Making Rhizcomics

Abstract

This webtext reflects on the methods used to create and rationale for my digital monograph, including a reflection on the book and my writing processes, a detailed timeline toward its publication, and recommendations for authors of similar projects. I argue that writers of digital scholarly monographs must pay special attention to the eventual form of their work at every stage, from writing a proposal to eventual publication.

Instructions

Use your keyboard arrows to navigate between slides, moving up, down, left, and right. Use your mouse track wheel to scroll down longer slides if they do not fit on your window.

Individual slides can be cited by referencing the anchor at the end of the URL (e.g., "#b/2"). Each anchor is a stable link.

This video offers a brief introduction to this webtext's purpose, organization, and navigation. Much of the same information is conveyed textually on the slide to the right. Readers are encouraged to use whichever mode they prefer to access this introduction, knowing that because form and content are imbricated, the repetition is not without difference.

Video introduction to this webtext. See next slide for transcript.

This webtext reflects on the various methods I used to create a digital monograph as well as the rationales behind those methods. It is divided into a series of sections and subsections. This introduction offers general reflection on the monograph. A lengthy and detailed timeline describes each step and when it happened. A practices section relates what I learned from that process and includes recommendations for future projects. Those recommendations are then distilled into a short list in the general recommendations section. The conclusion offers paths for future research.

I argue that writers of digital scholarly monographs must pay special attention to the eventual form of their work at every stage, from writing a proposal to eventual publication. The length of a digital monograph necessitates increased reflection on the scholarly modes it employs, offering new opportunities for the creation of scholarly digital tropes.

As you can already tell from the abstract and introduction, there is some repetition involved. This repetition encourages readers to browse and skim without depending too much on what has been discussed previously. It may also frustrate readers intent on reading it linearly. I'll offer some suggestions on nonlinear reading approaches below.

This web page can be navigated most easily using the keyboard arrows on your computer. Sections can be accessed by moving up and down and subsections (where available) by moving left and right. The site links to itself repeatedly and readers are encouraged to jump around.

This introduction includes two subsections that can be accessed by moving to the right: background and rationale.

There are two longer sections (timeline and practices) that include multiple horizontally scrolling subsections. I’d like to outline them briefly.

The timeline walks through each step of the process methodically, beginning with a discussion of pedagogy—how I learned to write, draw, and code for this project. Next, I outline the process of transforming the initial dissertation project into a book. Once I had a publisher, I needed to prepare them and myself for the work ahead. My composing process was substantially more recursive than my typical process. I continued tweaking arguments even while converting the project from a series of Word documents into a website. Once the initial web version was created, I worked with the press on a revision plan focusing on clarifying the argument and making the organization more patent. My final steps were to fix accessibility and make technical and copy edits.

The practices section breaks down the larger practices into smaller bites: designing a large multimodal project (publishing, medium decisions, constraints, accessibility, sustainability, and visuals), scripting for multimodal projects (nonlinear composing, drawing on scripts, and markup systems), making it digital (designing the interface, versioning, and workflows), and revising (testing and comment systems).

There are a few easy ways to approach this webtext. Casual readers could start on the general recommendations page. These recommendations are terse and direct, thus being more interesting than a list of every step found in the timeline. In addition, they will link back to slides from the practices and timeline sections. A reader more interested in creating their own digital project might read through the practices section first. The practices section links back to slides from the timeline section, and thus those can act as footnotes offering more context for the practices. A committed reader could read through the entire thing linearly and gain a great deal, as each section builds on the one before it. The best reader will explore based on their own needs.

All readers will benefit from finishing the introduction by proceeding through the background and rationale slides to the right before continuing to other sections.

It's difficult to say when the project started. There was a dissertation and some work that preceded it. In a sense, it's all connected. Regardless of where we draw the line, though, I had to learn specific skills before embarking on a multimodal project. This section is about how I learned the various skills required to create this project. However, I also include some reflections on how it has changed how I teach technology more broadly.

My interest in comics began before I could read and I didn't start learning Adobe Muse until 2013. Before I learned Muse, I learned how to draw, code, design, and digitally manipulate images.

