8 Things We Learned About Making Open Textbooks from Making Media Innovation & Entrepreneurship

When Rebus Community agreed to support the JMC Media Innovation and Entrepreneurship open textbook back in January of 2017, we had no idea the impact the project would have on our knowledge of the processes involved in open publishing.

Michelle Ferrier of Ohio University and Liz Mays, who is Rebus’ marketing manager and also an adjunct faculty at Arizona State University, had the idea for the open textbook.

For Rebus, the thought was one of us would personally go through the process of creating an open textbook start to finish, with Rebus’ support, and we’d use what we learned to inform Rebus Community resources, processes and software in development.

Ten months later, V1 of Media Innovation & Entrepreneurship is available, and it’s time to share some of those learnings with the broader OER community.

Why Media Innovation & Entrepreneurship Was Needed

But first, some context. In recent years, faculty of media, journalism and communications have begun to teach concepts of entrepreneurship in their classrooms, in response to the disruption the journalism industry has experienced. In some departments, entrepreneurship is a full course (or series of courses); in others, it’s part of an existing course.

As faculty who had been asked to teach entrepreneurial journalism and media innovation concepts in our classrooms, Michelle and I, along with others we knew, felt there was no one seminal text that covered the key concepts of media innovation and journalism entrepreneurship that media, communication or journalism students need, especially as the development of content and technology businesses is very different than traditional brick-and-mortar entrepreneurship.

When we embarked on this project, it was our aim to build the resource we ourselves needed, and make it freely available for use to others teaching media innovation and entrepreneurial journalism courses across the globe.

Michelle and I came to the project with two somewhat different drafts for a table of contents. First, we merged our TOCs into a new skeleton. Then, we held a webinar with others who teach these subjects to help us add to, refine, whittle down and revise that into a working, public Table of Contents, which was posted with the Rebus Community Forum call for contributors, along with an Author Guide and Precis. (Note: We stopped maintaining the TOC and moved to other working spreadsheets after beta launch.)

For each chapter, we identified a number of qualified and diverse potential contributors in our networks, and by around April, we were reaching out one by one. Here’s a sample of the baseline verbiage we used to reach out. Some chapters were easier to find authors for than others. (In fact, there are few sidebars and chapters that are still being written to add to the 2nd edition.) While some people didn’t have the capacity to participate, or weren’t interested in helping, by and large, author recruitment went smoother than I had expected. Which brings us to lesson one.

Lesson One: Don’t just build a textbook; build a community.

Throughout every phase of this book–TOC, author reachouts, peer review, beta testing and more–we leveraged community input and contributions around the book, and those efforts paid off. This project could not have happened as quickly had Michelle not been as well-connected in the discipline as she is, and able to bring a good many of her peers and colleagues to the project. At Rebus, we’ve seen the value of faculty leveraging their professional networks. Their involvement helps to drive participation.

I came to the project to build a textbook. Michelle came to the project to build a community. It turns out we could never have built the former without the latter. At the end of the day, 20 authors, 10+ peer reviewers, 12 official faculty beta testers, classrooms full of students–at multiple institutions across the U.S. and as far away as Chile, Greece, and Ukraine–were involved in the making of this book. Nearly 40 others have been following the updates, with an interest in adopting parts or all the book.

Contributors said they found the community aspect of the process rewarding.

“Many thanks…for the gift to join you in such a fascinating intellectual and collaborative journey for making the Open Textbook on “Media Innovation & Entrepreneurship” a dynamic learning, writing, and mentoring experiment,” wrote contributing author Betty Tsakarestou. “Along with the open textbook and the upcoming instructors’ manual, I think you are also helping to create a very dynamic network and community of practice.”

Lesson Two: Building an open textbook is going to take a lot longer than you think.

Once we assigned sidebars and chapters to authors, that’s when we learned the next major lesson. How long does it take to build a community-created open textbook?

