The Publishing Experiment of Portals

Updated Jan 2018

My 2017 short book, Portals: A Treatise on Internet-Distributed Television, was a publishing experiment in a couple of ways. First, it was—more or less—self-published. Second, I released a digital, open access format as well as version for sale in paperback or ebook. Self-publishing and open access are two different things. My motivations and the lessons (so far) of the experiment need teased apart.

Why self-publish?

Portals is an odd project. It grew out of another project and didn’t fit conventional academic forms of publishing well (too long for a journal, on the short side for a book) and spanned fields to an extent that I wasn’t sure what journal would be a good fit. There are certainly titles distributed online that could have coped with its length, but the benefit of publishing in these outlets seemed limited.

The book is about a state of the moment topic that had little existing theorization and was rapidly evolving. Anything I wrote would have been problematically dated by the year duration—at minimum—required to work through a conventional book publisher. (2018: There's not a ton different. Seeso is gone, but it was just an example). Such “developing” contexts are usually first fleshed out with case studies. Portals offers general frameworks and lines of analyses that I believed could be helpful for those working on such case examinations. What I had to offer was meant to be provocative—to create a starting point for conversations of either a foundation or a site of disagreement.

I chose the self-publishing route because it matched the needs of the project and, let’s be clear, because I am a full professor at a university that doesn’t have any type of RAE scheme that will necessarily afford this book less value because I self-published. I was not certain how, and if, this book would be regarded in terms of my “career” or issues like merit pay increases. I believed my existing publishing record made Portals likely to have little negative implications for my career. Though not without risk—the risks seemed minimal.

Why open access?

Mostly curiosity. I had been studying the consequences of digital distribution and the various revenue models and business strategies it allows for video for more than a decade. But I don’t make videos, I write books. This seemed a chance to play with some of the ideas (more accessible=more readers; freemium=more buyers) that I had been working through in the realm of television.

Also, I had published (as either author, co-author, or editor) six books (two with second editions) with five different publishers and there didn’t seem much value in repeating an experience that I found increasingly unrewarding. Among other frustrations, I know very little about who buys my books. I have a sense of pace of sale based on annual royalty statements. Oxford told us the first edition of Understanding Media Industries sold most of its copies in Canada, and we only learned that because we badgered our editor as we tried to understand our readership in preparing the second edition. But I don’t know more than sales of different formats (paper, hard, ebook) in either the U.S or the “rest of the world.” Open access would allow somewhat more refined information (big data!).

I created the purchasable version because I wanted to see if anyone would choose this and if so, how many, given the read-only open access version existed. I assumed few would buy it, but I was curious. The decision to employ a degree of DRM with a read-only version—rather than an easily downloadable PDF—was part of the experiment as well.


I started talking about self-publishing in spring 2016—testing out whether this seemed a terrible idea (no one thought it terrible, but most were skeptical. I was told a couple times that I’m not Henry Jenkins, a fact I was aware). I skimmed a book on self-publishing. I looked around the field for other cases and exchanged some emails with David Bordwell who has developed a quite advanced enterprise. My colleague Dan Herbert mentioned that another colleague, Markus Nornes had published a book unconventionally. Markus told me that University of Michigan Press has a separate arm for such ventures called Maize Publishing. I met with Maize Publishing and my project seemed a good fit.

Maize Publishing is distinct from the University of Michigan Press. Portals did not go through a conventional UM Press process. I needed letters from two UM faculty supporting the project—basically outlining how it was more than just a vanity endeavor. Dan and Aswin Punathambekar generously read a proposal and early draft and provided these letters. I also sent an early draft to colleagues familiar with the terrain of internet-distributed television and asked for feedback. The lack of blinding unquestionably limited the feedback (and those I sought—because they are particularly expert—are junior colleagues likely aware that I may someday be asked to write tenure letters). I don’t know to what extent punches were pulled, but the feedback was very helpful for a substantive final revision.