Each of those major skills covers a plethora of minor skills, not all of which I learned. In other words, mastery is not necessarily important here. Instead, my learning was problem-based. I learned each of these skills to solve a particular problem I was having. When I introduce colleagues to my work, I often see how overwhelmed they feel. As though they (or more likely my students) are expected to learn all of these new skills. I try to emphasize instead that there is no comprehensive exam on multimodal scholarship that I'm preparing for. Each skill was created by a particular problem. Similarly, I no longer teach software by focusing on particular tools or tasks. I introduce problems to my students and try to help them generate solutions, often teaching software along the way.

I didn't need to learn every programming language or everything about the languages I used. The JavaScript manual alone is thousands of pages long. Instead, there were some basic skills I needed to learn in each area. But it's not always clear what the basic skills may be. Here I'll walk through just a few of them.

Video depicts the author learning to draw a live model in a figure drawing class. [No audio]

Early on in the dissertation my argument began to solidify: Comics as scholarship offers several advantages over traditional scholarship (along with several disadvantages). Requirements three and five began to coalesce here, and I wondered to what extent making materiality apparent is already to insert the figural. I knew I wanted to create a comic dissertation. I did not know how to draw. Step one, then, was to take a drawing class and enlist a studio artist as one of my committee members.

Since my goal was not to become an artist or enter an MFA program, one class was enough to introduce me to the basic skills. The class itself was a wonderful reminder of how it feels to try to learn a new skill—something my writing students must constantly feel. I was alternatingly frustrated and elated. While working on a still life, I felt tears well up and nearly ripped the drawing in half out of frustration. Suddenly I realized that this is exactly how many of my students feel while learning writing. I highly recommend that all teachers try to learn something they're not good at, just for a reminder of how their students often feel. I cannot emphasize enough just how much empathy this process gave me for my own students' struggles.

By the time I left the class, I had a newfound confidence in my drawing abilities along with a healthy inferiority complex about my lack of expertise. I could draw, but no one was likely to be terribly impressed. That was fine for my purposes.

Throughout the dissertation I learned new visual skills. The drawing class had focused on various traditional media (charcoal, pencil, washes, etc.). I needed to transfer those to comics, which traditionally employs pencil and ink. I read a variety of books and websites to learn comics methods. An interview with Jonathan Hickman on his methods provided me with a simple, step-by-step guide to follow in imitation of an artist I respected (who was also on my dissertation committee).

My methods and style changed with practice. Eventually, I created a style that felt confident and personal—I wanted my readers to be able to pick my drawings out of a lineup and attribute them to me.

It reminded me of the work I do in the composition classroom. I want my students to learn basic skills around argumentation and style, but my greatest hope is that I help each of them develop a distinct voice. Drawing and writing are different means of composing ourselves, and both involve a certain amount of idiomaticity.

At first, I'd hoped to make an entirely comics-based dissertation, something like what Nick Sousanis made a few years later. However, I was advised against this by my dissertation committee and the members of my Research Network Forum roundtable at the Conference on College Composition and Communication, Lindal Buchanan, Ronda Dively, Diane Penrod, and Hephzibah Roskelly. They offered some insightful reasons to create comics interludes instead:

1. Writing a dissertation is already a long enough process without adding drawing into the mix. Some drawing would be applauded as innovative. Too much would detract from the intellectual work of the dissertation.
2. An entirely comics dissertation would face increased scrutiny in terms of its content. Readers might expect a less serious argument because comics are inherently less serious (for some readers). Consequently, my dissertation needed to be overly serious in terms of its content.

Behind these suggestions, a few assumptions lurked. Comics are unserious. Comics are not intellectual work. Comics take more time than text. Combining these assumptions yields the conclusion that the extra time spent on comics would be unjustified.

By the end of the dissertation I'd realized that comics are serious, intellectual work, but that they indeed take a lot of time to make. But if that work really is serious, intellectual work, the extra time is more than warranted because it means more serious, intellectual scholarship.

In the end, I created a print-based dissertation with comics interludes (a PDF set up according to the print requirements of the dissertation including relevant margins, fonts, and CMYK colors). The text and comics began as separate sections, but as the argument progressed from beginning to end, the text became more visual and the images more textual, reflecting the complex imbrication (requirement one) of image and text I was beginning to see. The gradual reversal from semantic text and sensuous image to sensuous text and semantic image allowed me to avoid letting the visuals become decorative (requirement two).

