If it is possible to have a favorite pseudoscience, mine would have to be personality tests. If you’ll allow me, I have a few speculations on why.
Firstly, we love to talk and think about ourselves foremost, but loathe the effort required for real self-reflection. Taking a Buzzfeed quiz about what fruit you are is a ridiculous, but meaningful way to feel identity. Following that, it becomes a way to connect to others. What fruits are all your friends? Is there some sort of fruit compatibility? Some go much further: maybe your fruit type says something about the job you should work, who you should marry, and what sort of a parent you’d be.
Many companies and organizations believe in the predictive power of personality tests. As a child, I remember taking supposedly scientific personality tests to find my “learning style” in school, which experts largely regard as a myth. More recently, I had to take a personality test for a software job. It featured such questions as, “Susie and Bob buy a picture book while on vacation. Do they read more or less books than the average person?” How this relates to programming is hard to say. This company ended up rejecting me after I submitted the personality test, only to then contact me again a few weeks later to move on to the next stage because “their hiring needs changed,” a strange contradiction if the test results meant I was a bad team fit.
I mention this because personality tests are a form of data collection. Ultimately, as engineers, it’s up to us to implement such systems, which we often do without question.
Over the past few months, my friend had me watch an anime movie with him that would go on to become an obsession for us. Umamusume is a franchise about Japanese race horses, depicted as anime girls (obviously), that is shockingly full of soul. The stories and personalities are based off the real racehorses. We did our own research into personality tests over the past few months and made our own ultimate “which Umamusume are you” quiz. We combined the TIPI (a tiny version of “the Big Five” or OCEAN test), the AQ-10 (initial autism screening test), and fun interest questions we wrote ourselves.
I’ve worked on plenty of small software projects, but what was different this time was we actually sat to discuss the ethics of our site. We talked about what the modal to disclose the autism screening portion should be, and the language to use for it. We considered the benefits of helping normalize neurodivergent behavior by including it, when the idea originally started as a joke. We debated how data collection should work and be disclosed, as we want telemetry to improve the test and visualize results, but are also strong privacy advocates. We ended up having to make another control flow in our site to ensure users that rejected sharing their analytics would still get an equivalent experience. Your implementation of any project reflects the thought – or lack thereof – put in for the end user.
Raising ethical and privacy questions is difficult. It is a habit that needs to be formed from working on your own, and takes a little confidence in a group, as it often has implications on how the system will work. However, it is the mark of an engineer that is thoughtful and intentional in their work. Though it may seem like more work for no material benefit, that care makes all the difference in who you are and how you create things. Though subtle, that difference can be seen by others, and it’s the mark of a talented engineer that can think beyond the screen.
If you’re curious to take our Uma test yourself, you’ll find us at umapoi.moe sometime in early April!
Leave a Reply