Comment by sqs | Hacker News Reader

Comment by sqs | Hacker News Reader

Parent Full thread

My first question was about how writing prose differs from writing code. The post covers this:

> despite being apparently similar activities (both mostly involve slinging text), the temporal timing of software development & writing are strikingly different. Thinking back, I don’t recall early-morning programming being a trend among programmers (programmers are infamous for preferring to come in late and late-night programming sessions which may wrap around the clock, especially in college). It’s fascinating that the stereotypes about writing vs programming line up so perfectly with the RescueTime data.

I am very curious to learn/think more about why the 2 activities are stereotypically (and empirically) on opposite schedules.

Personally, when writing prose (mostly docs, blog posts, and product plans), I prefer to wake up early and do it in the morning. I much prefer coding at night. My hunch is that this is because interruptions are rarer at night and more bothersome when coding.

For anyone else who is both a writer and coder: do you follow the same schedule for both?

For me, writing prose (mostly short stories, journals, memoirs) is more emotionally taxing. To write well I need to find ways to silence the self-censorship that will inevitably prevent me from expressing the things I really want to say. This often happens when I am in a state approaching mania. Donald Glover says he writes in the morning because it is "closer to dreaming".

Coding is primarily a task about cognition, so focus/concentration wins. I find I code best in the late morning, when I'm well rested and have had a good breakfast.

Donald Glover says he writes in the morning because it is "closer to dreaming"

This is really the crux of it, in my experience. The first things I write down in the morning are usually continuations of subconscious creativity that began in the last dream I had before waking up. And usually my dreams are emotionally rooted in whatever problem I was solving the night before, which is especially helpful in making the stimulus of dreams fruitful.

First thing in the morning, my recall is very good, and I am lucid about the essential concepts needed to explicate what I am really trying to solve in abstract terms. After hours of working on a problem that's going nowhere? Not so much. But then a break or evening hike, and I am just as creative again until bed.

That said, for skilled activities like coding and playing a musical instrument, the temptation to be too creative before warming up and trying some experiments can probably only hurt your productivity, because if you sink your teeth into an idea before figuring out where your real problems are likely going to crop up, you're just committing to something that's probably unworkable.

So if I have lots of code to write, I instead simply collect a list of bugs in the morning that I intend to squash, grab a bite to eat, and then start hacking away. Then very often, by early afternoon, there's the chance to reach some kind of minor epiphany, where it becomes painfully obvious that something simpler can obsolete the need for the kind of hacking done in the morning. ;-)

If you're doing independent research that requires coding AND creative writing, I recommend going back and forth between the two approaches, perhaps alternating by day.

Two tools I've found useful for getting text on page:

* https://getcoldturkey.com/writer/ -- just a very simple full-screen text composer

* https://otter.ai/ -- an app for doing transcription of human speech. I find it very useful to go to a quiet area and talk into it and then use that as my very rough draft of my thoughts. You'd never use it for stenography, but the quality is good enough to make it clear what you meant to say.

and if your voice of self-criticism says "that seems like a crutch", reflect on the fact that you are a tool-using human who will probably wear shoes tomorrow and many days thereafter.

I do both, and I also tend to find it easier to write prose early in the day and code later.

I think it’s because of the depth of focus that’s required and how fragile it is. I can have a productive 30-60 minute writing session all the time, but unless it’s very trivial work, it takes me so long to get fully in the zone of maximum productivity when coding that it’s really not worth doing unless I have a few hours. And my morning rarely offer that, so I use them to clear the decks so I can do deep work like coding or whatever later in the day.

I’m not a natural morning person, but i routinely get up at 430am. I’ve deliberately adjusted to this masochistic schedule over the course of 15 years because I find it so superior in terms of effects on my life. I’m not a natural daily exerciser, healthy eater, relationship builder, etc. either. Sometimes it’s good to push yourself to do things because they’re good for you.

I wake up before 4am (between 11pm the day before and 2am usually). While everyone else spends the last 6 tired hours of their day on their hobbies, I spend the first 6 best hours of my day on my hobbies (writing, coding, etc). I'm definitely nocturnal, I'd just rather not be sleepy, so night is the beginning of my day.

I think it works for coding too.

I don't like morning and prefer start coding late. But if I can wake up in early morning and start work in first half-hour, it improve my coding skills: I write better code and do it faster.

But it depends on personal qualities and daily regime very much.

And one interesting thing: it show enhancement only for first times. My current experiment is wake up at 5-5:30 a.m. And in first time I had improvements to my productivity, but now I work like in the past before experiment.

Morning coding works best for me.

Night coding also works, but tends to mess with my falling asleep and decrease the quality of my sleep. If I'm in a tight spot I find it more effective to not work later, but to sleep a bit less and start a few hours earlier.

I get most if my productive day done before everyone arrives. Then I just muddle through the day at limited capacity until either the evening or next morning.

I think there are 2 keys to this:

* I think there is contention on some mostly unconscious cognitive resources. Social processing seems to be an antagonist for deep focus for me. I focus deeper an longer when no one else is around.

* There seems to be memory/history effects in the short term. In the morning some "buffers" and "staging areas" are clear making it easy to build a good focus and flow. As the day progresses it looks like these can get scrambled.

I do a lot of both writing and coding and I also follow this schedule. I find coding at night to be very efficient, but when I try to write at night, I end up rewriting the same paragraph 10 times before giving up.

No idea why these activities are so different, but I feel the thought process is very different. Coding requires active concentration. I find writing to be more of a passive activity. Too much concentration, and the words stop making sense -- something like this effect: https://en.wikipedia.org/wiki/Semantic_satiation

There was this notion that at night the brain is more tired, meaning it cannot both focus on the code and daydream (or similar). From personal experience, this is at least partially true.

Reply on news.ycombinator.com