As many have already said, a dissertation is very different from a book. I talked with scholars I knew and respected. I attended a scholarly monograph roundtable to learn about the intricacies of book proposals and converting a dissertation into a book. I relied heavily on various manuals, including Anthony Haynes' (2010) Writing Successful Academic Books, which was invaluable.

With a little preparation under my belt, I drafted a book proposal. I let it sit for a few months, returned to it, and scrapped almost all of it. The proposal was trying to be too many things at once because I wasn't yet sure if the end product would be online, print, or both. I realized I needed to spend some time talking to various publishers to decide on a medium before I could really create a compelling proposal. The imbrication of form and content (requirement one) was more than just a theoretical concern. When form is forgotten, it makes finding the right publisher a great deal simpler. We often speak as though form is the purview of the publisher and content is the author's responsibility, though we know that's not entirely true. This project highlighted for me the ways form and content are a negotiation and conversation between publisher and author not just in digital scholarship but in all scholarship.

During this time I also took part in an MLA pre-convention workshop, "Evaluating Digital Work for Tenure and Promotion," led by Alison Byerly, Kathleen Fitzpatrick, Stephen Olsen, and Katherine Rowe. The workshop taught me how to communicate with my department and university about my work so that it would be appropriately valued—vital skills for any untenured faculty doing digital work.

I should also note that my department has very clear tenure and promotion guidelines that explicitly state that digital scholarship is of the same value as print scholarship. I am quite fortunate in this regard. This liberated me to choose a publisher and envision potential versions of the monograph in various media.

A photograph of three styli: an artist's pencil, a pen with nib, and a digital stylus.

I set up meetings with publishers at two conferences (MLA and CCCC). After talking with editors, I decided on three presses to focus on and created a distinct proposal for each. Here it became clear that each press would have its own medium and constraints. One wanted a PDF, one a pure-digital webtext, and one a web-based text that would become a print text.

I crafted the proposals according to the relevant expectations and attached a sample chapter from my dissertation. In each proposal, I explained that the end project would differ substantially from the dissertation in terms of format and argument (again, reflecting requirement one). I could have worked up individual samples for each press's preferred medium, but this seemed unnecessary to the editors I spoke with. I discuss such medium decisions more extensively in the Practices section.

The Sweetland Digital Rhetoric Collaborative and University of Michigan Press logos, from early in our working relationship. (The logos for both units have since changed.)

Two presses either rejected the proposal or wanted it to shift its focus so radically that it would no longer be the same project. One of these two publishers declined the proposal because they did not feel they could support it technologically. Both were hesitant about the amount of theory.

The Sweetland Digital Rhetoric Collaborative series from the University of Michigan Press stood out as a good fit in terms of both form and content. They would allow me a great deal of freedom in terms of form and were interested in a more theoretical argument, allowing me to shoot my argument through with the figural (requirement three). In addition, they were willing to offer an advance contract, which is especially important for digital work (as I discuss in the Practices section).

A quick warning here: Things move slowly with academic presses, especially when there is a separate entity (the Sweetland Center for Writing, to which the Sweetland book series editors are affiliated) to the actual press (University of Michigan Press, which publishes the Sweetland series). While I was offered the advance contract in June of 2013, I did not receive the contract until January of 2014. There were a variety of really good reasons for this on the publisher's side, but the time in-between was a bit nerve-wracking for me. Nonetheless, I began writing a draft of the monograph while waiting for the contract.

I hate a blank page, so I began by copying and pasting content from my dissertation. In the end, the original dissertation work (heavily edited) became two chapters of the monograph. Having written an outline for the proposal, I had a general sense of the structure of the project. I composed in Word, making notes to myself with highlighting. These notes described the various visual and interactive elements I later planned on integrating. I found it too easy to ignore the very visual requirements I'd given myself (requirements one, two, and three) at this point when my form and content felt so far apart. Writing notes toward the visuals reminded me of the eventual imbrication of form and content and allowed me to propose visuals that would be more than decorative.

It wasn't until I was roughly two-thirds of the way through writing the initial draft that my overall argument really coalesced. I knew a good bit of it before I started, but as every writer knows, writing is thinking.

