In this five-part series of blog posts, I am recounting, by way of documentation, the process by which the modernist community has begun to create a free, digital anthology of modernism. This installment summarizes the proceedings of the first “board meeting” for the project.
During the MSA 2014 conference held in downtown Pittsburgh in November, 2014, a group of people interested in this open-access modernist anthology met at the Omni William Penn’s Terrace Room to map out a preliminary model (or at least a set of expectations). This group included (from left to right) Jeff Drouin, Alex Christie, me, Claire Battershill, Stephen Ross, Chris Forster, Matt Huculak, James Gifford, and Marit MacArthur:
In between fielding many questions from our waiter, who couldn’t quite come to believe we were actually ordering food, rather than simply taking up a table, we had exactly the kind of conversation you’d imagine having in the middle of a terrific conference: chaotic, energetic, lumbering, and ultimately very inspiring. Please keep in mind that much of what follows has been superseded or amended or completely overturned since this meeting, but it nevertheless reflects the brainstorming process, and we hope these records will allow us to come back and revisit the course of our discussions (largely in the service of reminding ourselves what we are doing!)
In the initial conversations on the MSA email list and on Twitter, I suggested The Modernist Sourcebook, conveniently hashtaggable to #modsource. This name was seen, I think, as lacking zazz, so we free-associated a bit. Matt Huculak suggested Sylvia, as a reference to Sylvia Beach’s Shakespeare and Company. Some of us were worried that, even though it was the most inspired and thematically relevant suggestion, it would be too obscure, impeding our visibility on search results. (Since then, I’ve wondered if we can use Beach and/or Shakespeare and Company for branding cues, perhaps by using the bookstore sign to source a font in the eventual website banner or by using a stylized image of Beach’s head as an icon for returning to the home page.) (I also want to suggest that we apply Sylvia as a name for any tool or script or app that we might, God forbid, have to design specifically for the anthology.)
Claire suggested Free Modernism, which we found phonaesthetically pleasing and snappy, but which was critiqued for its similarity to other projects following the “Free X” template. To retain the sense of open-access and freedom in Claire’s suggestion, I considered various synonyms and landed on Open Modernism, which Matt, to emphasize the plurality of our definition of modernism, immediately amended to Open Modernisms. Perhaps not a lot of zazz there, either, but hopefully it will work for now. It can shorten to OMP (Open Modernisms Project), which will nicely harmonize with the MJP and the MVP without disappearing into those other abbreviations (as MAP, Modernist Anthology Project, might (in addition to overlapping with MAPP)).
As Chris Forster argued strongly throughout the meeting, we all agreed that maintaining strong sense of purpose was necessary. We decided it would be to create an online anthology of modernist source texts. In spirit, it is a re-conceived Lawrence Rainey anthology, but with a few twists:
It’s free for students.
Instructors can create custom anthologies.
The offerings will be updated to align with current definitions of modernism.
By crowdsourcing our “Wish List” from teachers of modernism all around the globe, we can respond directly to instructor needs, not distort them or ignore them to meet publishers’ bottom lines. Unlike currently existing free online texts (think Gutenberg), this new anthology will be thoroughly vetted, proofread, and edited by scholars of modernism.
We also discussed subsidiary affordances—ones that are perhaps not as core to our mission as the above three mission statements. These include
reproducing any unusual typesetting (such as that seen in manifestoes)
giving each custom anthology internally consistent pagination (for ease of use in the classroom)
providing access to the texts in multiple formats (.html for viewing in a browser, .epub for viewing in an ereader, .pdf for printing, and perhaps plaintext for use in basic digital humanities tools)
It was understood, however, that this final affordance (plaintext for text mining, visualization, et cetera) may represent the thin end of the wedge in terms of falling away from our core values and mission, leading us to the next topic.
Expansion versus Core
What we were calling “primary secondary” works or “primary criticism” for pedagogical purposes (not for research) is the core mission. These works include nonfiction written by important modernists who normally write poetry or fiction; significant works of philosophy, science, or social science; and literary criticism written during modernism. These works help you teach modernist fiction and poetry, such as criticism written by T. S. Eliot, Ezra Pound, and Virginia Woolf and manifestoes written by F. T. Marinetti, Mina Loy, and Wyndham Lewis.
After meeting the core goals, we can we can add sound files (as Marit emphasized) and poetry (as this is included in the Rainey anthology). A later iteration might include fiction, but this goal should not influence the early decision-making process about workflows, hosting, et cetera. Overall, we outlined a plan for potential future expansion radiating in this way: 1) prototype versions, 2) “new Rainey” anthology, 3) sound and poetry, 4) fiction. Not everyone agreed with these outer layers, it must be noted, as some argued that the clarity of “primary secondary” texts should necessarily constrain the project so that we can make it happen quickly.
