Make, Try, Revise:
Before this class, I thought game design was a lot more linear. I thought you came up with an idea (that had to be awesome from the start) and then coded it up, and ended up with something hopefully great by the end. What I naively didn’t realize was that there’s so much pivoting and changing based on play testing and working on what you’re interested in. If you don’t think the main premise of your game is exciting, change it! If people think a mechanic sucks, change it! However, the only way for us to figure that out, is to try making something and then have people test it.
Sketchnotes… love/hate relationship?
This is my 3rd CS247 class in a row this year (247A, 247B, 247G) and I hated sketchnotes to start. I also hated them last quarter. I think I still hate them, but just a little bit less. Why? I’ve grown to be a better artist which makes me proud of my sketch notes. I also learned that I don’t have to read or watch the entire assigned document, I just have to find things that are interesting to me and sketch them. I’ve found that I get a lot more value and enjoyment by trying to sketch out things that others will find visually appealing, than just writing text that matches exactly what the reading says. It’s also a chance for me to practice synthesizing ideas, which is nice.
3 hours is too long:
Additionally, with this being my 3rd CS247 class this year, I hate the 3 hour class length. To be blunt and completely honest, I don’t think it uses class time efficiently and leads to somewhat pointless lectures where students just do their own work in class instead of paying attention. We’re promised “lots of team time” (which in this class we got, but not in A or B) but in reality we get very little, leading to two 3 hour classes per week in addition to trying to meet with our groups outside of class to get more team work done. I think the class could be 1.5-2 hours long and the students would get the same amount out of the class. I have a lot more thoughts about this and HCI as a major concentration so if you’re interested feel free to email me: nhafer@stanford.edu! I’d love to chat more about it because I know y’all put a lot of effort and thought into these things.
LLMs are here to stay:
In lots of classes at Stanford, LLM = honor code violation. But whether you like it or not, they’re here to stay. I think HCI should embrace that and teach students how to effectively engineer prompts and use LLMs to our advantage to make things faster and better. I like the 247 classes because we get to focus on the intentional design. Like in 247A and B we made Figma mockups but didn’t have to worry about the JS, CSS, React, etc. to make it work. I think that content is valuable to know, but I hated CS147 because they expected you to know how to make an app, but the point of the class was design and iteration, not making an app. Anyways, the TLDR is I think we should talk more candidly about how LLMs are here to stay and how we can leverage their extreme power to make and test designs more quickly, efficiently, and effectively.
