By Martin Wall, STC Carolina Member

MW: What kind of unique insights does your background in Computational Statistics provide to your writing of manuals and instructional books?

RW: In both computers and in statistics, there is an emphasis on precision and efficiency.  There is also an emphasis on getting rid of extraneous details and getting down to what really matters.  Those characteristics extend to my writing and have a strong influence on it.  As Einstein once said, “Everything should be made as simple as possible, but not simpler.”  My goal is to distill my thoughts down to their essence, while keeping them sufficiently complex to convey the topic.


MW: As an author of both long manuals and short blog articles, can you describe the differences (other than the length) in your writing method between the two?  What do you find advantageous and challenging about both mediums?

RW: As any good writer will tell you, it’s easy to write, but it’s hard to rewrite.  The editing process for the creation of a short article, such as a blog post, often takes longer than the initial draft.  It is a challenge to write a short article about a complex topic.  Because of those space limitations, I will often go back and edit the drafts of my blog posts multiple times.

For a longer document, such as a book, the focus is on the completeness of the topic.  I provide more detail and am less concerned about brevity when I write books.

One of the best aspects of writing on social media, such as in a blog post, is that it invites discussion from the reader.  A reader might post a comment that says:  “Tell me a little more about this item that you mentioned,” which will lead to a follow-up article.

MW: Several of your articles deal with the concept of making sense of charts and graphs. What do you find are the essential components for explaining how to use and interpret those tools correctly?  While you obviously use visuals to demonstrate, how do you tailor your writing to cater to those visuals?

RW: A picture can be worth a thousand words.  Graphs are a concise way to communicate a lot of information.  When I write about a graph, I focus on two things.  The first is how to construct the graph by using a computer.  The second is how to convey information through graphs?

MW: You also use a number of examples, such as the example in your article regarding Histograms to illustrate your point.  How do you formulate the language used that supports the examples?

RW: When I write or speak, I believe that a well-crafted example is often the best way to communicate the point that you’re trying to make.  Sometimes I start with the general topic, then present a specific example.  At other times, I will use an example as a “scaffold” that builds towards the general concept.  It can flow both ways, depending on the topic, format and audience.  If I’m writing in a blog post, I’m more likely to start with an example.  If I’m writing documentation in a manual or book, I’m more likely to start with the general concept and then show examples.


MW: Your biography at SAS says that you’ve been using SAS since 1997.  In the past twenty years, how has the process of providing instructions regarding SAS programs evolved? 

RW: When I first came to SAS, the entire series of SAS documentation manuals would fit on half of a bookshelf.  It consisted of about fifteen books and I still own some of them.  Over the last twenty years, the SAS language has grown dramatically. Documentation now includes many more graphs today; it was difficult and expensive to produce color graphics back in 1997.

Our documentation today is freed from the bonds of being in a paper volume and the writers are freer to make longer comments, include graphical outputs, etc.  Now that the documentation is on the web, it doesn’t matter, in terms of access, whether the document is 500 or 1000 pages long.  Twenty years ago, each topic in a manual would usually use one or two examples that were the best overall demonstrations of that topic. Today, the same topic might include a dozen examples  Having additional examples enables us to lead incorporate examples that cater to specific niche applications.

MW: What advice would you give to aspiring technical writers who are also SAS programmers regarding how they should be creating articles or manuals explaining how to use SAS?

RW: Know your audience!  Choose a target audience and write to the needs and experience of that audience.  My target audience is typically people who are statistically sophisticated and have a moderate level of experience with SAS.  A second piece of advice is to write the article that people want to read, not the one that you want to write.  It’s easy for a writer to write for themselves, but we are at our best when we write for the reader!  What does the audience need to know?  What content are you going to provide that will entice the audience to read the entire article?