Introduction
CanLit Guides (canlitguides.ca) is a blended learning digital resource produced by a team of researchers at the journal Canadian Literature: A Quarterly of Criticism and Review (canlit.ca). We were both graduate student members of that team, with one of us working as the workflow coordinator (Jamie) and the other as a lead writer and developer (Mike). The project emerged as an educational expansion from the wider journal ecosystem (Figure 1). The guides present digital materials that draw on the journal's scholarly experience to add to embodied courses, especially by introducing students to key areas of interest in the critical study of Canadian literatures. The editors of Canadian Literature eagerly enabled graduate students to develop research, teaching, and digital writing skills by mentoring them as they created much of the content of CanLit Guides. All academics working on the project keenly supported open accessibility, collaboration, and the creation of new digital genres by working in spaces and through processes that "emphasize collective acts of composition" (Weingatten and Frost 2011, 48). We soon learned, however, that it is one thing to want to be collaborative, inventive, and pedagogically innovative in a digital space, but quite another thing to ensure that work gets done in such a way that the whole of the project is greater than the sum of its parts. Brown et al notes that digital humanities projects in general tend to never quite finish, since large collaborative ventures are "considerably more prone than traditional humanities projects" to having issues with "doneness" (2009, 1) due to sprawl. With CanLit Guides, we felt the desire to embrace the specific needs of the scholarly community of CanLit educators and scholars along with, and at times in tension with, feeling the pressure to find ways to complete the work efficiently. These issues obviously motivated the team of writers already, but the project's time-sensitive and pedagogically focused funding further reinforced them. In the end, the collaborative creation produced a range of robust guides, a workflow process, and key structures, which we call digital genres, that will facilitate the ongoing growth of content.
Figure 1: The Canadian Literature ecosystem, including CanLit Guides (diagram by Zoya Mirzaghitova for Canadian Literature).
We do not present this case study because we have solved the issue of doneness. Rather, we do so because we struggled with this problem throughout the writing and development of the project and would have benefited from examples of considered reflection on prior projects, especially from the perspective of graduate students embroiled in them. We offer our critical reflections on several challenging areas, and our solutions to them, in order to model for others one way through such ventures. In what follows, we begin by situating the current CanLit Guides (as it is now available online) within its developmental history. We then turn to a discussion of our digital genre development and the advantages of creating a workflow to facilitate their creation. A theme that runs through this paper is that we constantly made tradeoffs at CanLit Guides: on the one hand, we needed to strike a collaborative balance between our desires for the freedom to be creative and the needs of the collective to be productive and predictable. On the other hand, we also needed to find a balance between the needs of instructors to have a clearly structured and helpful supplement to their classwork (without determining how they might utilize it) and the desires and needs of students for accessible and engaging materials in a digital environment. The balance we struck between individual and collective productivity, and between distributed pedagogical needs, was produced through an ethical evaluation and construction of structures and processes that fostered as much freedom as possible in all areas within the constraints of collectively determined goals. We found that collective composition is enhanced by using a democratically created writing process, by having clear objectives and timelines, and by always remembering why, and for whom, a digital document is being created.
It is helpful to place this project into context. CanLit Guides began as a means to mobilize a massive digital archive of about two hundred past issues of the scholarly journal Canadian Literature (1959—2008). The archives represent, in many ways, a time capsule of both the history of the journal and of academic criticism about Canadian literature, a history that the journal became increasingly self-aware of as it approached and recently passed its 50th anniversary. Moreover, the results of CanLit Guides speak to the issue of digitization of archives. It is one thing to digitize fifty-years of a quarterly journal of criticism, it is quite another to use those digital resources to create something that students can use to help them study literature. And students do use it. According to Laura Moss (2014), using data from Google Analytics,
in the year after its launch, CanLit Guides had 18,701 visitors from 121 countries who looked at almost 75 thousand pages. Unsurprisingly, the most traffic came from Canada (14,705) and the US (1,335). The top ten traffic sources rounded out with many visits from India (340) and the UK (337), followed by Poland (251), Germany (260), then Australia, Spain, the Philippines, and Sri Lanka (11).
What is more, because our archive was so big, students working with CanLit Guides could get a feel for how a text or an issue is currently being discussed in the scholarly literature but also, and just as importantly, they could come to see how that conversation changed over time.
