Month: February 2018

Attempting to recreate the experience of individuals can be very difficult. For example, Dangerous Embodiments, an attempt at capturing experiences of different individuals accomplished very little. As I wandered through the virtual world, I kept expecting for something to be different as I changed my avatar, yet nothing did. It almost felt as though I was trying to force myself to feel a certain way when choosing different avatars. It felt as though I was trying to force myself to get into the mentality of being my chosen avatar.

As we played around with MakeHuman I recognized how difficult it is to create characters that accurately embody the mission of a game. The level of detail to which I could configure the players infringed upon my ability to accurately represent the players to the extent I would like. As a creator, it is important that the game communicates what was intended to be communicated, which is something that I felt Dangerous Embodiments lacks.

The MakeHuman creation platform is certainly a fantastic way for a low-budget team to create models that share traits with their subjects, but MakeHuman has its shortcomings. By being a jack-of-all-trades, MakeHuman is clearly a master of none. For instance, while the three race sliders for Caucasian, African, and Asian represent a strong majority of the world’s population, the system doesn’t make it very easy to model other people. MakeHuman models also have a tendency to start looking inhuman pretty quickly. Some slider settings just do not mix well.

Nevertheless, the larger question might be this: how perfect can a digital model ever be? Current technology does allow for a great deal of photo-realism, but that’s at the cutting edge. Moreover, trying to model subjects that have limited or non-existent representation in photographs or artwork makes the job hard. Even if these media exist, how much can we believe what they portray. No looks good in a mugshot, and everyone looks fantastic under studio lighting.

Perhaps it’s more important that the model evoke the proper feeling. The digital embodiment probably shouldn’t be the focus, rather it should be good enough that it doesn’t distract the player. The model should get out of the way to let the player experience the game. That doesn’t mean the player model is unimportant. The character may well be crucial, but it shouldn’t need to undergo closeup scrutiny.

The creation of virtual bodies for a historical recreation is a difficult process. Every aspect of the design of the figure must be intentional; no single feature or component of the body can be represented in a certain way without interpretive or historical reasoning. This process is further complicated when one attempts to create a digital embodiment of a historical figure. In many cases, the creator may have no evidence of what the person being represented actually looked like, and must carefully exercise historical imagination to represent the body in a respectful and ethical manner. I grappled with some of these issue when I attempted to create a virtual representation of James Moore, the first master of the Gressenhall House of Industry. I found myself questioning every minute detail I changed; why did I make that choice? Can I justify that? What does it mean to represent him this way?

Issues also arise in how the player is meant to experience the character. In a first person view, the player is immersed in the body and sees through the characters eyes. In a third person view, they can see the body itself at all times, and are therefore constantly aware of it, but they do not see through it. It is also important to intentionally design an environment that reacts to and interacts with the character. If the body is incidental to the space in which it moves, there is not meaningful interaction between the space and character.

While working on the MakeHuman platform, I engaged with the pauper I had been researching in a new way. Being responsible for her body meant thinking about her recent pregnancy in physical terms. I had to negotiate between considering the extra weight she might have gained from pregnancy and the lack of nutrition in an 18th century pauper’s diet. Where in her body would each of these manifest? How would her labor tasks from before her pregnancy have affected her musculature? I realized I didn’t know and would have had to circle back to do new research.

Even though I couldn’t shape her appearance exactly as I wanted, I was able to engage in some substantive process-based questions during my MakeHuman session. After an avatar (a product) is built, though, those process-based experiences can disappear for the user, as the experience playing Dangerous Embodiments taught me. However, allowing a user to create an avatar without any historical information seems equally, if not more, at odds with process of building empathy in a historically authentic way.

In terms of viewing digital embodiment and VR as empathy machines, the Dangerous Embodiments software did not necessarily achieve that end for me. As we discussed briefly in class, the game modeled an open world exploration experience (with no goals or interactions) and gameplay was the same regardless of chosen character. The default view–following the backside of the character so you can never see their face–actually distances you quite a bit from your character with whom you attempt to empathize; in effect, the whole empathy element of the game becomes refigured by superficiality. However, there might be some merit to this kind of distanced and equalizing embodiment. At one point playing as the female slave in the plantation world, I discovered a door that would not open, and later came to find that the door was also glitchy for the male plantation owner as well (I had expected the door to be accessible to the male owner). Perhaps the act of bringing my own expectations to the game is educative in of itself, even as gameplay remains unchanged through different characters’ perspectives.