In hindsight, we suggest people to allow more than a year, and that’s if you plan to be aggressive about staying on timeline.

One thing that lengthened our editing process was that we didn’t anticipate the questions authors would have about how their chapter ought to be different for an “open” textbook than a traditional textbook. We had never thought of them as any different, but realized we could have made clearer suggestions for the voice, which we wanted to be accessible to students rather than in a formal style. Quite a few authors sent us substantive outlines to edit to make sure they were on track before their full chapter, which added another round of editing that more clarity up front on our part could have eliminated. Also, in a transparent process, I learned it’s a good idea to give all contributors roughly the same deadline, assuming you’ve allowed ample time for editing and revisions.

After a frantic summer of editing and formatting, we found ourselves barreling toward peer review (which went quite smoothly) and then getting a “minimum viable product” up by Aug. 1 for the faculty who would beta-test it in fall semester. All the while we were scrambling behind the scenes to live-edit that version for when students would see it, mid-August.

We ended up in a similar scramble to implement the tail end of revisions from the extensive feedback we received in the book’s first semester of use, just in time to enter the book into the prestigious AEJMC Tankard Award competition, which required that the book be “available in print” in 2017. For teachers wanting their students to be able to get a print copy (that they didn’t print off at home), the timeline wasn’t much later.

If you want to produce an open textbook in a leisurely way, consider spreading it over a year and a half.

In hindsight, here’s how I would adjust to make a new open textbook happen on a roughly yearlong academic timeline. (I’ll note though that every open textbook has different nuances that may add or remove time from your schedule.)

Our Timeline	Ideal Timeline
Jan-April: planning	Nov: planning
May-June: writing	Dec-Jan: writing
June-July: editing, peer review, formatting	Feb-May: editing, formatting
Aug-Oct: proofreading, beta testing, open review	June-July: peer review, make available for beta testers to consider, proofreading/fine-tuning, accessibility
Nov-Dec: POD first edition	Aug-Nov: beta testing, open review
Nov-Dec-Jan: 2nd edition additions	Dec-mid-Jan: POD first edition (most of this is waiting after you upload)

Lesson Three: Find collaborators who have different skills than you.

One of the startup culture concepts j-entrepreneurship faculty stress in their classrooms is the need to build a team around you with skills that are different from, but complement, your own.

Prior to Rebus, I had been a professional editor for 10 years full time and for another 7 years on the side. Michelle too, with her background in journalism, brought this expertise. But our editing skills were different and complementary, which enabled us to tag-team and divide the work fairly, without stepping on each others’ toes.

With a national profile from researching and practicing the teaching of media entrepreneurship and innovation in higher ed, Michelle served as subject matter editor and line editor–she did the intense “heavy lifting” types of editing and rewriting. I wore the hat of managing editor and proofreader, making sure project deliverables came in and were tracked; proofreading the book like my proofreaders UK mentors had taught me; and dealing with image logistics and footnotes. We also had help with additional copy edits on four chapters from some great volunteers–in retrospect, we wish we had put out a call more widely for volunteers who could assist with the editing. Supporting the process of finding collaborators for editing, images and similar non-writing, non-reviewing components is something Rebus will be working to expand in the future.

Lesson Four: Modularity for OER is a “capital D” decision.

Early in the project, we had to make some decisions, decisions we didn’t recognize as Big D decisions, and probably should have.

First, it never occurred to us to build the book in any way other than a modular way. This was because media innovation and entrepreneurship are taught in vastly different ways at different institutions. In some cases, there is a full class in journalism entrepreneurship, media innovation, or business of journalism. In others, these concepts are part of a media management, magazine writing, student media, or other reporting course. In some programs, students might be from the broader fields of media, communications or P/R, rather than just journalism. Also, at Rebus we care a lot about the remixability of open textbooks, and a modular format works really well for that. It means future faculty can take parts of the book or most of the book as a baseline and write just the pieces they need to accommodate the specific way they teach their class.