The project has just begun transitioning into its second phase of development, which will be less graduate student driven. It will now enable specialists to produce new materials within the generic frameworks, and it will encourage scholars to use the workflow processes, both of which were established in phase one of the project. The first stage funded and mentored graduate student writers to collaboratively develop the content and structure of the guides, whereas now the writers will be specialists who are unpaid but who will gain academic credit instead. While there will be less collaborative development at the beginning of the workflow in the second phase (in brainstorming and structuring chapters), the later processes of collaboratively editing, revising chapter content, as well as adding visual and coding elements, will remain, and will continue to fund and mentor graduate students.
In this paper we focus on the graduate student experience of working on a large, collaborative, digital project, and the way that the team at CanLit Guides came to figure out issues of digital genres and digital workflows. It is our hope that the lessons we learned in phase one of the project will help any group of scholars, be they established scholars or emerging ones, who want to work collaboratively in a digital space.
The creation of CanLit Guides
CanLit Guides began from a seemingly simple issue. While it was wonderful to make digital content like book reviews, editorials, and academic articles openly accessible online, those steering the project—editors Margery Fee, Laura Moss, the late Judy Brown, and managing editor Donna Chin—realized that getting students to engage with these discourses of the discipline required further support and mentorship (and often more than any instructor could reasonably provide). The editorial team realized that this digital archive presented an opportunity to produce a blended learning resource that scholars and high school teachers could use to supplement, but in no way supplant, their classrooms (i.e. it is not a MOOC). CanLit Guides grew to provide students with information, resources, and exercises that introduce them to literary theory, Canadian literary history, and to works of Canadian fiction, poetry, and drama, in part by drawing on the open access archive of back issues of Canadian Literature, and in part by providing peer-reviewed, open-access content written directly for students. In the end, we developed a resource that introduces students to key issues in the discipline, and mentors them in developing approaches to these issues by modeling and encouraging an active interaction with past scholarship and commentaries.
The journal Canadian Literature developed the guides with funding from the University of British Columbia Teaching and Learning Enhancement Fund (TLEF), which supported the hiring of two graduate students (Mike Borkent and Cara Woodruff) to begin developing the project. The in-house web designer, Matthew Gruman, also worked to custom build the website to support and functionalize the range of pedagogical materials the team developed. Over time—with ongoing TLEF funding and, later, other teaching and experience-focused funding sources at UBC—the staff grew, adding undergraduate copy editors and two other graduate student writers (Karen Correia Da Silva and Alissa McArthur). Two years into the project, a workflow coordinator and writer (Jamie Paris) was hired to help streamline the collaborative production process. All of the graduate students took on independent roles writing, editing, and coding content as needed, and the journal editors regularly contributed, as well as vetted and edited, content. For all involved, the process of moving from traditional scholarly writing and materials to a digital resource was both challenging and rewarding.
The project required a hierarchized, or what we call "asymmetric," collaboration between the editors and the graduate students, as editors mentored and directed the students at crucial moments, while also relying on them to develop materials and to engage carefully with the issues. This asymmetric collaboration better utilized the knowledge and skills of both junior and senior academics to meet the needs of teachers of the discipline for an authoritative resource while also meeting the constraints of the medium and the time pressures on the editors. The collaborative production process settled into one much like a wiki in which several authors worked together to develop parts and to revisit and edit their own and other's content at several points in the development. This was done off-line in word processing programs with track changes and other forms of reporting and interaction (including sharing and archiving documents on our mutually accessible internal server). By the time any content was fully digitized and accessible online, it had moved through this communal development process to become a high-quality, peer-reviewed product.