Once I finished the manuscript, it was sent to a subcontracted typesetting firm (Apex, I highly recommend--actually, a lot of presses subcontract to Apex). I submitted the manuscript October 27. Copyediting was minor and came back November 15. I returned the copyedited manuscript November 22, and received the first proofs December 3, reviewed them and constructed an index. I received the second, final proof January 3 and the book was available open access and on Amazon three weeks later. So, about three months elapsed between completing the manuscript and its release.

During those months I sent the unedited manuscript to colleagues who accepted my request to offer a promotional endorsement for the cover. I worked with local photographer (Rob Gingerich-Jones) to come up with the image for the cover (to keep costs low, we went with a stock cover design that allowed some personalization with an original image). I retained the copyright; the file is archived in HathiTrust; the online edition will be distributed via a Creative Commons Attribution Non-Commercial No Derivatives (by-nc-nd)] license.

The Numbers

Beyond my time—which I really can’t calculate—Portals cost me $2000.

$1600—Typesetting—outsourced to Apex (copyediting, proofs, layout, ebook layout). Pricing is per page, so a lot of the low cost comes from Portals’ brevity.

$400 + 50% net sales—Maize Publishing—project management, print distribution, formatting of open access version

in kind debt Proposal review: Dan Herbert, Aswin Punathambekar

in kind debt Manuscript review: David Craig, Ethan Tussey

in kind debt Notes and conversation: David Hesmondhalgh, Ramon Lobato, Joshua Braun

gift Cover image: Rob Gingerich-Jones

Sales/hits to date

3 months after release: Portals had sold 25 books in the US, 35 books in the UK, and 24 ebooks globally. (I don’t have access to this data, but Maize Publishing provided at my request). 

The metrics for the open access site—which I can access—aren’t ideal. The book had just over 1,500 unique page views, but an average view time of 1 min 28 seconds. With no more detail on viewing time than average view time (no data about the individual hits that make up the average), it is difficult to discern how many—if any—were there long enough to read. The overwhelming evidence suggests that most of the hits are not reads.

1 year after release: Portals has sold 426 copies; 373 in the US and 53 in the UK. At least 31 copies have been sold to libraries.

The open access book has received 6,982 unique page views, average view time is still quite low. 


After three months, I’m most fascinated by this data. The metric of unique page views is pretty meaningless. Based on the average view time, few are reading this version. Rather this suggests that most click to the open access version to see what it is, and maybe look around a little to assess whether it is worth their time. I liken this to when I request a book from the library and flip through it before deciding to buy, which I do if I think I’ll read it in near entirety. I’m surprised that it has sold 84 copies (and more surprised by 426 in year one. For comparison Redesigning Women sold 477 in year one). The combination of the data from the open access hits and book sales suggests little cannibalization. 

So far, I’m pleased with the experience and the experiment. Notably, costs were low because it is a short book—a typical monograph would run closer to two to three times this price. I’m glad to have worked with Maize given this was my first foray into self-publishing and open access. Once I know more about the book’s readership, I’ll have a better sense of whether it is worth taking on the duties they performed myself. I’ll offer another post reflecting on the business of academic self-publishing and deal with this subject there.

If I do this again, I would build the reviewing costs into the project and pay reviewers more than a round of drinks next time we meet or in a review of their work—a debt I sincerely hope they ask me to repay.

It is also too early to have a sense of issues such as marketing and helping the book and audiences find each other. Twitter helped announce the book’s availability but doesn’t seem to have played a significant role. There is also a reference to the book and a link to its webpage in an article I wrote for The Conversation (that was redistributed by Quartz, Business Insider, Salon, and others) that looks to have generated considerably more traffic (which I can see in the metrics on my website). I was also doing a series of talks when the book released that helped establish some initial word of mouth. Notably, I already have much more of a sense of awareness about the book than when I publish with a conventional press.

Did I skip anything? Send me your questions.