This new thinking involved some major reorganization. I had been writing in short, two- to five-page sections, each with a bold heading. For the reorganization, I selected all of the section titles using the style menu in Word and pasted them into a new document. Then I wrote a single sentence paraphrase of each section and printed this short version out. I wrote all over the outline until I had some idea of where things needed to be moved to.

Because I had always been composing for the Web, each section was already intended to stand alone to some degree, anticipating readers who might jump from section to section. This made the re-organization much easier, but I still had to revise much of my earlier signposting.

A screenshot of the various section titles in a Word doc.

Throughout the composition process, it was important to plan out workflows. You can read more about my workflows in the Practices section, but I'll describe them briefly throughout the timeline. Beginning in Spring of 2013, I created an article for Digital Humanities Quarterly's special issue on comics as scholarship. (This work is discussed, in part, in the other Inventio piece in this issue!) Publishing that article gave me a chance to practice a lot of the things I wanted to do with the book, including integrating drawings and animations in Adobe Muse. It was a shorter, more manageable piece, and it allowed me to discover a variety of techniques and hone my practices, most of which are reflected in the Practices section below.

A screenshot of Adobe Edge Animate used to incorporate interactivity.

After finishing this reorganized draft, I sent it to my colleague, Joshua Hilst, for notes and began my own revisions and visualization. I printed a copy of the draft, single-sided, and bound it. I worked my way through the draft, line-editing, making notes about organization and argument, and sketching on the blank facing pages. This editing and sketching let me turn some of my visual ideas into actual images. This stage was crucial in addressing all five requirements, allowing me to think through the visuals visually. From those drawings and my notes on visuals in the draft, I created a list of images and animations that needed to be created or found. This document became a checklist throughout the project.

A photograph of handwritten notes and sketches on a printed version of an early textual draft of the monograph.

Upon receiving my colleague's notes, I began revising the written draft. At various stages, I would save my draft as a new file to distinguish between major revisions. My final Word document was called Rhizcomics3.docx. I continued the numbering in my first Web version, Rhizcomics4.muse. My final web version was Rhizcomics14. I say this to emphasize how different the Web writing was from the word-processor writing.

A screenshot of the various versions of the files that made up the monograph.

Happy with my text, I began moving it into Adobe Muse and adding images. Having clear workflows and keeping notes and to-do lists was essential here. First, I designed a single page, one that would incorporate many of the features I planned on including in the final draft and addressed all five requirements. I set up a master page and an initial menu bar that would be revised multiple times. I tested this single page with colleagues and friends before moving on to create the rest of the site.

The first step was putting the words into the site. I created a page for each section and placed the text in before moving on to the next one. At this point it was still possible to reorganize the pages fairly easily, so I did not add links to next or previous pages.

Next, I drew, scanned, digitally painted, and optimized my images. Typically, this visual work was all going on simultaneously, with me drawing one image, painting or optimizing a second, and integrating a third all on the same day. It was helpful to have my workflows overlap in this way so I could revise it as I went (e.g., I noticed the line drawings weren't thick enough on one drawing, so I used a larger nib on the next one).

As I created the site, textual revisions inevitably crept in. Drawing made me rethink my arguments and conclusions. In most cases, I avoided doing my revisions in Muse (web design platforms are terrible word processors) and instead did them in a new Word doc. This enabled me to compose easily and prevented the revisions from getting lost in my 60,000-word Rhizcomics3 draft. Also, it kept the Rhizcomics3 draft "locked." Once I moved on to the Web, that file was just for reference.

A screenshot of an early version of the monograph as a website.

I finished an initial web version and worked with our department's undergraduate research assistant Kyra Lindholm on editing it. I had integrated a comment system, so she simply commented on each page as she went. As she edited, I integrated a search function and created a new table of contents system. The table of contents was dynamic, complex, intricately organized, and ultimately scrapped. At this point I realized trying to meet my requirements could actually interfere with usability, especially requirement five (making the materiality apparent). In an effort to create a materially reflexive project, I had made navigation next to impossible. I discuss these revisions more in the responding to reviewers section.

A screenshot of Adobe Muse showing the dozens of pages that make up the monograph.