To acknowledge and facilitate the collaborative development of the guides, the editorial team made the difficult decision to avoid attribution of authorship on individual pages, unless specifically adapted from someone's previous teaching materials, since the entire team would work on outlining, drafting, and editing a document together. Instead, all who worked on the project were given credit for their work by acknowledging their roles on the masthead. Students who worked on the project were required to sign over their intellectual property and authorship rights to the project, since they were being paid directly for their labour and to facilitate the corporate development of the project. In phase two of the project, writers will be given an authorship credit so that they will be rewarded for their authorship in other ways more common to the Academy, like merit pay and promotion. Why would a DH project, which is typically invested in tracking networks of connection to mobilize peer-to-peer review, take authorship rights away? Largely the decision was practical. Students will, over time, graduate or get busy on other projects, and as such individual components of the project needed to be transferable between the different writers. At UBC, a student owns copyright on the teaching materials they produce unless they formally relinquish it. Thus, CanLit Guides kept a delicate balance, whereby the journal carefully collated the names, roles, and dates of contributors on the masthead, to give credit and allow students to claim the work on their CVs without giving the students copyright over their work. Furthermore, because the journal, Canadian Literature, holds copyright to material created for the project, writers can edit previously published content created by students who have left the project. Thus, the decision supported the goals of collaboration and sustainability. Finally, because so much of the work was done collaboratively, it was difficult to say who were the "authors" of any one page.
The issue for graduate students (and editors working towards tenure or promotion) was not one of production, however, but one of credit. It is easy to cite one's work on a curriculum vitae when you can give a URL for a page that you personally created or co-wrote (such as on a blog or encyclopedia entry); it is harder, but by no means impossible, to cite the guides themselves as projects one worked on (masthead dates and guide launch dates help corroborate contribution claims). Perhaps more importantly, this ambiguous collaborative mega-author makes it difficult for others to assess the labour involved (a problem further exacerbated if they have little or no experience with digital projects). Documentation of labour is crucial in overcoming the appearance of ease that is often ascribed digital developments. These issues continue to haunt many digital projects in the humanities and the scholars behind them. For instance, how do we classify this work? For the graduate students working on the project, being paid directly for our labour complicated traditional academic distinctions between scholarship, service work, and research assistantships. Arguably, CanLit Guides falls into all of these categories.
Digital Genre Development and CanLit Guides
As a collaborative digital project, we spent significant time developing the structure, look, and flow of the information we wanted to present. Over the course of several years, we developed a range of digital genres for the project. Understanding the digital genres we created to make CanLit Guides work requires placing the process of genre creation into the context of the process of soliciting user feedback on our original product. After creating some initial content for CanLit Guides, Mike developed guest lectures (under supervision from Laura Moss) that focused on different levels of undergraduate education, class sizes, and guide content types. Each class featured a range of lecturing, reading, and activities drawn from the guides. We solicited feedback from students and instructors through surveys and focus groups. We also had ongoing input from statistical analysis of user traffic (facilitated by Google Analytics), so we could see how students were working with the guide while Mike was teaching. Surprisingly, the user traffic also revealed that we had a global audience (mentioned above) that was far more varied and engaged with the resource than we had initially assumed. The team took this feedback and began restructuring the guides to better reflect the needs of our broad scholarly and pedagogical community.
One of the things our user feedback told us to do was to develop more discrete content parameters for chapters and their composite pages. The introduction of the workflow required further clarification of the constraints upon both form and content and the solidification of types and forms of content we were developing. In short, we needed to clarify our genres. As Das, Kleut, and Bolin observe, "[t]he new burden of providing fixity in [digital] communication is being met by an increased reliance on genre" (2013, 32). The new media environment presents a stimulus for creative pedagogical resource development, but it also needs to develop in a way that fixes the new in relation to the old. Our initial model had not sufficiently embraced genre, which made it too easy to slip and slide between types of content, weaving literary, cultural, and theoretical interests together into complex arguments and sliding back and forth between knowledge development and pedagogical application. While certainly interesting and critically rich, this approach made the material too idiosyncratic and specific to particular interests and, thus, potentially made it harder to integrate into different teaching styles and class dynamics. It was also very conducive to sprawl. In essence, we had transposed our teaching and specialty areas online, without clearly transforming them into a resource that was structured sufficiently well to supplement the work of numerous instructors in their own embodied classrooms. The question became less "What do we want to write about?" and far more "For whom are we writing and why are we writing for them?" To some degree every work of academic writing must deal with the issue of audience, but we had to do so in a way that allowed our content to remain authoritative and accessible to a diverse, global audience.
One central value of the academy—and one which Kathleen Fitzpatrick argues in Planned Obsolescence is the central concern of scholars assessing digital objects—regards the "authorativeness" of the digital object (2011, 101). At the same time, as she notes in her essay "The humanities, done digitally,"
scholarly work across the humanities, as in all academic fields, is increasingly being done digitally. The particular contribution of the digital humanities, however, lies in its exploration of the difference that the digital can make to the kinds of work we do as well as to the ways that we communicate with one another (2012, 14).
