Why and How I Write

One of the things I’ve enjoyed in the TechWriters forum has been folks talking about their writing process. A few folks mentioned the Fieldstone writing technique by Weinsberg. I’ve enjoyed Weinsberg’s other books, like “Being a technical leader” and “Secrets of a consultant”, so I decided to give it a read.

You can summarize the Fieldstone writing technique as: (1) collect related ideas in a file, (2) write stuff that energizes you, (3) switch projects to wherever you have the most energy.

As I started reading the book, I was fairly certain that I was a Fieldstone writer–it sounded like how I write–but as I got into it further I began to suspect I really wasn’t. Probably the key line was one that roughly stated, “If you can’t write down each though you have within five seconds, you’re not a Fieldstone writer.” Well, I’m probably not then.

Another theme in TechWriters is tooling. Most folks seem to use Roam Research to capture their ideas and slowly accumulate them, creating their own personal little wiki that they curate into blog posts (or books!) as they come together. Listening to folks talk about how they use Roam, I was inspired to wish I was someone who used Roam and curated a careful little wiki of ideas.

As a fairly prolific writer, I think it’s interesting to describe how I write, and in particular acknowledging my “secrets” to writing at volume. I guess we can start with secrets first:

  • I spend a lot of time writing.
  • I write when the energy’s there.
  • I publish everything I write, even when I think it’s bad.
  • I only edit stuff that I want to. Usually I don’t edit at all.
  • I rarely have anyone review my work. I have experimented with getting more proof readers, but generally I finish writing something exactly when my energy to write it reaches zero, and don’t have much energy left for editing.

This entire process is optimized for managing my energy and for volume. Where someone else might be collecting notes within Roam for a good post in the future, I’ll just go ahead and post something mediocre today. If it’s an important enough idea and I have more to say, then I’ll write another article about the same topic in the future.

That said, I’ve also spent time refining how I write so that I can create plausibly readable content despite the fact that I rarely edit and filter infrequently. The actual mechanics of how I write are:

  • I have a Google document titled “[larson] stuff to write” which has a list of topics I’d like to write about.
  • Within that document, are links to other Google documents with the provision title and occasionally a couple notes or links. There are usually a few dozen ideas on the list, most of them half-baked.
  • If I have an article idea, I email myself the idea to add to the list later. I’m not strict about doing this, if it’s a good enough idea then it will come back to me.
  • A few times a month I will sit down and collect ideas from emails and off the top of my mind onto the list of articles.
  • Every article goes through an outline phase before I write it.
  • When I’m working on a larger project like StaffEng, I’ll often have two or three articles in an outlining phase.
  • My outlining process is to write the introduction section, identify and bullet other sections, write the conclusion. I rework the article flow a lot at this level of abstraction when it’s very cheap to edit and change ordering.
  • I
  • I rarely write more than one article at a time, although I will occasionally put down longer pieces (5+ pages) because I’ve temporarily lost energy for them.
  • Before publishing, I will always edit the introduction section.
  • If it’s a piece intended for a book or other publication, I’ve recently started running Grammerly on it to catch anything I may have missed.
  • I figure out a publishing date based on other queued content for my blog. Typically I try to do my best available piece on a Thursday, which will cause it to headline the next week’s Tuesday mail digest. I bury worse pieces on Friday through Monday where they’ll be lower in the mailing list. The weekly mailing list is probably my best distribution mechanism despite being only a year old, this is because it’s folks who actually want to read what I write.
  • Beyond the mailing list considerations above, I don’t worry too much about the timing of when I publish onto the blog. Nothing has ever suggested to me that this matters much.
  • I don’t publish my work to social news aggregators, I don’t invest into the communities and voting rings that get you onto the frontpages and defer to other folks who are better at that. This is an absolute privilege of having written for a decade-plus that folks are willing to pick my stuff up, but also keep in mind that most of my stuff still isn’t ever picked up – the majority of what I’ve written has never been submitted to any news aggregator and probably never will be.

Digging in a bit, there are a few implicit beliefs of mine within the above that are maybe worth recognizing:

  • I write to get ideas out of my head and refine my own thinking.
  • I also want to be read, but generally don’t
  • I’ve generally chosen to write content that’s applicable to a narrow audience, so trying to maximize pageviews isn’t a meaningful goal.
  • Most common distribution channels aren’t effective at distributing narrow content. Word of mouth, Twitter, LinkedIn, mailing lists, and RSS are effective at that.

Another theme on topic is just the sure privilege that writing represents. A few good pieces on this topic:

This is particularly relevant during the 2020-2021 pandemic in my opinion, as there are so many fundamental privileges that make it possible for some of us to write and “be professionally productive”:

  • The privilege to stay inside
  • The privilege to do work where “hours worked” and “salary” are decoupled
  • The privilege to get medical care

And so on.