I’ve always been a quantity over quality artist, and what I mean by this is that rather than being intentional about what I’m doing or meticulously planning things, I typically make a lot of garbage and hope for the best. This isn’t a particularly efficient way of working, but I personally find it quite effective. There are three main benefits of half-assing my art process that I would like to point out here:

When I’m trying to approach a highly specialized topic for the first time, one of my tactics is to find a recording of a research-geared workshop about it and watch it as my introductory crash course. The benefits of this are as follows:

• I am learning about the subject from (hopefully) a credible expert in the field.
• Workshops and talks usually try to be self-contained, which means basic background info will likely be given and I won’t have to pore through 10 different research papers, searching for an obscure definition, in vain.
• Talks typically have the benefit of including visuals and informal intuition conveyed by the speaker, which generally won’t make it into academic papers, because they’re not rigorous. However, the visuals and informal statements and intuition are invaluable for gaining a better understanding of the information.
• The speaker will generally include a bibliography and mention related works, which is a great jumping off point for further investigation and saves me from having to figure out what the seminal sources are myself.

However, there are also some drawbacks:

One of my quests during the research project I did last summer was to find a way to simulate pointers in Python, or at the very least, get some sort of named reference-like behaviour from the language. Unfortunately, I quickly learned that this is quasi-impossible, because Python is a language that aims to abstract as many implementation details away from the user as possible, and gives programmers very limited ways in which to interact with the language.

While all NP-hard optimization problems are identical in terms of exact solvability, they may differ wildly from the approximative point of view. If the goal is to obtain an answer that is “good enough”, some problems become much easier (such as KNAPSACK), while others (such as CLIQUE) remain extremely hard.

In some ways, photography is to visual artists what piano is to musicians. I can get you to play a somewhat accurate rendition of Mary Had a Little Lamb in about 2 minutes, and if it isn’t in tune, that’s the piano’s fault, not yours. Similarly, if you have a camera, you can create an image in about two seconds, and if the image quality sucks, it’s probably the camera’s fault, not yours. But have you ever tried playing a bowed instrument? I did, for over a year. I doubt I ever actually played a note in tune, and my tone quality bordered on unlistenable. Drawing is similarly difficult; my first two years of drawings are generally not good.