I gave a talk this week on things I have learned re: scientific writing. The talk was in French (my language at work) but I thought I’d share here (slides in French, tweets in English) in case it is useful to others.
Thread:
Thread:
As I presented to those at my talk, what worked for me may or may not work for you, but for whatever they are worth, here are my best tips.
1. Be honest with yourself about your issues.
1. Be honest with yourself about your issues.
At one time, I found it easy to say that I wasn’t writing because I was busy. I was definitely busy.
But the real barrier was that I was scared. (Did you know that if you never publish your work, no one will ever learn it has flaws?? It’s true! 😂)
Once I admitted to myself that it was really fear & intimidation—not time pressures—blocking me, I started writing more and more.
Once I admitted to myself that it was really fear & intimidation—not time pressures—blocking me, I started writing more and more.
2. Put on your own oxygen mask before assisting others. I really like helping people. I like the immediate, positive feedback of doing something that makes a colleague/student happy. But if I don’t put on my oxygen mask, I am no longer of use to anyone.
3. Decide what you have to say before you start writing, choose a preferred journal and read the Instructions to Authors.
It’s much easier to get somewhere when you know exactly where you’re going.
Getting lost or turned around while writing is not fun.
4. Start in the middle. Exact order varies, but my usual order of writing is: tables/figures, methods, results, discussion, intro, abstract, cover letter.
5. You don’t have to (and probably can’t) write the whole manuscript in one big flood of words. You can make a surprising amount of progress in 10-20 minutes a day. I keep a To Do list at the top of a manuscript in progress, and a code at my stopping point so I can find my spot.
6. Make use of useful tools/approaches. There are many, and they all have pros and cons.
Find a system that works for you that has:
a) reference management software
b) an outlining method
c) a versioning system
d) a way to keep all the pieces/files together
e) a way to collaborate with co-authors that doesn’t make you or anyone else frustrated
a) reference management software
b) an outlining method
c) a versioning system
d) a way to keep all the pieces/files together
e) a way to collaborate with co-authors that doesn’t make you or anyone else frustrated
I personally use:
a) Paperpile
b) declarative topic sentences [1]
c) project_doctype_otherdetails_yyyy-mm-dd (or no date but named versions in google docs)
d) folders (or hyperlinks in google docs)
e) google docs [2]
a) Paperpile
b) declarative topic sentences [1]
c) project_doctype_otherdetails_yyyy-mm-dd (or no date but named versions in google docs)
d) folders (or hyperlinks in google docs)
e) google docs [2]
[1] Outlining with declarative topic sentences for every paragraph means we nail down the ideas *first*, and *then* we work on the rest of the writing. This is a major efficiency when working with trainees for whom English is not their first language (that’s 100% of my trainees.)
[2] I really hate the multi-branched change-tracked Word documents. Google docs is not perfect but it’s very accessible and works reasonably well for big teams. Everyone edits the most recent version. We’re all working together.
Google docs is why I use Paperpile—it plays nicest with google docs of any I’ve tried. Zotero also now plays with google docs but last time I checked, it lacked a major feature of Paperpile: the ability to copy & paste from one document to another and have the refs come with.
7. Think about your reviewers. Make their experience as pleasant as possible. Give them something that is well-organized, clear, and, please please please, free of abbreviations and acronyms.
You don’t save that many words by inventing twelve new acronyms, and it costs you a lot in readability and reviewer goodwill.
Again, a well-organized outline of *declarative topic sentences* will help your reviewer follow your writing.
8. Make use of open science methods that may be new to you & can significantly increase the value of your paper. (I suggested: preprints, our institutional repository for accepted manuscripts, our institutional data repository, R markdown. There are others!)
9. If you’re going to write in English, know that it is not just a simple translation of how you’d write it in French. Here’s a link to an open document in which I keep a running list of common errors I see among francophones writing in English.
10. If you want to write well, read. At minimum, read your own writing aloud.
Bonus: know that it’s hard, but you are capable. Sometimes the only way out is through.
Or, if you prefer, I offer my family’s motto.
Anyone: “What’s your family motto?”
My kids: “Suck it up, buttercup.”
Anyone: “What’s your family motto?”
My kids: “Suck it up, buttercup.”
Wishing you good writing.