Once I had her comments, I went page–by–page making edits and deleting her comments along the way.

The original table of contents shown as clusters of pages radiating out from a central navigation page.

Submitting a Draft

(April 2015)

By early April 2015, I had finished a first complete draft of the book. The file was called Rhizcomics8, so "first" is a bit of a misnomer. I exported the entire thing to HTML and hosted it on my personal university server space. I submitted it to the press as a link and a zip folder containing the web-ready files (HTML, CSS, JavaScript, and various assets).

Waiting

(May 2015–July 2015)

It may seem strange to add this. Four months from submission to receiving reviewer comments may not be much longer than average. However, I think it's important to state that every time a draft left my hands it took longer to get back than I had expected.

A large part of this is due to the nature of digital media. Reviewers were harder to find, the book took longer to read (as reviewers noted), and I'm sure writing the review itself was more complex in that they needed to address visual and interactive elements on top of the usual reviewer responses.

I emphasize how long this took not to complain but to give future writers clearer expectations. If you're making multimodal scholarship, estimate how long everything should take. Then double it. You'll probably still be short. This stuff is complex, but the thorough review process is just as, if not more important to, multimodal scholarship as it is to traditional print scholarship.

Reviewer Letters

I received the reviewer letters in August. Both were positive and both had substantial critiques. My editor offered me two choices: respond to the letters with a revision plan for the editorial board to review or make the revisions and submit the new version to the board for review. I chose the first option for one major reason. With the amount of time involved in revising for the Web, I didn't want to make unnecessary revisions.

I wrote a careful letter describing my responses to the reviewers and my planned revisions. Working from Anthony Haynes' (2010) invaluable Writing Successful Academic Books, I responded in four ways. I pointed to positive comments and described how I would make the book do more of that. I recognized the truth of some of the negative comments and described how I would revise accordingly. I applied the concept of symptomatic misreading to some of the negative comments and outlined revisions that would remedy the misreadings. That left me the opportunity to respond to one or two negative comments by disagreeing and sticking to my guns.

One example of a symptomatic misreading was the table of contents. My original table of contents was visually dynamic. The pages were organized in a series of tree structures radiating from a central circle. Readers could zoom in and out and toggle a breadcrumb trail that would show a linear path through the text. Unfortunately, both reviewers felt the book was disorganized, with one calling the table of contents "expressive" and unhelpful. It was clear my desire to make the materiality apparent (requirement five) was in tension with basic usability.

I wrestled with these comments for a while before I realized that my table of contents was actually too organized. The reviewers weren't saying it was disorganized; they were saying they felt lost—there's a big difference. I took the linear trail and converted it into a simple outline of every page and chapter, complete with numbering.

The readers I showed this new version to all breathed a sigh of relief, saying they finally understood the organization. The interesting thing is that the new one actually contains less organization but retains the overall structure. I think readers were responding to the numbering and the familiarity of the system itself more than the actual organization (which didn't change).

Revision Steps

I worked first with the tech specialist at University of Michigan Press. After a few months he sent me an itemized list of required changes, most of which focused on accessibility. When we met to discuss these issues, it was clear there were two major criteria pulling in opposite directions: sustainability and functionality. The scroll effects I had created seemed particularly unsustainable. On the other hand, this functionality was an integral part of the argument. Losing the scroll effects would alter the argument itself. The press wanted something they could maintain in perpetuity. I wanted something that was fully functional now, even if that meant it might deteriorate over time.

While the initial plan was for me to submit a web-ready monograph, the publisher began to have concerns about the complexity of the site, especially in terms of accessibility and sustainability. We decided to split the book into two versions. University of Michigan Press would house a sustainable version and Sweetland DRC would host a full functionality version on their website. The press would agree to upkeep the simpler version forever, and I would agree to upkeep the Sweetland version for the foreseeable future. The difference in time requirements owes to the University of Michigan likely outliving me. I began to become concerned that the simplified version would divorce form from content and invalidate my self-imposed requirement. We moved forward with this plan, only later to recombine these two versions into one.