We would add to Fitzpatrick's argument that the digital is not only changing the way scholars communicate with each other; it is also changing the way that instructors and teaching assistants are communicating with their students. The CanLit Guides project is caught up in questions of authority and audience within this new communicative scene. While the guides present an authoritative front by being produced by a major journal, which gives them a disciplinary stamp of approval, they also seek to supplement rather than supplant embodied classroom pedagogy, deferring authority to instructors and their teaching institutions as well.
Changing digital practices create the conditions for new digital genres that better meet the specific pedagogical needs of the flipped or hybrid classroom. These new digital genres could not simply be repurposed lecture notes in a different platform, but had to undergo "remediation" (see Askehave and Nielsen 2005). In fact, we have developed a bricolage of sorts, developing digital genres that are not wholly novel, but that do not simply re-present print texts (although the archive texts we draw from are "simply" PDFs of the original articles and reviews). While the content of the different pages often seems similar to print genres like question sheets and encyclopedia entries, their global accessibility, as well as the influences of visual and interface qualities on reader comprehension, subtly alter the writing and presentation. The traditional content needed to be reworked in order to be remediated without over-simplifying it. Clarifying our digital generic forms ended up clarifying the critical content. As digital genres influenced by the varied audiences and mediated potential for multimodal communication, the guides have the challenge of trying to be attractive to multiple audiences while also remaining coherent and authoritative. This meant we had to carefully envision and assess the classic generic issues of form-style and audience in relation to pedagogical needs, or what Lloyd Bitzer would call our exigencies (1999, 221). Furthermore, clarifying just what a guide was and what its contents should be called proved unexpectedly difficult. However, we knew that this clarity would help collaboration both at the development stage and in future, when new contributors would require guidance, even templates, for their work.
To develop digital genres we attended to the limitations of the medium and the needs of our implied and diverse user community. As Bernal (2005) notes, for digital media "location is ambiguous, and to be made socially [and we would add pedagogically] meaningful, it must be actively constructed" (661; qtd, in Coleman 2010, 491). Constructing a digital resource meant recognizing the central role the medium played in how our materials were presented (including screen sizes, browsers, and lower-grade CPUs) and for whom they were being presented. We worked hard to make our materials as accessible as possible, including by adding image ALT tags and descriptive captions to facilitate reading programs for the visually impaired, and by adding in-text glosses for culturally specific terms that might confuse international readers. We also presented links to articles in the Canadian Literature archive and to external articles in the works cited for each page to encourage students to access and engage with the source material. Hyperlinking to other resources, videos, and other online media extended the impact of the content of the guides as they modeled a critical disposition toward cultural practices and artifacts beyond literature as well. By making ourselves as accessible and connected as possible, we hoped to make guides that would be attractive and user-friendly for many diverse audiences.
While meeting these various needs of digitally-savvy and globally diverse students, we also needed material that would attract specialist teachers. The necessary balancing act of writing for both novice readers and knowledgeable specialists bears a strong resemblance to the phenomenon of crosswriting for the dual audience of children's literature: children and those who choose books for them (see Nodelman 2008). Striking the right balance is difficult, and the benefit of a collaborative project is that there are always numerous sets of eyes, ears, and hands working to strike it together. Similarly, we needed to position ourselves as an authoritative pedagogical resource in a digital world replete with information. We were not writing an archival tool, a Wikipedia-like encyclopedia entry, or a blog. Nor, however, were we simply digitizing lectures and assignments. We are a flipped learning tool (King 1993), but we are also not behind a pay-wall. We are open access and, thus, accessible globally. By developing peer-reviewed, informed, nuanced, and yet accessible materials, we hoped to appeal to teachers seeking pedagogically focused literary and cultural materials.
