pursuit of efficiency
I haven’t written much since high school.
Even during high school, the idea of writing something gave me a bit of a knot in my stomach because I never did particularly well in my English classes.
I think STEM people can somewhat relate. We gravitate towards STEM subjects because the solutions are not as ambiguous compared to those in the arts.
Recently, I started practicing for one of my art history writing assessments without the help of AI, since the inclass assessment would be done on paper.
Although it was only a couple of short paragraphs, putting my thoughts into a cohesive paragraph after not doing it for years was pretty hard. It took me over an hour to get a decent working draft.
Though, when I read over what I’d written, I realized that despite finding the process a bit difficult, I had really enjoyed writing it.
It was fun to write, which made me wonder what things should we use AI for, and what should we do ourselves?
Use of AI
I personally think AI is great at making things efficient.
I can code much faster and do so much more with AI. Even for school courses, when I’m stuck, I don’t need to wait for office hours or Piazza replies, and nowadays, I find that AI is correct most of the time.
I feel that I can learn more, more efficiently, and focus on actually understanding concepts instead of being stuck or waiting for someone to point me in the right direction. The same goes for reviewing literature or finding related papers. AI is great at that.
It also helps spot mistakes in code or emails that we all make from time to time, which I think is really useful.
But recently, I tried vibe-coding a solution for Toronto city bikes because I started biking almost every day to school and around downtown Toronto.
The bike stations near my place are always empty, and the stations where I want to park are always full, which always make me walk around everywhere to find a spot.
I personally wanted to figure out why this happens and see if there is a solution or at least an optimal way to plan my trips.
The ML solution I vibe-coded seemed to work. But I didn’t really understand what was happening under the hood in terms of the data and the model, which was what I was interested in in the first place.
So I found myself stuck in a bit of a dilemma.
Because I actually cared about this problem, it didn’t make sense to outsource the process of digging into the data and the problem to AI.
Would it be more efficient to do so?
Probably.
But if you care about something, you want to know more about it.
I find that it is a bit of a contradiction for me to say I care so much about something but do not want to know more about it.
This reminded me of a post that I read some time ago and I borrow his ideas below as I could really relate to him.
Meaning of Value
I did a lot of visual arts during high school. My school had an amazing arts program, and although I was more of a drawing & painting person, I would try different mediums. One medium I tried was pottery.
I would throw cups and bowls quite often and since I went home every summer break, and since I knew my mom liked collecting bowls, I would always spend time throwing some bowls to bring home.
The pieces I threw were far from perfect and were not very practical.
They’re not as thin or light as those sold in fancy ceramic shops or Japanese kitchenware stores. The glaze isn’t very consistent either.
But the bowls I brought home are, at least in my view, much more valuable to me and my mom than those nicer ones you can buy.
And I think what makes those bowls valuable is the time I spent making them.
Because our time is so limited, choosing to spend it on something or for someone makes whatever it is much more meaningful.
AI and Efficiency
Coming back to AI and efficiency, if there’s something we value and care deeply about, we would want to spend time on it.
There’s nothing wrong with being efficient, but the more time you spend on something, the more valuable and special it might feel to you.
This isn’t to say we should use less AI or do everything ourselves to create more perceived value for ourselves. I enjoy making things efficiently and I think the majority of jobs are better done with the help of AI.
It’s more that for the things we care about, whether that’s time with friends and family, your career, your personal project, or the arts, efficiency doesn’t always have to be the objective.
It’s a very weird thing to say as an industrial engineering student but not everything is meant to be an efficient process.
And life is really not efficient in any way; however, that doesn’t make it any less valuable.
To summarize, we live in an world with in pursuit of efficiency. Efficiency is highly valued and is rightfully so.
But sometimes, I think it’s okay to slow down just a little bit and waste a bit of time on the things we care about because doing that makes those things more special.