy), it is a tool: focus on the principles of empiricism, emergence, time boxing, prioritization and self organization.  Use this tool to iterate on your process.  You need to provide leadership and guidance for Scrum to be effective, and plan continually to attack uncertainties.

What I want next from Clinton Keith is a session of case studies!

Game Writer's Roundtable

Moderator: Richard Dansky

Richard Dansky is also the guy collecting submissions for the GDC Austin game narrative review competition.  I attended the talk just to see what it’d be like, I guess.  I felt that the most vocal participants were the most opinionated (no surprise, but caused content to feel unbalanced) and learned that just about everybody in the room felt that writing is the most slighted occupation in the games industry.

Everything was pretty straightforward, from methods of filling in backstory to the player’s role in creating story.  The opinions that spoke the most to me were ones focusing on writers recognizing themselves as part of a development team.  One guy asked all the writers if their stories energized their teams, causing them all to play a role in realizing it (sharing the vision).  Another said that his first job was to listen and see what he could draw out of the existing design.  Resonant comment: writers give gameplay an emotional wrapper.

Wake Up! Your Team is on Fire! Why Your Real Problems are Cultural and how to Change Them

Speaker/s: Michael Saladino

Saladino was my manager at Pandemic when I interned there as a producer last summer.

His talk promoted the identification of (or reduction of problems into) feedback loops and how to understand them to reduce delays in a system.  Feedback loops can be positive (encouragement to grow), negative (stasis), and exist in your team: examine loops on the team through the lens of delays.  He espoused the values of having a shared vision and promoting team interconnectedness through transparency, visual workspaces, allowing hard questions in public, and avoiding “closed doors,” and more, which all fit nicely into his recommendation of Scrum and Agile practices (monthly retrospectives providing actionable feedback, team self-organization).  He examined the concept of crunch  as a feedback loop, encouraging us to address the core issue in order to effect 2^nd level change and stop the damaging cycle.

Related to, but partly outside of the systems talk, he encouraged some guidance/leadership principles as well.  The first was leveraging the size of your company to find who you want to talk to when you’ve noticed their work (and they’ll love talking about their work), as well as forming monthly discussion calls with like-minded individuals across the company.  He encouraged attendees to be teachers: present new ideas, create workshops to bring knowledge into your team, challenge people for growth via 1:1’s, and acknowledge that the best teachers are also students themselves.  Lastly he talked about the importance of building a culture that fosters these things:

1.      First, lead yourself.  Act on your ideas/promises.

2.      Second, lead your supervisors.

3.      Third, lead your peers – by working together, we win.

By living this, you earn soft authority, and the culture begins to perpetuate.  To effect change, run experiments, find the right approach.  Evolution gives people time to convert information into knowledge into understanding and ultimately into wisdom.

He ended his talk by giving us homework: Think of your worst issues that always seem to be around year after year.  These are likely systemic!  Identify the driver and effect change.

Concrete Practices to Be a Better Leader

Speaker/s:  Brian Sharp

I really enjoyed this talk because I felt a lot of other talks were very technical, or talk about game development in terms of different systems.  Brian Sharp talked about human beings and it dovetailed nicely into Saladino’s discussion about culture. Sharp started by saying that leadership is neither science or craft, it is a practice.  Practice: Training is the opposite of hoping; intellectual comprehension is not enough for mastery.  Leadership itself is synthesizing a vision with the team and guiding them to it.

1.      Empathy.

a.       Remember you are dealing with a shell, and there’s a human underneath.

b.      Listening and understanding earns trust and respect.

c.       “This person is just like me; they want to be happy.”

d.      “This person is just like me; they do not want to suffer.”

2.      Humility.

a.       A lot of people prefer to be right rather than compassionate.

b.      Arrogance impedes us and our work.

c.       Need for validation/affirmation is an impediment.

d.      Humility will make you happier and allow you to learn.

e.       “May I be of service?”

f.       Everyone is my teacher.

3.      Heartfelt Communication.

a.       Communicating with the head consists of facts, logic, and checklists.

b.      Communicating with the heart includes feeling and emotion, recognizing that words affect people, and noticing how others feel about what you’re saying and responding to that.

c.       Create the terms and conditions that will allow your team to succeed.

d.      Focus on the basics (references “Difficult Conversations” by Douglas Stone)

4.      Confidence.

a.       Training is the opposite of hoping.

b.      Face insecurity and work with it.

c.       Training instead of hoping (or longing, idealizing, wishing).

d.      Confidence is a gift to those around you.

e.       Confidence is not about me, it is about those around me and it is my responsibility.

f.       Self examination: Why do I feel anxious or uncomfortable about something?