An entirely different experience of digital embodiment for me was engaging with the MakeHuman software. Given the ability to sketch up pauper avatars in a matter of minutes, I quickly became concerned with the level of detail at which users could exert control. Put in conversation with historical accuracy and authenticity, this sort of granularized avatar creation urges users to make guesses without adequate background knowledge. What nose height or eye distance should we give Mary Jones, a question we ask while we have virtually no physical descriptions of her in our archives? Clearly it’s a difficult question, and yet the software encourages uninformed choices in the name of experimentation– lest our pauper be given default characteristics.

The Virtual St Paul’s Cathedral Project has ambitious goals if it wants to “enable us to experience worship and preaching at St Paul’s Cathedral and in Paul’s Churchyard as events that unfold over time and on particular occasions in London in the early seventeenth century.” On the website, these goals only come to us in constituent parts, through playable audio recordings, image captures, text-heavy descriptions, and a painstakingly slow flyover video. As such, it is not immersive or a “fun” game-play experience online. However, the version physically installed at NC State sounds more immersive. The promised experience of being surrounded by St. Paul’s sights and sounds does seem more achievable in a physical room than a flyover video. Hopefully the effects of the room will give them an atmosphere to strive for online as they continue to develop the project.

It is also worth noting that the visual and audio models meticulously represent aspects of St. Paul based on thorough historical research. Furthermore, the project’s creators are well aware that it is a work in process, so it seems unfair to evaluate it as a final product.

Sources:

“Overview.” Virtual St. Paul’s Cathedral Project: A Digital Re-creation of Worship and Preaching at St. Paul’s Cathedral in Early Modern London. Accessed February 9, 2018. https://vpcp.chass.ncsu.edu/.

The purpose of the Virtual St. Paul’s Cathedral Project was to “explore public preaching…enabling us to experience a Paul’s Cross sermon as a performance…in real time”. At first sight, the project seems to be a product-based question because the player “experiences” a digitally-created performance from the 17^th century. However, upon further investigation, this project was more intended to explore process-based questions. The makers dwelled on how they transformed a rough rendering of a church yard into a historically accurate church yard with sounds and sights. They mainly grappled with the “formative discussion and development” section of their project where they make clear distinctions of where they used historical information, made approximations and recreated lost experiences. This helped them contend with their own assumptions of how the Paul’s Cross sermon looked and sounded like. I appreciated how they acknowledged their own biases in interpreting this event and were open different models with other interpretations of the same event, but I had some hesitations regarding whether the process-based question they pursued was effective in allowing greater public accessibility.

Since the purpose of the project was not meant to be a game-like experience, it seems irrelevant to determine whether or not they fulfilled the humanities scholarship or “fun” experience balance. However, looking at the website alone, it wasn’t very accessible or cohesive. It was not very user-friendly and on the website pages, there was a lot of text with the same terms bolded. The acoustics part of their project was not added to the visual “fly over” videos, making my experience of the fly-around not very unified. Thus, although not all “serious games” must meet a game-like experience, I wonder what the best way is to engage with humanities scholarship while remaining accessible to a larger audience.

I explored the University of Arkansas’ digital recreation of the House of Prince of Naples, one of the houses in Pompeii. This particular model is focused on the murals on the walls, and two versions (one during the day, and the other at night) illustrates the placement, location, etc of the murals in a way that a picture or other 2D reconstruction would struggle with. Their goal seems to be more academic, with this model being part of a larger effort to build a database of visual art references. With that intention, this model serves that purpose.

Given the more academic context in which this is presented, it seems unfair to judge it for it’s accessibility towards a more general audience. However, it still is engaging. I absolutely geeked out over this experience, having learned a bit about these murals and the styles in a Roman archaeology and art class I took last year. This would have been a handy tool for my professor to use when teaching. Furthermore, if desired, it could be remodeled to fit a general public audience. A choice to further explore the styles and subjects of the murals by a pop up or other mechanic, would inform the audience as to what they are walking through while still being able to explore the space.