In order to meet the needs of instructors, we isolated different content types (history, theory, literary form, skills, etc.) and made the pages of each chapter more modular and self-contained so users could flip non-linearly through chapters and still grasp coherent ideas. For example, while we initially linked two or three pages together to develop a complex portrait, now we had to divide that approach into self-contained units. While chapters still maintained a linear logic and topical coherence (in a sense, they always had a thesis), each page became a more discrete, coherent unit as well. Depending on the theme of the chapter, the pages contained different types of content, often grounded in contexts described and arguments made in the pages of archived Canadian Literature articles and other scholarly sources, and all chapters ended with exercises that encouraged students to apply the knowledge they had gained in the chapter. For example, a chapter on a particular critical perspective might have a broader introduction, a more detailed engagement with a select concept (and often an argument made in an article from Canadian Literature,) a series of questions one might consider asking about a literary work, literary engagement with a few poems (including framing questions), and perhaps another application exercise, such as an article summary. Sometimes, a chapter focused on historical or cultural content might have several longer pages broken down by era, practice, or discourse, but which would shift to questions and possible application exercises as well.
The flow of the guides was to begin by introducing the students to a topic and then to give them the tools to complicate it. For example, our guide on "Gender, Sexuality and Canadian Literature" (http://canlitguides.ca/guides/gender) opens with a chapter that introduces students to issues of gender and sexuality and shows how Canadian Literature participated in the development of feminist and queer criticism in Canada (Figure 2). Importantly, the chapter encourages students to see feminist criticism as essential to the study of Canadian literature while, at the same time, seeing such criticism as diverse and ongoing rather than monolithic and established. The chapter on sex and gender ends with a page (http://canlitguides.ca/guides/gender/1/5) that tells students about Cree and Irish artist Kent Monkman and his alter ego Miss Chief Eagle Testickle, a trickster figure who is openly and subversively sexual while, at the same time, actively reversing the colonial gaze. Monkman's work is then contrasted with Edward S. Curtis, and artists like George Catlin, who attempted to freeze Indigenous peoples in culture and in time, believing that Indigenous cultures were static systems of belief, rather than dynamic and malleable interactions with the changing conditions produced by globalization and colonization. The questions get the students to consider the ways in which Monkman's work encourages us to see First Nations cultures and sexuality as fluid and dynamic. By doing so, the chapter structure encourages students to apply what they have learned about sex-gender and queer issues to contemporary art and to a new conceptual domain. Furthermore, the guide goes on from this chapter (as seen in the side-bar of Figure 2) to give more comprehensive chapters on historical and theoretical perspectives on gender and sexuality in Canada and its literatures, and chapters on literary works by Susanna Moodie, Ethel Wilson, Daphne Marlatt, Ann-Marie MacDonald, and Shyam Selvadurai, that facilitate the application of knowledge across the guide.
Figure 2: Screen capture of the opening page of the guide "Gender, Sexuality, and Canadian Literature" (http://canlitguides.ca/guides/gender).
By structuring the chapters and guides in this manner, we hoped to facilitate an active, dialogic pedagogy in the classroom by off-loading the work of background lecturing onto the guides and always leaving students with questions and ideas for further engagement with a given topic that could inform classroom discussions and student writing. At the same time, the guides introduce students to the discourses of the discipline by integrating and interrogating them in accessible language and in a digitally attractive and networked form. Through these strategies, we hope to have developed guides that appeal to both student and instructor audiences and that maintain an authoritative yet open (searching, questioning) approach to literary criticism in Canada. We have attempted to balance the action of pedagogical crafting and the exigencies of archival and discursive integration with pedagogical needs, all within the limitations of digital media. Thus, the guides actively engage with what Coleman (2010) describes as the three categories of cultural politics, vernacular cultures, and prosaics of digital media within a pedagogical model.
WorkFlow and CanLit Guides
We created our own digital genres out of a need for more restraint and predictability in the development process. To work well as a team, we also needed to track easily where everyone was at with different aspects of the project. The democratically developed workflow process sought to track creative processes while encouraging collaboration. Perhaps most importantly, however, the workflow process was iterative. The process began by recognizing that the students and the editors working on CanLit Guides were producing a lot of text, but that the project was getting so complex so quickly that nobody was quite sure what everyone was doing on the project at any given moment, nor when to stop developing in particular directions. The project was being very generative, but was also conducive to sprawl, which made it difficult to contain and release the material as discrete and coherent guides. Thus, a workflow coordinator (Jamie) was hired to develop a clear understanding of the labour being done on CanLit Guides and to find a means of encouraging and directing productivity. Through analyses of analogous workflows for writing environments in print media (places like newspaper, magazine, and academic book publishers) and in digital media (such as webpages, blogs, wikis), he found that the terminology that content creators used to describe their processes were different, but that the basic process of content creation, editing, proofreading, and publishing were the same. Recognizing this overlap made it easier to discuss the parameters and processes of workflows to print natives (the editors) and digital natives (the graduate students) alike.