Having decided to split the project, I created a to–do list broken down into three separate periods:

1. things I could do while copyediting was going on
2. things I had to do before splitting the project into two versions
3. things that I would do after the split

It was very important to push the split back as far as possible so that I could avoid doing the same revisions twice (e.g., if I split before copyediting, I would have to do the same edits in two documents). This was both to save me time and to avoid errors—creating two versions doubled the chances of making mistakes.

When working on a large multimodal project like a digital monograph, scholars should create opportunities for themselves to test their methods at smaller scales. For me, this was an article for Digital Humanities Quarterly. In composing that smaller project, I tried to be as reflective about my techniques as possible, knowing I wanted to streamline them for the digital monograph.

As part of this reflective work, scholars will need to take notes on their process and test it in small increments. This means testing each part of the process before completing any one aspect. Scholars should design a single chunk of the project from script to finished web site before finishing their entire script. This will foreground any errors or new available means that aren't always obvious during scripting.

For larger multimodal projects, scholars must cast a clear vision of their argument early on and revise that argument constantly. These are simple things we often tell our composition students, but it is too easy to forget them. It also changes radically when composing for new media. It was not until relatively late in the process that I realized I was imitating comic gutters in the way I linked between pages, but this realization helped me to hone my overall argument (see this page of Rhizcomics for more about how this shaped my argument).

Publishing

With a digital project that may change format drastically depending on the publisher, it's important to secure an advance contract. While such a contract does not guarantee publication, it does ensure that the publisher will review the project and shows their interest in it. Untenured faculty are often the ones doing the most innovative work, and their untenured status means they are the ones that most need the security of an advance contract.

Talking to a publisher about medium and format (in addition to the standard decisions about argument, audience, etc.) allows authors to check their assumptions and get on the same page as the publisher.

It is important to be as detailed as possible in these early conversations and get things in writing. Here it is less a concern of people going back on their word, and more about avoiding miscommunication. Getting things in writing offers authors a tangible statement to refer back to in later conversations.

Medium Decisions

It is important to find a publisher before deciding on a medium. This seems counterintuitive. As we know, the medium shapes the argument (requirement one). Different means will be available in different media. However, different publishers will prefer different media and have different requirements—online vs. print isn't the only distinction: There's a big difference between a PDF, a Scalar project, and a website created from scratch.

In a perfect world, scholars might be able to have a clear vision for exactly how the project should look and be able to take that to a specific publisher with the expectation of getting published. For most scholars, though, that's unreasonable. Most will need to shop their project around before it finds a home, and that means being willing to see it fit into a variety of formats. That flexibility multiplies publishing opportunities, but it also limits the amount of progress that can be expected at the proposal stage.

Proposals are generally accompanied by a sample of some of the project's content. With a multimodal monograph, however, scholars need to show their technical ability alongside the project's content. This doesn't necessarily mean a different sample for each publisher. All of the publishers I talked with thought a multimodal sample chapter was enough, whether it was in the format of the proposed project or not. Given that sample, it was important to describe what modifications would be necessary to meet the publisher's desired format.

A Picture is Worth a Thousand Words

The more visuals and interactivity you use, the fewer words you get. My total came in at 62,000 words—the short side for a monograph. Nonetheless, my reviewers initially felt it was too long. Re–organization was a solution to this problem, but the problem is inherent to the medium as well. Short paragraphs are a must and pages start to feel long after 1,000 words or so. With careful organization, scholars can get by with longer pages (the page you are reading is an example of a very long one).

Accessibility

Screenshot of monograph displaying alt text.

As I mentioned earlier, this project created a few tensions in terms of accessibility. I wanted to take advantage of as many available means as possible while catering to as large an audience as possible, all the while arguing that form and content are inseparable. Talking with Stephanie Kerschbaum helped me to think more carefully about my audience, values, and goals. Universal accessibility is a myth—a laudable goal that can only be approached asymptotically. Rather than abiding by a set of practices that will offer accessibility in all cases, I recommend that scholars have conversations with colleagues more versed in accessibility issues. Each project creates its own problems and potential solutions. In terms of best practices, please refer to Access/ibility: Access and Usability for Digital Publishing (Eyman et al., 2016).

