Some thoughts on writing

Dan Saunders
Jun 19, 2018 · 3 min read

I keep a journal for a number of reasons, chief among them to practice my writing. Some days, I don’t know what it is that I want to put down, but I convince myself to fill a page anyways. I think of it as a sort of eat your vegetables ritual which I (try to) do every morning.

Thanks to a recommendation from a friend, I’ve been listening here and there to the Ologies podcast, in which a different -ologist is interviewed every week or so. In the episode on mythology, John Bucher convincingly argues that, in order to be a great writer, one first has to be a faithful reader. This is a fairly obvious practice to recommend, but it resonated especially well with me: improvements in my writing skills seem to stem from what I’m reading!

This is not to say I think I’m a good writer. I’ve had only the typical training, and the rest of my skills are self-taught. In order to challenge myself, I’m focusing on more free-form, public writing, starting with this blog.


Writing becomes more enjoyable through practice. “Practice” here connotes consistency and repetition. At first, it may be forced, and in my case, it was a self-development effort born out of frustration with my own mediocrity. I tried to convince my future self that my words were an accurate depictions of my thoughts and experiences from that time. It’s an excellent way of tracking progress at work, in relationships, and in the development of your own worldview. In time, some nuance developed, and I’ve begun to feel proud of the “voice” that is coming out of the pages. A page’s worth of journaling can take no more than a few minutes, but it feels best when careful thought is dedicated to each sentence.

I have a tendency towards technical writing, which makes sense in light of my perpetual studentship and typical reading material. This is not (necessarily) a bad thing; such writing can be beautiful or inspiring in the right hands (to name a few: Jane Goodall, Carl Sagan, Roger Penrose, …), and various layperson-friendly mediums exist for the dissemination of scientific findings. Several prominent scientists dedicate a chunk of their time to communicating their work, on Twitter or in journalistic settings.

However, I don’t want to constrain myself to purely technical subjects, and neither should the average “technical” person. This is born out of another frustration: the difficulty of communicating what I read and how it modifies my understanding of the world. A goal of science, if I understand it correctly, is to nudge public opinion in the direction of “truth” as suggested by scientific experiments, and it’s desirable to take part in this effort.

I’ve had some success in writing up short summaries of books I’ve read, and sharing these with my friends. I want to do this better; namely, technical jargon and terminology should be discarded in favor of sharing useful information clearly. This isn’t always easy to do; it’s can be difficult to communicate simple heuristics for understanding parts of the world from, say, math-laden textbooks (somewhat paradoxically).

But, an effort can always be made! I like the distill.pub venue for exactly this purpose, although, quality-wise, it seems a bit out of reach for people like me. So, that is one purpose for this blog, and why I will continue to make an effort to write.


Another thing that motivates me is the notion that good writing is beautiful. The topic doesn’t matter much as long as the story is told well. We are inclined to only consume media that support our philosophy of life, politics, or life’s work, but any good story is often so because of its structure and nuance and choice of words. You needn’t be convinced by, or agree with, everything you read; it’s often more useful to read something disagreeable and think about whether or not your beliefs need revision.

So, I’ll try to write for myself and for potential readers to appreciate, not for content alone, but also for form, elegance, and convincing.

Dan Saunders

Written by

MSc student in computer science at UMass Amherst. Likes machine learning and brain analogies. https://djsaunde.github.io