Agile: No Silver Bullet

Speaker/s:  Chris Oltyan, Rich Vogel, Clinton Keith, Michael Capps

I was pretty underwhelmed with this session, it felt a bit scattered due to its question/answer panel style.  There was some good re-iteration of solid practices though:

1.      Beauty of wall scrums is that everyone can manipulate it; electronic scrums are in the hands of one guy at the keyboard, which means the meeting ends up losing energy.  Tools can still help a lot, one team uses wall stuff for the stand-ups and software for the actual pushing and tracking of tasks.

2.      Scrum is meant to be changed.

3.      Plan for time-consuming polish and iteration; get into the habit of locking down.

Game Design Challenge: Real-World Permadeath

Speaker/s: Eric Zimmerman, Heather Kelley, Erin Robinson, Jenova Chen, Kim Swift

I’ve heard several friends say they probably will never go to another challenge session because they don’t learn anything from it these days.  I’m not so jaded, I thought there was a lot to learn not necessarily from the content but from the questions raised thereof.  This year’s challenge was based on the idea that someone in the game must die or be dead in the real world.

1.      Last Game and Testament, Heather Kelly & Erin Robinson

The premise is that you use the software to create a scavenger-hunt style game based on your will. The game includes stories and videos about your stuff and you get points for guessing who certain items would go to.  Objects that have meaning to everyone are bonus items given to those who know you best (i.e., have the top scores). 

I liked it because (to use their words) it tells the stories that live in objects, which are otherwise unceremoniously meted out to family members – that is, your stuff is just stuff that somebody “deserves.”  I liked the idea of having fun creating this for your family instead of writing a will, and the idea that your family might even have fun playing.  This is in line with my personal philosophies that funerals are supposed to be celebrations of life and the fulfillment of a natural cycle – it can be sad, but it shouldn’t be negative.

2.      HeavenVille, Jenova Chen

I’m guessing once he hit on this idea, the comic potential was too much for him to resist.  Lots of people laughed at the apparent trivialization of death as people are converted into stocks in HeavenVille after dying and the stocks go up and down based on how well known they are or how well you promote your dead friends.  Jenova hoped to present death as a positive; a confrontation and understanding of death that makes you appreciate human life.  In this respect I think HeavenVille only partially works; the confrontation is there, but the appreciation felt kind of meaningless and silly.  However it raised great questions about how our digital culture will absorb death into our daily lives.

3.       Karma, Kim Swift

Kim’s idea was basically a grief counseling game, which you are prescribed after you find out you’re going to die.  You gain karma by helping other based on in-game suggestions.  I liked the general concept, but I don’t think I see how the game might be fun for the person who is dying, though I can see how it might help you cope with the news.  I didn’t see why you’d play this game through to the end, and it wasn’t clear what you do in the game when you aren’t helping other people.  I could see a lot of patients feeling patronized by being prescribed the game, as well.

Where Did My Inventory Go? Refining Gameplay in Mass Effect 2

Speaker/s:  Christina Norman

Christina Norman gave a great talk about what drove design changes in Mass Effect 2.  She went over a lot of details, so I’ll try to extract the headlines.  The team were motivated to make changes through the fact that they didn’t get to do everything, and through fan and review feedback.  Interestingly, they conducted a quantitative analysis of reviews (almost like the Tomatometer you would see on rottentomatoes.com) in an Excel sheet to provide a perspective on Mass Effect 1.

She identified the main design theme of ME2 as creating a feeling of intensity.  They designated a design team for gameplay changes and rallied the team around this theme, which was a touchpoint for all the systems they went on to change (inventory, combat, and other systems that were already changing due to performance upgrades).  One of the interesting things they did, was that in order to focus on shooter combat, they turned off all RPG systems, allowing them to evaluate and iterate the shooting game by itself (of course, RPG stuff was developed in parallel, separately). 

Other things changed in the name of intensity: it’s okay to pause combat as an option, but not a mandate (in ME1 you’d pause to select powers, in ME2 they are much more accessible); Class designs each have features that make them distinct; global cooldown on powers and inclusion of ammo clips forced combat resource management.