AN21 Panelists Offer Publication Advice to Prospective Book Authors

Writing and publishing one’s own book may seem like a daunting task, especially for first-time authors. However, certain strategies and approaches can simplify and facilitate the process. During the 2021 SIAM Annual Meeting, which took place virtually in July, a panel of SIAM book authors engaged in a lively discussion about the nuances of book publication. The panel—which was chaired by Elizabeth Greenspan of SIAM’s Book Acquisitions division—included Emily Evans¹ (Brigham Young University), Nicholas Higham² (University of Manchester), Ralph Smith³ (North Carolina State University), and Paula Callaghan (SIAM). All of the panelists reflected on their respective techniques, offered guidance for anyone who is thinking about writing a book, and spoke about the advantages of publishing with SIAM.

The panelists began by urging potential authors to think deeply about the motivations and desires that inspire their prospective projects. For example, one might wish to address new teaching methods, novel approaches to existing research topics, or unique software developments. Higham encouraged attendees to consider whether they would use their imagined books themselves. “I’ve written four SIAM books and I use them all quite regularly,” he said, adding that authors who do not plan on consulting their own books might not have found the right premise or angle.

Conversation quickly moved to the topic of book proposals, which reflect a book’s aim and scope and identify the way in which it differs from existing literature. Some authors choose to submit proposals early in the writing process, but Smith opts to wait until he has drafted roughly 75 percent of the text. “It takes me until that point to see most of the final picture of the book,” he said. Greenspan noted that the amount of manuscript that authors submit to SIAM varies, ranging from a table of contents and preface to a complete draft. More senior individuals who have long records of publication might submit only the former, but junior researchers should submit at least one chapter. “We do get a lot of proposals where the book is half done or even sometimes more,” Greenspan said. “Sometimes people just don’t want the pressure of signing a contract without having written most of the book.”

Before submitting a proposal, Smith recommended that textbook writers teach a course with a draft of their textbook to gauge student reactions and ensure that the content meets its targeted audience — he even allows students to edit and offer suggestions about the material. In fact, virtually all SIAM textbooks begin as lecture notes. When translating course notes into book format, authors must keep the knowledge level of their intended audience in mind.

Like Smith, Evans touted the value of teaching-based feedback. She gave her drafted textbook to a colleague, asked him to teach from it, and solicited candid comments based on his experience. She also had additional coworkers and graduate students review the text, incentivizing the latter with extra credit for each mistake that they found. “Students are much more likely to express their opinions if they’re getting something out of it,” Evans said.

While authors often ask their students or colleagues to look over proposal materials, a fundamental part of the proposal process for both textbooks and monographs is the outside reviews that the publisher gathers after proposal receipt. Unlike commentary from students or colleagues, these reviews—as with journal articles—are anonymous and come from individuals who are experts in the field but have no personal involvement in the project at hand. This feedback can also jumpstart the writing process for authors who are suffering from writer’s block.

As such, employing multiple reviewers is beneficial. “Send as many people through the book as possible,” Smith said. “It’s amazing how different people look at it with different lenses.” He noted that colleagues might have ideas for better approaches and students are particularly quick to identify concepts or explanations that they do not understand. However, he warned that the review process can be tedious when one reaches out to multiple critics. For instance, he once spent an entire month revising one section based on a single comment from a SIAM reviewer.

Evans remarked that in some cases, reviewers should do more than simply read the text. “When you’re publishing a book, you need to have someone—and it doesn’t have to be the same person—do every single one of the exercises,” she said. This guideline becomes even more important in the context of programming, as a third party should run and test all software.

Discussion then turned to the perks and drawbacks of working with coauthors. Higham, who has written books by himself and with other people, commented that each path has its own benefits. Some writers choose to recruit coauthors because they cannot complete the prospective book themselves and need additional expertise. For sizable projects, sharing the work may mean that the final product publishes in a more timely manner. Nevertheless, Higham suggested that coauthors first write a journal paper together to ensure that their work styles are compatible. “Do think very carefully before you start a coauthor project,” he said. “The dangers of coauthorship are that it might slow things down.”