Jamie conducted interviews with the editors, staff, graduate students, and undergraduates on the team to ascertain how the collaborative process unfolded and to identify and explain bottlenecks and other obstacles in this process. After conducting this process, he created a mock workflow and asked everybody for feedback and suggestions. The process of creating the workflow, then, was democratic and student driven, and everybody had at least two points of input so that they did not feel as if a process of working and collaborating was being imposed upon them. Moreover, nobody's comments were privileged because of rank, and we found that the undergraduates, graduate students, and editors all had valuable contributions to make to the process, given their varied expertise with print and digital platforms and processes.
Once instantiated as part of the guide development process, the workflow remained iterative and subject to revision. Importantly, this ensured that the experience of the team was always being listened to, and it allowed people to recommit after each iteration of the process to the workflow itself. The collaborative re-visioning of the workflow also reduced interpersonal conflicts within the team by allowing them to voice their concerns before the issue(s) that were bothering them became obstacles to the development process. The workflow rapidly stabilized around the aptitudes and idiosyncrasies of the team members, maximizing their contributions; in time, the team made fewer suggestions about the process itself. The iterative nature of this workflow also meant that it differed from other models in select ways. It might not work for all development teams, but it worked for ours, since everyone genuinely committed to the process. The iterative and subjective workflow model promoted productivity and quality by drawing on the strengths of each team member.
The final (for now) workflow (see Figure 3, Figure 4) consists of four basic processes, each having subtasks within them. The rough content of guide and chapter foci is decided upon through a collaborative brainstorming session. The general topics for guides and chapters were informed not only by area expertise of team members but also by input from the archives and the wider academic community. Early on in the project, Mike isolated and compiled areas of particular strength in the archived materials, so that we had a working database of topics and writings to work from. Furthermore, we surveyed instructors of Canadian literature across Canada to ascertain their areas of interest and preferences. We were able to further refine our topic development through keyword searches of the archives and n-gram analyses of language use data (for instance, one can trace the rise of the concept of "multiculturalism" in the journal and in the broader popular discourses. As one might expect, the debates in the journal arise approximately at the same time as the first usages more popularly, but the journal debates surged while popular usage took longer to develop). Through these various means, the team collaborated in choosing content for the guides.
Figure 3: CanLit Guides Workflow, showing multiple chapters in parallel development.
Figure 4: Complete Chapter Workflow.
Chapters were developed after the tasks for the chapter's completion were assigned and a writer was chosen (usually based on who had most familiarity with the subject). The writer was in turn often assigned a direct or indirect journal editor who could mentor them through the process of researching and writing that particular chapter. At this phase of the project all those involved in the chapter's creation were contacted by the workflow coordinator and a rough timeline was created so that everyone knew both when they when expected to complete a task and when the person they were sending the chapter to would be expecting it. Once the chapter was written, it was sent out for peer review by an area specialist, either internal or external to the journal and UBC depending on the content. The guide would only go to final revisions if it passed through peer review. Once through peer review, the writers put the guide through a revision and first copy edit process. At this point it was common for the chapter to be revised substantially (although less so for content) by another author. Furthermore, the chapter would not leave the content creation and revision phase until it had editorial sign off, which also often introduced further revisions. Once the text was revised and approved, we would find the media necessary for the page (pictures, videos, etc.), code the page in HTML, and do a quality control check. Finally, someone who had not worked directly on the file would be brought in to give the page a final read in full HTML markup; they ensured primarily that the guide had the look and feel of a CanLit Guides page and that the previous revision processes had not introduced any mistakes.
