The Avatars Project: Postmortem 

The Guildhall is dedicated to supporting research efforts at Family Research Center, spanning the past 10 years, to raise awareness of dating violence and allow teens to practice assertiveness skills. The Guildhall first created a Half-Life 2 mod (2006) for which a research article was published, Reducing Sexual Victimization Among Adolescent Girls: A Randomized Controlled Pilot Trial of My Voice, My Choice. The Guildhall later implemented the project in Unity 5.1 in 2015 with added VR support.

The Guildhall is now developing The Avatars Project, a further expansion of the first VR simulation. The Avatars Project will simulate the experience of a high school or college student at a party with a counter part to simulate high-stress social situations and train students how to navigate those situations. In addition, the project will add a third scenario where the avatar and the participant will sit in a school setting for the same purposes.

The Guildhall team was comprised of three faculty artists, a contract programmer, two student producers (myself included) and a faculty producer.  The project was completed over the period of 8 weeks in conjunction with the SMU Department of Psychology.

What went well

Takeaway:

This project was part of a class called Directed Focus Study 1 at the Guildhall.  Myself and another production student were given the opportunity to run this small project. This was our first experience being largely responsible for an entire project by ourselves.  We were developing a porch scene and a school scene.  Each one of us was given a scene to produce along with some other responsibilities. This process took place over 8 weeks and we met as a team once a week. What made the project a success was working with several passionate developers. Having artists and a programmer with a ton of experience between them who were excited and engaged with our product was very fruitful.  When problems arose, they jumped at the opportunity to solve them. For example, the drinking animation in the Porch scene was troublesome at the beginning.  A combination of our developers stepped up to create assets and program certain elements to generate a solution that was acceptable to the clients.  The drinking animation is visible in the trailer above.  By trusting in our developers and giving the project our best shot as young producers, we ended up hitting our development milestones and finishing the product on time.

What went wrong

Takeaway:

While the project could be considered a success, as a learning experience on this short project there were many things that could have been improved.  Personally, I felt like meeting once a week with a team was not enough to be able to effectively keep up with the project.  So much changes in a week that it's hard to stay current.  Additionally, almost all the people working on the project were faculty members.  When we had time to check in with them they were often in class.  This was unavoidable due to the nature of our schedules, but it worked against us. Additionally, we had no level designers working on this project. Our programmer did an amazing job of getting everything in the game and making it look believable in VR, but having someone more familiar with the engine and level design would have sped up the development of this product.

What I learned

Takeaway:

The biggest takeaway from this project for me was finding the right tool for the job.  None of our developers were in the same room.  They were close to one another, but not sharing the same space. We also didn't check in on the developers but once a week.  This meant we needed to find a tool for task tracking and progress management that would work well for those specific conditions.  What we decided early on was that we would use Airtable as a task management system.  We embedded it on the team's Wiki, which was where our documentation lived.  As the Wiki was frequented often by the developers we figured it would be a good solution.  We liked this solution over a physical scrum board because we would be unable to look at the physical scrum board after school hours, but we could check the Airtable from home.  Our decision ended up being a functional solution, but it was not the most optimal solution we could have found.  At the time, we made the best decision we could, but in retrospect another tool like Jira might have been a better choice.  This retrospection on the Avatars Project drove home the importance of finding the right tool to best fit your developers' needs.