My dad is a scientist, a semiconductor physicist to be exact. But surprisingly, I don’t recall myself being particularly interested in becoming a scientist myself, partially because what my mom does seemed much more exciting to me. She used to be a software engineer, writing Visual Basic back in the days.
One decade later, I somehow find myself happily in grad school, but still not completely satisfied with the idea of being a scientist. Instead, I want to be something more like a “Scientist(-engi)neer”, a first scientist then engineer. Kind of like the combination of my dad and mom. Combining Wikipedia’s definition for scientists and engineers, I came up with this definition of a scientistneer.
Scientistneer – someone who primarily engages in a systematic inquiry to acquire knowledge that describes and predicts nature (including physical, mathematical and social realms), but also designs and builds materials, structures, and systems while considering the limitations imposed by practicality, regulation, safety and cost.
The first thing I learned in grad school is to establish a structured way to ask questions, get answers, and report on the answers. This is the scientific method process, as illustrated in the graph below.
Here is an explanation of how this process plays out in my PhD life:
I study social technologies. And therefore I make observations in the real world. I have been inspired by platforms such as Secret, Yik Yak, Happn, Airbnb, and have published papers on each of them.
I then go to social science and HCI literature to understand what we already know, and what’s new. This way I can then formulate interesting questions that our previous knowledge base doesn’t adequately cover (or perhaps I missed it in my literature review).
Then I formulate my hypotheses, that generalizes from specific phenomenon and focus on a self-contained and likely repeating phenomenon that has implications in many other situations. Then I design either experiment, survey, or interviews to test my hypothesis (sometimes repeatedly).
Finally I report my findings through a coherent story, perhaps using some existing theories or developing new ones to allow us organize the findings into the rest of existing knowledge base.
All these processes are collaborative, I bump ideas to my peers and my advisor all the time to avoid doing research in a vacuum.
My favorite stage of the scientific inquiry is the step from making observations to thinking of interesting questions. It’s a process with a lot of freedom and set the tone for the rest of the process.
I think this is also the reason I like photography – to see the difference in the ordinary (new lenses and angles); and to see the ordinary through the strange (a form of generalization and pattern finding).
So, three years into grad school, I finally begin to really understand this illustration of the PhD.
If you are thinking about getting a PhD, read the full illustrated guide to a Ph.D.
But are we done here? The responsibility of a scientist may stop here. But not a scientistneer.
Design & Build
What is the use of knowledge? I am not asking the question from an epistemology view point (but you are welcome to ponder that).
From the definition of a scientist, not only the knowledge we acquire describes our world, but also it can predicts the world’s future state. This is a very powerful thing, but its power is not used unless we design and build something informed by this knowledge.
Another thing that design and build can achieve is to alter. Knowledge cannot directly alter the state of our world. This is the root source for many forms of frustration in PhD life. No one cares about my research. No one would ever read my papers. None of these matter. Why am I doing this?
Design and build process is almost like a therapy in this sense. The tiny sense of achievement you feel when you hack something together that actually works, that does something for you, or for your user (be it one person, two of your friends, or may be eventually hundreds and thousands of them… who knows).
It matters. You have altered the state of the world with your knowledge. Even just a tiniest bit. It matters not only to you, but it does something for someone else.
Productivity & Creativity
But I’m no engineer. You may say. Me neither.
That is not an excuse to not to build. Each year it becomes easier and easier to build something. Because collectively the technological ecology makes it easier and and more efficient to create, driven by (cynically) cycles of investment and business profits. You can build websites now easily with Squarespace, prototype with Sketch and Flinto, deploy with AWS or Heroku.
And once in a while, you stumble upon something that really makes your heart tick.
Some technologies I picked up recently and absolutely love are:
- Flask (A lightweight python webframework)
- I can stay in python where I trained my fancy machine learning algorithm and implemented all the pre-processing modules
- I can take user input through a web interface and returns the result in a beautiful webpage rather than dumping them in a text file
- Jekyll (A simple, blog-aware, static site generator)
- Jekyll is the engine that powers GitHub Pages
- I built this website with it
- Super easy for me to add content: I just type into my Sublime editor in plain markdown, and push a button. Everything is organized and published online.
- I just picked this up this week
- Fell in love with the concept of functional programming through this Medium series.
- Pure elegance (I will write about it more when I get more familiar with the language)
Hey, they say the best way to become creative is to create a lot of work. Considering that in science, every project takes forever, it’s not possible to work on a lot of projects. But when you build, you can define the scope of each project to be very small. Therefore you can stay on a stream of creation, and from there maybe some of your best scientific ideas would find roots.