In my August newsletter, I shared my angst about whether what I teach in my technical writing classes is really informed by research or is simply based on instinct. My headline read: “How do I know what I know?” (My eight-year-old grandson saw that headline and said, “You just know what you know, Grandma! At least, I do.” Oh, to be always as self-assured as an eight-year-old!)
Fortunately, I’m no longer in angst because I have just read Letting Go of the Words, Second Edition by Ginny Redish, a fellow Fellow of STC. I had not read the first edition because the book’s subtitle is “Writing Web Content that Works,” and I don’t write Web content for a living nor do I usually teach people who write Web content. However, my editor suggested reading Redish’s material on personas, and I got hooked on the whole book.
Ginny’s book confirmed that most of my cherished language beliefs do have a solid research foundation, at least for web readers. I recognize that there are differences between Web and print-based reading, but I believe that most of the web principles that Ginny shares will generalize to the printed page as well (this may be an intellectual leap that we could argue).
Here are three principles from Ginny’s book that I also love to teach:
- Using color is problematic for at least two reasons: Some readers have color deficiencies, and colors carry cultural connotations that you might not be able to predict (pp. 50-52).
- Don’t center text, even in table columns, because centering violates the reader’s need for alignment (pp. 59-60).
- Readers seeking information look first at your first paragraph and then at the first words of list items (p. 139), so put the “bottom line on top” (i.e., lead with the most important information).
Although I do agree with most of Ginny’s points and practices, a few left me wondering:
- She argues in favor of “turning sentences into more visual forms, like lists and tables” (p. 62), and I concur; but I wonder why she lowercases the first letter of each listed point. This practice contradicts what I see in almost all lists in technical writing today (again, I don’t know if any research argues one way or the other).
- She recommends sans serif typefaces for both Web and print-based text (p. 62). I’ve always taught that sans serif is best for Web writing, email, and slides, but serif is best for print-based text. I was fascinated to read that the research that claims that serif typefaces are better for sustained reading (print-based text) is now more than 60 years old. Ginny claims that “research on Web content has not shown a consistent winner between serif and sans serif for either reading speed or comprehension.” However, her assertion that “preference almost always favors the familiar sans serif fonts [for the Web]” is not as definitive as I would like.
Despite these minor concerns, I am glad that I have studied Ginny’s book―it is packed with excellent tips and persuasive case studies and makeovers. In addition, I was reinforced in three more principles:
1. Before writing anything for any audience, always imagine a conversation with the reader:
- What do they think?
- What do they know?
- What do they need to know?
- What do they want to do with the information that I might provide to them?
I don’t need research to tell me that considering my audience’s wants and needs is valid―it’s just compassionate common sense.
2. Rethink my assumptions—I had made an assumption that I didn’t need to read Ginny’s book because it was framed to help Web writers. It turns out that most of her principles apply to the kind of writing that I do teach, and I wish I had read it earlier.
3. Stay close to STC publications and colleagues—their work is brilliant and can inform my practice and reduce my angst.
Elizabeth (Bette) Frick, PhD, ELS is president of The Text Doctor LLC in Boulder, CO. She teaches technical writing in corporations and edits medical documents. Bette is an STC Fellow and has been independent for 23 years.