Creating Your Own Constraints

While new media create new affordances, they also require new constraints. Often these constraints must be created by the author. I've already listed the five requirements I had for Rhizcomics. It was important that visuals never be decorative but that they be frequent. The form could never be separable from the content. In addition, the tone had to balance scholarly rigor with the conversational voice of the Web. These constraints came from the arguments I was making and from the medium itself. Done right, creating constraints forces authors to create new techniques and tropes, furthering their research and changing scholarly conversations.

Sustainability

I'm very concerned about the sustainability of Rhizcomics. As it was created with proprietary software, I am at the mercy of Adobe in terms of long-term sustainability. This is by no means assured. Edge Animate has already been discontinued, and I find it unlikely that Muse will still be around in a decade. At the same time, the final product exists in non-proprietary format (e.g., HTML, CSS, JS). The TCU library was kind enough to hold on to these files for archival purposes. I've also saved the program file for the specific version of Muse I used. However, Muse tends to create fairly opaque code and it will be very difficult to edit those files in the future. I hope to continue working with the press to create a sustainable version of Rhizcomics in the next few years.

My failings here underscore the importance of thinking about sustainability throughout the project, especially when choosing web-authoring software. I would recommend that authors work with code directly as much as possible to ensure that their work will be sustainable.

When writing for a multimodal project, scholars must keep the target medium in mind from the outset. Writing for new media necessitates the creation of new tropes and composing techniques. Anne Frances Wysocki (2004) defines new media as texts that "keep their materiality visible" (p. 19). In this sense, every new medium becomes old as we get accustomed to its devices and lose sight of its materiality. Creating new tropes and techniques that make materiality visible is essential to making media new.

A photograph of handwritten notes and sketches on a printed version of an early textual draft of the monograph.

I created a system of highlighting to indicate links, images, animations, and design choices. Just like every other piece of the writing process, these techniques were created recursively. This system is described in more depth on the next slide. I didn't realize I needed to be highlighting all of my links until I'd already created a few dozen. This meant the system was in flux right up until the moment the project moved from Word to Muse. This system was distinct from the one used after copyediting described earlier, but certainly contributed to it conceptually.

That said, there are a few general techniques scholars can use in composing scripts for digital or multimodal projects.

Nonlinear Composing

Because I was constantly editing different sections, it was never as easy as just continuing to write on the last page of my document. Rhizcomics argues for writing from the middle, and that's the way it was written. To keep from getting lost, I picked a key word that I knew would never get used in the text ("Gilgamesh") to mark the last place I'd been writing. I could then just search the document for that word and pick up where I'd left off.

Drawing visuals on scripts

I cannot overstate how important it is to compose visually. After finishing the script, I went through a hard copy of it and drew ideas all over the facing pages. Drawing completely changes the composing process, and forces authors to make visual decisions early on in the process.

I created a markup system and saved it as its own file, in which I explicitly stated which color of highlighting denoted which element. I specified a few key elements:

Links

In addition to any external links, I noted relative links within the project. Organizing these by section title, rather than numbering, facilitated easy reorganization. Each link also included a tooltip that described the link's destination.

Navigation

Beyond linking within sentences, there may be generic navigation elements that need to be considered (e.g., next and back buttons). In addition, it's important to consider larger organizational concerns throughout the process. Some of my pages were asides that didn't fit into the main narrative thread and were linked to from various places. Others were more complex still. The Energeic Rhetoric section, for example, performed Thierry Groensteen's concepts of general and restrained anthology by linking repeatedly to a few pages and anchors. I began thinking through these ideas during the initial composing stages, but they were repeatedly revised throughout the entire process.

Formatting

I found it helpful to create a document devoted to the format of the eventual site. I wrote about what I wanted the interface to feel like, the colors and shapes I'd like to include, and how to distinguish between different sections visually. Of course, these decisions, too, underwent heavy revision.

Visuals

I began with thinking about the rhetorical effect of the visual elements I wanted and worked backwards toward content and medium, just as we would teach our students to do. I thought about these decisions from the scripting phase all the way through completion, adding visual elements, changing tone, and creating interactions where I saw fit. Sometimes I knew exactly what I would want to create right from the start and other times I was much less descriptive, leaving the creativity to my future self.

A screenshot of the document including markup in various colors.