Evans, who wrote Foundations of Applied Mathematics, Volume 1: Mathematical Analysis with two coauthors, conceded the risks but offered a different perspective. She found that having coauthors motivated her when she did not feel like writing because it held her accountable to other people. Collaboration also provides an opportunity for writers to enlist individuals with unique capabilities, such as pedagogical proficiency. “If you’re picking coauthors, you need to pick people who have strengths or expertise that very much complement your own,” Evans said.

Regardless of the number of authors, the panelists agreed that writing a book always takes longer than expected. “In all three cases I had an initial schedule, and in all three cases I didn’t quite finish when I said I was going to,” Smith said. “One of the things I could caution all authors is that the last 15 to 20 percent of the book seems to take 40 percent of the time.” If writers are testing their material by teaching it in a course, they can link their writing schedules with their semester syllabi for pacing purposes. Smith also advised attendees to allow sufficient time for the incorporation of feedback.

Though authors should of course attempt to meet their deadlines, Higham indicated that many publishers would rather have writers submit their books a few months late than compromise the quality by rushing to finish. “Don’t agree to a deadline unless you’re really happy with it,” he said. “You might need more time to put the book aside.” Greenspan noted that people who submit proposals and contracts in a book’s early stages are more likely to miss deadlines, while those who deliver proposals for nearly-written books tend to finish much closer to the agreed-upon target.

All of the panelists conceded that preparing the index is one of the most time-consuming parts of the process because indices take a long time to both create and debug. Higham urged writers to refrain from saving the index until the end, as working on it several months before the book’s completion allows one to catch LaTeX errors and other inconsistencies. Smith concurred and recommended that authors allot specific blocks of time exclusively for indices, which are important tools for readers. “For a general reader, the index is really important,” he said. “Please take the index very seriously.”

After successfully publishing a book, authors might eventually wonder whether they should consider a second (or third, or fourth) edition. SIAM recommends that at least 20 to 25 percent of material in a subsequent edition be new and different from the existing text. This material could stem from a dramatic breakthrough, a novel technique, or a new software. In some cases, future editions of certain books—such as those that address software-oriented topics like MATLAB—are guided by changes in the software itself. “There clearly has to be something different that makes the new edition warranted,” Higham said. “In particular, why would people want to buy the new edition rather than the old edition?” For example, one need not immediately consider a second edition if the field in question is not rapidly changing.

Higham keeps files for all his books and makes annotations about items that he spots after publication—including typos, references, and other updates—that he would like to incorporate or improve upon in future editions. This system allows him to maintain organized notes about his thought processes and helps him prepare for any subsequent publications. In some cases, authors can make minor typo changes and other corrections if/when a book enters a second printing cycle. “SIAM is very good at working with you on that, and that might not universally be the case,” Smith said.

As the panel wrapped up, panelists reflected on their publishing experiences with SIAM. Higham praised SIAM copyeditors who correct grammar, spelling, references, quotations, and sentence structure, and added that he cannot imagine publishing with an organization that does not provide copyediting services. “Copyeditors are extremely important,” he said. “They really do improve my best efforts. The final result is a book that is far better from what I delivered.”

Higham concluded with several remarks about the consistency and flexibility of SIAM publishing. He noted that most authors work with the same staff members throughout the entire publication process, from proposal submission to final edits (sometimes a few years later). “You get to know people and they’ll take you through the whole process, beginning to end,” he said. “If you have a reasonable request about your book, SIAM will always listen and accede to your request if possible.”

¹Co-author of Foundations of Applied Mathematics, Volume 1: Mathematical Analysis

²Author/co-author of Accuracy and Stability of Numerical Algorithms (2nd edition), Functions of Matrices: Theory and Computation, MATLAB Guide (3rd edition), and Handbook of Writing for the Mathematical Sciences (3rd edition)

³Author of Smart Material Systems: Model Development and Uncertainty Quantification: Theory, Implementation, and Applications