Questions of expansion (considered above() regard the generic scope (which texts to include), whereas the “tiered” system (under consideration in this subtopic) refers to procedural matters of creating the anthology and of unrolling various functionalities over time. Matt Huculak’s original insight about a “tiered” process was seconded by Stephen and others during the meeting. Stephen aptly summarized this process as “scan, dump, and deal,” though precisely what these mean was debated and left relatively undefined. However, my general understanding of the tiered process included these steps:
Gather PDF or image files of first edition publications so as to address copyright for editors.
Create HTML or Markdown file via probably an OCR translation into plain text. OCR must be hand-corrected.
Add coursepack functionality (and/or additional layers of markup/encoding).
Even at this level of generality, the questions multiplied: Do we do this among ourselves (the dozen at this meeting?) or broaden it to MSA in general and affiliates (BAMS)? Do we want to specify file formats more strictly?
Early in the meeting, we all agreed that a somewhat standardized set of metadata would be required not only for establishing the provenance of the text used to create our edition (hence linked to our copyright management), but also for establishing identifiers that will be necessary for creating the website and the coursepack. Matt suggested 5 pieces of information: author, date of publication, publisher, provenance/archive (institution), and uploader (who provided file initially so we can track them down in case of a problem or question). I also stressed that we should indicate initial publication context, such as a little magazine or novel or collection. That would be 6 pieces of metadata. Already, we see a bit of “metadata creep,” each turn of the screw making the sticklers happier but the rapid prototypers warier.
Chris suggested that thorough quality control may require TEI. Most of us immediately and emphatically rejected that idea, but Chris’s underlying concerns are still valid. Stephen and Matt stressed facility and speed, whereas Chris, with his practical knowledge of setting up digital archives, was already, at this early stage, thinking about how we would deal with errors, variants, and other editorial details. (Balancing this trade-off between speed and quality will continue to animate our later discussions.)
It should go without saying that we will try in all good faith to follow copyright laws.
The Canadians in the group persuasively argued that we should host this project from Canada in order to take advantage of their copyright laws. For most texts, going by “50 years after the death of the author” (or, if published posthumously, then 50 years after publication) takes us further than the American standard of “texts published before 1923.” Where the American law works better, we can arrange to have it hosted in America. In the vast majority of cases, each entry will meet copyright law both in America and Canada.
No matter what, though, we decided—per Stephen and Matt’s advice—to include a notice on the anthology website that transfers all liability for copyright infringement on the end user.
Stephen’s institutional machinations
By the time the meeting started, Stephen had already made critical contacts with various institutions and existing projects:
University of Victoria: He has worked out an informal agreement in which the University of Victoria libraries host our materials. By working with the UVic libraries, we can take advantage of various kinds of support, from technical support to librarian consultations to resources that would otherwise cost us money. (We may have further access to UVic resources beyond the libraries.)
MSA: As this project began with an exchange on the MSA email list, it is only fitting that we work with MSA on this project. MSA will put its backing behind us both in terms of reputation (we are one of their affiliates) and in terms of money (a bit of financial surplus might be drawn to us). As incoming MSA president, Stephen is admirably positioned to negotiate this relationship, and he convincingly argued that an open-access anthology is a resource that would serve MSA’s members extremely well.
ModNets: In addition to MSA, we will be affiliated and authenticated through Pamela Caughie and David Chinitz’s ModNets. They will guarantee our accuracy and fidelity, as well as provide another way for teachers, students, and scholars to find our site.
Jeff Drouin is also promising to access his MJP for texts, links, and other forms of affiliation.
Claire suggested, to unanimous agreement, that we create a small set of prototypes: about 1 per person interested in doing so for a total of 5-10 prototype texts (hopefully more toward 10 than 5). Each prototype should represent some kind of unique limit case or genre, trying to present every kind of problem we’ll have later down the road (emphasis on trying).
As the meeting began to succumb to the usual exigencies of conference-going, Claire, Chris, and I felt that there were some issues that still needed to be ironed out before soliciting the prototypes. These issues (most of which have been foreshadowed in this blog post) will dominate the next blog post, Part III, which will summarize the emails exchanged among us and Andrew Pilsch. This exchange was intended to present the main issues in an informed, logical way to the larger board. Parts IV and V will record the email exchanges among this larger group prompted by the “report” generated by Claire, Chris, Andrew, and myself.