Interaction

Again, when thinking about interactive elements, I tried to work backwards from rhetorical effect. Knowing I wanted surprise, humor, ease, or dread gave me a target to hit when brainstorming possible effects. Sometimes I knew immediately what I wanted to have happen (on a page about reflection, a visual reflection of the text scrolls by in the margin) but had no idea how to make it happen. At the composing stage, that's perfect. Figuring out how to make it work is a problem for another day.

Notes to Self

There will always be things that need to be said that won't fit in any system you create. For me, these notes varied between small reminders (flesh out this section with references to these specific scholars), larger considerations (what if I moved this section to the beginning?), self-doubt (this is crap), and self-praise (you've got this, now take it home).

When looking at a completed project, it's difficult to see where to begin for scholars trying to develop new media skills. But the project wasn't produced by someone who had those skills; it was produced by someone who was developing them. In other words, I didn't know how to make Rhizcomics until after I'd already made it.

Most of my coding and design skills were learned for free from w3schools and Codecademy. I still use w3schools frequently to remind myself of syntax. Codecademy helped me learn new coding environments, but I find it cumbersome in terms of reference.

I selected Adobe Muse as my web design software for the project. Muse allows users to create compelling sites with no knowledge of coding required—which is a strength and a weakness. It's easy to create amazing sites quickly and keep them organized, but it's very difficult to extend Muse for tasks it wasn't designed for. In addition, the code it creates is fairly unwieldy. I chose Muse because I felt it would be the simplest solution to creating and maintaining such a large site.

Looking back, I am now completely persuaded by Karl Stolley's (2016) argument for lo-fi web design. Stolley wrote that "Digital works should long outlast the software that played a role in their creation." Because its final form is in a series of HTML5 files, Rhizcomics will likely outlast Adobe Muse. However, the code in those files is needlessly obscured because of Muse's export methods. If I had to do it all over again, though, I would probably just code it in a scripting platform (the project you are reading was created with brackets.io). The higher time cost in terms of creation would pay off in terms of sustainability and accessibility.

Regardless of the software, medium, or coding language(s) chosen, each scholar will need to decide for themself how much they need to know before they can begin. We're all familiar with the PhD student who can't write the dissertation until they read everything. For some reason, though, it's much easier for us to see that impossibility and still require mastery from ourselves when it comes to new skills. But, skills aren't developed in a vacuum. They need a practice to grow. My advice, then, would be to err on the side of less training and more practice. There is a fairly low education threshold for beginning new projects, and those projects are the best places to learn new skills.

When revising a large digital scholarly project, the revision process is at once more involved and simpler. The affordances of the media mean it will need to be tested more thoroughly with a larger variety of audiences. At the same time, commenting systems offer a simple way of collating your revisions and feedback.

Test with a Variety of Audiences

When I write for traditional scholarly outlets, I tend to send my work to a colleague or two to get feedback. In multimodal scholarship this must be done at various stages and with distinct audiences in mind. Scholars must create an interface and learn to teach their readers how to use it. Rather than thinking in terms of any universal accessibility or universal audience, scholars create their ideal audiences and should do so with care.

A screenshot of the monograph showing the two commenting systems.

Commenting

I wanted to integrate a commenting system into my final website, but I realized early on that the commenting system could also help with revision. In fact, I used two commenting systems, each with their own affordances. The University of Michigan Press required me to integrate hypothes.is, an open-source annotation system. I also integrated Disqus comments on every page. Hypothes.is works great for commenting on individual words and phrases, but lacks the visibility of Disqus. In addition, I found it much easier to make revision notes through the Disqus interface. Disqus emails me for every comment and this resulted in a kind of to-do list for my revisions.

I hoped these two commenting systems would complement each other in the final product and encourage conversations among readers. Rhizcomics was meant to serve as a communal scholarly monograph that would continue to be written through its comments system, answering Craig Mod’s (2011) claim that the "book of the past reveals its individual experience uniquely.” In the months before publication, I reached out to two dozen colleagues and asked them to comment on Rhizcomics, hoping they would create the seeds of conversations. While reported being willing to help, very few actually did. Some seemed uncomfortable commenting without having read the entire monograph. In the first few weeks after publication, readers seemed more willing to enter the conversation without having read everything. It remains to be seen whether and how readers will continue to comment.