By using this "assembly line" workflow process we were able to double our outputs from the previous year, avoid redundant developments, and reduce disagreements. Moreover, an unexpected yet welcome side benefit of this process is that it encouraged "non-linear" collaboration as well. Team members knew, for example, who would be doing the media for their chapter in advance and thus could discuss their ideas with that person or even make recommendations. For instance, by lucky happenstance, several interesting images of settler posters from the Canada West magazine, which were originally going to serve an illustrative function in the guide to Indigenous literatures, were instead employed to introduce students to, and model, a feminist reading of a cultural artifact by way of introducing the guide to "Gender and sexuality" (http://canlitguides.ca/guides/gender/1). This fortuitous event, produced through parallel development of multiple resources, of course was not the norm. Rather, what we learned quickly was that almost nothing worked exactly as planned. People sometimes fell behind, got sick, or did not meet deadlines. However, the structure of a workflow allowed the project to take such setbacks in stride, since writers and editors could shift to other tasks in relation to other chapters, and turn to one they had initially planned to develop later. Having multiple documents in development at different stages allowed for flexibility while sustaining productivity. But such flexibility in the "assembly line" required stability both in the workflow and in relation to the digital genres and the content being developed.
Conclusions
The story we have told about CanLit Guides is one of compromises made democratically among a team, within an ethical but asymmetrical power structure, all with the goal of enabling us to "emphasize collective" work (Weingatten and Frost 2011, 48). As graduate students, working on CanLit Guides was an important opportunity to interrogate how we operate in and out of the classroom. We were challenged to carefully consider how we mobilize a range of textual and discursive resources to facilitate the development of skills and knowledge that students need to operate in the academy. These issues involved considerations of both general pedagogy and disciplinary specificity, which presents an incredible opportunity to learn about the major issues in the discipline of Canadian literature and about how one might approach teaching a wide range of literatures and topics. Furthermore, the project introduced us to how digital spaces can present opportunities for bringing various discursive and pedagogical practices together. Additionally, the editors mentored the graduate students on their writing, research, and pedagogy, as well as on the workings of the journal itself. The project presented a unique opportunity for the graduate students to work closely with leading specialists and award-winning teachers of Canadian literatures. The skill of transmediating knowledge into digital environments also serves us well as young scholars, since more and more material, both popular and scholarly, is being digitized, including online journals like Digital Studies. Knowledge of the practical challenges of digital presentation will help us work productively in the increasingly digitized practices of publishing as well as digital resource development which universities and other educational institutions are growing increasingly interested in.
Finally, this project arose from a specific cultural and academic history because it was put forward by the journal Canadian Literature rather than an English department. In the first editorial for the journal, George Woodcock offered a manifesto for the journal, arguing that it first only "exists, and must become its own justification. But its very existence may have been rendered possible only by the faith of people and institutions who have been willing to become — in one way or another — its hosts" (1959, 3). Something similar might be said of CanLit Guides. The project exists, but only due to the faith of the editors in the productive energies of the graduate students charged with running it, UBC's commitment to pedagogical innovation, and the readers and teachers who use it. The time and effort needed to produce CanLit Guides is justified by the community of educators and scholars who will integrate them into their own pedagogical practices. As two of the several graduate student writers who worked to pursue this open-access pedagogical initiative, to define the parameters, to unfold the medial structures, and to render the materials of scholarly discourse and Canadian literatures accessible and attractive to undergraduate students, we hope to have met the faith and hospitality of the editors at the journal with the enthusiasm and determination their ideas deserved, and to have built a resource that will be justified in its role in the digitally inflected pedagogy of the future. As we hope to have shown in this paper, developing CanLit Guides required the careful balance of individual, collective, and pedagogical needs in order to support a dialogical and digital pedagogy that instills a critical disposition toward Canadian culture, and especially toward the vibrant field of Canadian literatures.
Acknowledgements
This paper would not have been possible without the initiative of the editors at Canadian Literature in envisioning the need for guides, securing funding, and encouraging the development of the CanLit Guides project in such an ethical and supportive way. Thank you to Judy Brown, Donna Chin, Kathryn Grafton, Margery Fee, and Laura Moss! We are incredibly grateful to have worked with and learned from them on the CanLit Guides project, and to Margery and Laura for feedback on this essay. Our gratitude also extends to the many other people who worked with us on the project as writers, reviewers, and critics, all of whom are listed on the CanLit Guides masthead. None of this work could have been accomplished without the dedicated attention everyone provided to this emerging resource. Our thanks to Canadian Literature for their kind permission to include all of the images in this article.
The views expressed in this essay are strictly our own, as best as we can discern. So many conversations have happened around these issues that surely we are inadvertently channeling other members of the team at times. Nonetheless, all errors and omissions are our own.