Knowing this, we designed the book modularly. And we stand by that decision. Some initial users were quite enthused by it.

“I really get why it makes sense to be a modular “book” because we could rearrange our curriculum and almost build courses around it. Like we could focus on 4-5 chapters per course, depending on how it makes sense to break up the coursework,” wrote Mark Poepsel to Michelle and I, who has been using portions of the book. “Then courses would be built to the professor’s expertise or they’d be designed around resource availability. You’re making the IKEA of information.”

However, we got a lot of feedback in beta testing about the modularity of the book from those who were using it straight through–some things are repeated because we’re cognizant that students reading only an isolated chapter of the book will still need background on a concept that might also be introduced somewhere else earlier. While we couldn’t fix that repetition easily, it’s something we would like to smooth out for the second edition.

Lesson Five: Pressbooks — did I mention we broke Pressbooks? (Don’t worry, we fixed it too.)

Most people who know the folks at Rebus know that some of us–Apurva Ashok, Zoe Wake Hyde, Hugh McGuire, and I–also work for Pressbooks. So here I’m switching hats. This was the first educational book we ourselves had tried to format, to see what problems we ran into.

Some of those included:

• Because the book is produced from a single source, and will be available in multiple formats, including printable PDF and print-on-demand, anything that might otherwise have been a link from the digital formats also needed to be a footnote, because embedded links wouldn’t be clickable or visible in the print edition.
• How to handle multiple authors in the metadata (we’re working on a new way to handle multiple contributors in the metadata of a book)
• Wide charts and tables (temporary fix in place on this book; long-term fix in process)
• Import from Google docs, including footnote and link import (we learned a few things)
• Easier glossary making and figure numbering (on the Pressbooks roadmap)
• Editing (I recommend doing as much of the editing process as possible before importing to Pressbooks as it does not have tracked changes or similar functions that copy editors rely on at earlier stages of editing.
• Images (Apparently, in addition to throwing errors if you send too low-res of images, POD will actually throw errors or cause delays if your images are too high-res. For large images displayed in a small area (such as author images), this became a problem.

These are problems that lots of educational books would run into. This book turned into an opportunity to use our own product and find the flaws. We were able to put these issues into Pressbooks’ queue to improve the software not only for us, but for all the others using it for open textbooks and OER. Changing back into my Rebus hat…

Lesson Six: Someone (maybe us) should build a mechanism to track feedback from beta testing

We put the book through a formal peer review process, with most chapters getting their own single-blind peer review prior to the beta launch. The book was then formally beta tested in classrooms. We gave beta testers a guide and as many ways to give their feedback as possible. We enabled hypothes.is on the book. We asked for narrative feedback by email. We opened a “suggestion box” Google form and offered a place to suggest revisions on the Rebus Community forum. One student even reviewed the book as his honors project. (Read Kyle Kercheval’s review.) We then held biweekly calls with the community of practice teaching these subjects and using parts of the book. Through this process, we received hundreds of points of feedback. (And we still have three peer reviewers still working their way through the full book!) I can tell you from that process that there is no magic way to track and incorporate beta feedback–yet. Just a lot of heavy lifting.

We learned that having too many avenues isn’t practical but no one of these avenues was sufficient. This is another problem Rebus would like to help solve in the future.

Lesson Seven: Print-on-demand is still not instant.

Having a print book comes with many benefits, especially for students who prefer non-digital formats, but preparing the print format and ensuring its availability can be tricky!

No. 1: If you just want students to be able to purchase a print copy of the book in Amazon if they want it (rather than a university bookstore ordering a bunch of copies), you should know Ingram recommends you allow 6-8 weeks before you need them to be able to do so. (Createspace doesn’t allow CC BY content, so we went through IngramSpark for print-only distribution.) That’s two months of waiting added to the time it takes to bring an open textbook into the world. That meant for us, with a contest entry that required the book to be available in print by a hard deadline of Dec. 31, we had to declare a first edition a few weeks earlier than we would have liked.

