Dr. Drang's Writer Workflow

February 01, 2012 by Gabe | [mmd] |

I’m not sure if Dr. Drang needs an introduction. He’s a mysterious super-hero engineer holed up over at Leancrew.com. I don’t know his true identity and hope I never will. He’s often inspiration for me here and one of the two guys that really got me thinking about how other people write. He’s also one of the most generous people I’ve interacted with on the Internet.

Why do you write at And now it’s all this?

First, I’ve learned a lot by reading blogs and other free content on the internet, and it seemed only right that I offer up the things I’ve learned in return.

Second, I need to write for work, and I wanted to get better at it. It’s not that I thought I was a poor writer, but it often takes me a long time to get my words in a shape that satisfies me. I thought that with more practice I’d get more facile, better able to bring the right words on the fly rather than laboring over every sentence. In this I’ve had mixed success, I’d say.

Third, I started out as a teacher and there’s still something in me that wants to explain. If you substitute “show off” for “explain” I wouldn’t argue.

How do you capture your ideas and research an article for And now it’s all this?

For the articles in which I’m presenting a script, the research occurs while I’m programming—I usually don’t have to do any more to write the post. And since I usually write the article within a day of finishing the script, the web pages I used are usually right there in my browser history.

When I run across something I want to write about but know I won’t be able to do so for a couple of days or more, I bookmark it in Pinboard or drag a .webloc link into Dropbox. Most of these “articles” never get written, either because I cool on the topic or because someone else writes a good post about it before I do.

The engineering articles (and no, I don’t consider programming engineering—I’m old-fashioned that way) usually come straight out of my head with no outside research. The information in, for example, the snap-through buckling post (still my favorite post ever) is material I’ve been working with for so long I don’t need to look anything up. Sometimes, as with the torsion spring post, I’ll dip into my library to look something up. I’ve had some of my engineering texts for over three decades; the familiarity I’ve developed with them makes them a much faster reference than the internet.

Can you provide an overview of your writing process?

If the post is about a script, I usually start with two TextMate windows: one for the script and the other for the post. There’s usually a Terminal window open so I can run the script and capture its output. And, inevitably, a Safari window with many tabs open. If the project involves AppleScript or Automator or TextExpander, those are open too. And Dictionary—very helpful when I know there’s a better word but can’t think of it.

I generally write a post from start to finish, in the order it’s read, and I almost never outline a post. Although I tend to be long-winded, even my longest blog posts aren’t so long that I have trouble keeping them organized in my head. This isn’t to say that I always know everything I’m going to say when I start writing. I often go back to add or rearrange things after I’ve reached the end of a post and realize that its emphasis changed as I wrote it.

Most of my “process” consists of my sitting and looking at the screen, trying to figure out why the thoughts in my head aren’t turning into coherent sentences. And there’s also a lot of drinking of tea and getting up to go to the bathroom.

How long have you been doing it this way?

The dates of my earliest posts got screwed up during one of my blog engine changes, so I don’t know my blogversary. But And Now it’s All This started up somewhere in mid-to-late 2004, shortly before I switched back to the Mac from Linux.

Do you have a specific work environment or setup for researching and composing an article?

No specific environment for researching. Composition is usually done while sitting in a Poäng chair with my MacBook Air resting on a Bräda laptop support. Feet up on a footstool, tea in Thermos stainless steel tumbler on the nightstand to my left. The same position I’m in now.

Posts are usually written at night, while the kids doing homework or sleeping. Weekend posts are sometimes written early in the morning.

Does your workflow change based on the type of post?

A little. Some posts require sketches, which I usually do in OmniGraffle before I start writing. Sometimes I have to take photos; again, I try to have them done before the composition phase. I’ve taken many of my blog photos at work, where I have access to big sheets of blue paper to use for a background.

There’s one post that’s had a very different workflow, a post I’ve been planning to write for two years. It’s a review of a paper, “On the Mathematical Theory of Suspension Bridges,” published in the Transactions of the Royal Society in 1826. It’s what you might call a narrow-interest topic, but it’s fascinating to me because both the math and the engineering in it are so foreign to the way we do things now. For example, the author, a Brit still under the influence of the Newton/Leibnitz calculus war, solved the differential equation of the catenary using Newton’s fluxions instead of differential notation. Even weirder, our modern notions of stress hadn’t really been worked out yet, so the strength of the iron chains in the suspension cable is expressed as a length: the length of chain that could support its own weight if hung vertically.

Anyway, I have lots of handwritten notes that I took as I read the paper and worked out what it was doing. I also have references to contemporaneous accounts of how suspension bridges were being built in the early 19th century, and a comparison with with what I was taught about 20th century suspension bridge design. I’ve been sitting on this stuff—which I did for my own amusement—since late 2009, trying to figure out how to fit in a bunch of material and make it as interesting to my readers as it is to me. One thing I’m certain of: no one is going to beat me to the punch on this post.

What are your custom tweaks to your workflow?

My workflow is all tweaks. TextMate snippets for Markdown reference links. TextExpander snippets for grabbing the URLs of Safari tabs. A Python/JavaScript/TextExpander hybrid for embedding tweets. Scripts for formatting tables. Scripts for Apple affiliate links. A script for taking screenshots and uploading to Flickr. To me, it’s the malleability of computers that makes them useful and fun.

What parts of your workflow are you looking to change or improve?

I’m experimenting with your system of inserting all the links at the bottom before writing any of the text. That’s a more premeditated way of working than I’m used to, but I like the idea of staying in my text editor as much as possible. Several of my scripts were written to avoid context switching when adding links; inserting the links first may be the best solution.

Several months ago I started using my Flickr account as a sort of CDN for the images I put in my posts—mostly screenshots. It’s worked well, but it’s left my Flickr account a bit disorganized. I need to figure out a better way to categorize the blog images.

What parts of your workflow are you least willing to change?

TextMate and the Blogging Bundle. Cold dead hands.

Anything else you would like to share about your workflow?

Developing habits is more important than what those habits are. I don’t need to sit in this chair or drink this tea (Twinings Earl Grey) in order to write, but making the tea and turning on the light over my left shoulder get me in the mindset of writing, even—or especially—when I don’t feel inspired. Everyone’s habits sound silly and precious to someone else, but if work becomes part of the habit, then your rituals are worthwhile.

Unless your rituals include fetishistic coffee making, which is loathsome.