Thread by @iamscicomm

Welcome to Friday. I'm tired, you're tired. Let's make this a relaxed learning environment. I'm tweeting from living room. Today, we're going to learn about structuring an article.

*my living room. I'm a professional editor.

It's hard to learn about writing without looking at people's past mistakes. I'm not really in the mood for dumping on someone who doesn't deserve it, so let's take a look at something I WROTE.

Here's an article I wrote in October 2017. It's about some fun research where a virus and a transposon (a piece of DNA that reproduces on its own, they're awesome) MERGED. Like, became one. It's cool stuff.

massivesci.com/articles/genes…

I think it's useful here to look at article from a bird's eye view. Here's what a bird looking down directly on to your laptop would see if you were reading this article.

Also just for fun let's imagine the bird can read.

Similar to how you want to pitch an editor, you need to hook your readers with the juicy, fun science bits as early as possible.

Now, having said that, where do you think I put that wild, crazy concept of viruses merging with autonomous parasitical DNA? Sentence number one? Two?

nope allllllll the way down in PARAGRAPH NUMBER SEVEN

Don't do this. Don't make readers wait for the good bits. Put a good bit first. Of course, save some for later too.

Think of it like world building on a very small scale. If you're reading a book, and nothing interesting happens for the first 10% of the book, a lot of people are going to stop reading coughDragonboneChaircough.

What did I spend all that precious time at the top of the page with? Uh, let's see. The first few sentences are me waxing poetic. It was fun to write, not that much fun to read though.

Paragraph two has literally nothing to do with the rest of the article. I just wanted to get a word in on what I think of 23andme.

Paragraph three has some relevant bits about what genomes are like, but it could have waited til later, after actually illustrating why genomes are messy. It is not necessary to spend this much on Barbara McClintock or any time at all on the Nobel Prizes.

Here are paragraphs four and five. They're actually pretty good! But, a reader who has no idea what's going on might not make it this far, because I STILL haven't introduced the story at this point. These paragraphs would be so much better if the story had started before.

Here, at last, paragraph seven. SEVEN. Seven paragraphs of no story. We're like 2/3 of the way through the article. If you ever got 2/3 through anything with no payoff you'd be upset, amirite.

The rest of the article is totally fine. It's buoyed by the fact that the story has started. This has been a very long winded way to say: start the story first, everything else can come later.

Let's look at a better structured article. I only do callouts when it's POSITIVITY TIME.

Here's @qeastman writing for @massivesci about a new discovery about narcolepsy.

massivesci.com/articles/narco…

Here are paragraphs one and two. No preamble, you know right away what this is about. The first WORD is narcolepsy.

Paragraph 1: this is what narcolepsy is, and how serious it is. "How serious it is" is a good way to indicate to the reader why they should care.

"Why you should care" goes at the top of the article. My herpes/transposon article never even got to why the reader should care.

Paragraph 2: what the story is. No time is wasted. We didn't know what causes narcolepsy, which is interesting because it's such a serious, oft-talked about disease. But now we do. Already into it at the start of paragraph 2.

Now that Quinn (we got drunk at a conference together once so we're first namers) has the reader interested and compelled, he doesn't let up. The center of the article is the best part to get into the technical science bits that I put ahead of the story in my virus article.

This is great structure. Readers are hooked early with a compelling story, then in the center of the piece, they digest the heart of the issue, the science that you're trying to communicate.

Here's more of the central part of the article. Interviews, historical background, more shading in. Excellent.

The article ends with a summation, and a "where do we go from here?" Think of it like you're sending the reader off into the world, armed with their new knowledge. "Here's what might happen in the future. Good luck out there."

This is a generic structure of a pop-sci article, or really any article, essay, publication, book, screed etc.

A lot of people after graduating high school railed against basic 5-paragraph structure of an essay they were taught. Well I have terrible news for you: your teachers were right.

1. Intro
2. Three paragraphs of discussion.
3. Conclusion.

That's really good structure, distilled to an easy-to-remember skeleton.

*takes drag on cigarette* skeletons are hard to forget.

I think a big source of difficulty here, at least for me when I write, is trying to do way too much. Trying to get every opinion, every turn-of-phrase, every novelty I think I have stuffed into as little space as possible.

And that starts to break things. Weighs down your writing. Cut the dead weight -- irrelevant opinions, asides, etc. Put the story first, literally. Put the *relevant* facts and opinions second, and then conclude. Just like you in-this-essay-I-will-argued in high school.

Here's an updated skeleton:

1. The grabbiest information. The stuff that made you interested in the story in the first place.

2. All the explanatory stuff. All the things that contextualize and explain the juicy bit at the top.

3. A summation. What does the future look like?