No. 2: Some POD platforms only allow you to cite up to three contributors in the metadata (even though Amazon allows up to nine). This is going to be a challenge for the burgeoning cadre of open textbooks with multiple contributors. Also, we had issues with the metadata we used transmitting incorrectly, a problem we’re still working through on both sides. It seems neither Ingram nor Amazon can fix an Amazon entry that didn’t originate on the Amazon side.

Lesson Eight: Publishing versus authoring

The question in open textbooks, and the problem we’re trying to solve at Rebus, is how to help authors do, or find other volunteers to do, the things that publishers have traditionally had a staff to do, in the absence of a traditional publisher.

Among these are permissions, accessibility and other “invisible” labor.

We believe strongly that all those things still need to happen for open textbooks, especially those produced with support from Rebus. Editing, images, Pressbooks formatting, and accessibility are four areas where we want to specifically build up pools of potential volunteers interested in contributing their skill sets in these areas to open textbooks.

When we embarked on this project, we weren’t sure what accessibility for open textbooks really entailed. We have a much better idea of that now, after starting the Rebus Community Accessibility Working Group, but we didn’t have that information in time to tell the writers what to send us, so that we wouldn’t end up writing alt text for most of the images ourselves, for example.

At Rebus we have since been working on some guidelines for accessibility for open textbooks produced with Rebus support, and these will be part of the author guides for future projects.

Styling footnotes, from both an editing and a formatting perspective, was a monumental effort, and one that I can say with absolute confidence needs its own volunteer or volunteers.

Another of the inherent challenges to creating a CC BY-licensed open textbook is images. Our goal was to use all CC BY images. Some writers found that limiting (since traditional publishers help to source permissions), or did not quite understand the nuances (for instance, buying stock photos doesn’t skirt this problem).

We cut any images that weren’t necessary. Of the rest, some image suggestions were general enough for us to create new openly licensed images. I had the benefit of support for this from Rebus staff member Apurva, who with her design background was able to help create some illustrations for the book from scratch. (This is another job, by the way–one that needs a volunteer.)

In other cases, even if using an image would have been fair use, the fact we were setting the book up to be remixed downstream and would lose control of the content, made us err on the side of caution while we figure out all the nuances to this.

Some images were so central to the subject matter that we simply had to include them. How do you not include the Business Model Canvas in a chapter about startup business models, for instance? There’s no way to remake what’s important about that chart without exactly copying it. Thus, we had to hunt down permissions for some images.

Luckily, the Business Model Canvas folks, along with a few others, were awesome to us. They got the concept of what we were doing right away and gave permission to include their image in the textbook, and to them we are eternally grateful. There were others too who had to bend internal rules to work with our paradigm, but were ultimately able to grant us the permissions we needed.

Then there were the publishers that wanted to charge an exorbitant rate, which would increase with every format or new edition (exactly what does that look like for a book that is continually updated, and can be remixed?)

Still others were never able to grant us permission to use the simplest of things. These included one data chart that, according to the site’s TOS, could be embedded in our website with no problem but could not be published in something like an ebook or print book. With single-source formatting, this was unworkable. Our multiple calls and emails requesting a solution went unreturned.

In the end, the thing I thought would be hard about building an open textbook (getting contributors to write and review the content) was not the thing that was most challenging about building an open textbook. Our authors, reviewers, beta testers, and everyone else who played a role in refining the content for this book were awesome, and they built the content into the open textbook we envisioned.

What was hardest was getting these behind-the-scenes jobs, many of which take a lot of invisible emotional labor, done. We at Rebus will now be working to build up pools of volunteers to enlist in these efforts on future books. Of course, we will credit them with their efforts. In addition, the software we have in the works may help. If you have further ideas on this front, please reach out to us in the Rebus Community Forum.