Network Conversations - Federico Brega

[taylorcate]

Subject: Nutting Variorum: ideas for an application in archival description contexts

Dear Taylor,

As I told you in our first conversation via Instagram, I was completely mind-blown by your Wordsworth editions project. Beyond Wordsworth himself, what caught my attention was the use of a version-control environment for the "representation" of a complex object --the poem. By "complex" I'm referring to the fact that it is hard or impossible to determine which is the ultimate, definitive version of the poem, and this approach, as far as I can understand, allows for the "object" to maintain its "identity" or "singularity" while at the same time being able to be represented distinctly depending on the version that is accessed in the proposed digital flipbook.

Just to give you some context: I am Argentinean-Chilean and I work as an archivist in the context of the visual arts in Chile, and I've worked in the organization, description and cataloging of a few pretty straightforward institutional archives (in contemporary art museums), which could be fairly well represented with a combination of ISAD-G descriptions for archival fonds and series and a custom-made but rather simple database for the description of individual documents. I should add that most of this was done with proprietary software --a choice determined by the ease of access and by the severe limitations of our public and private museums regarding the implementation of IT departments within their workflow and, most importantly, their budgets. I've worked with these limitations for a few years, and I've developed some methods for collecting, crunching and recycling data with an open-source mind in a non-open-source environment. (Data recycling is an issue I'm passionate about, ours being a context in which many archival organization and description projects use very limited technology or just plainly lack a data-driven approach to archival processing, and also because these projects are very short-lived, mostly because of lack of funding.)

Mind you, I'm not at all versed in coding; I do most of my work in a very sui generis, home-made style, combining and recombining spreadsheet software formulas, bits of code found on the internet (mostly SQL) and some open-source text analysis tools. I warn you of this in case my coding ignorance transpires in my interpretation of your project.

Now, though, Jennifer McColl, a colleague of mine has presented me with a very interesting cataloging problem, way more complex and idiosyncratic that those usually found in institutional archives. It concerns Ronald Kay's personal archives. Ronald Kay was a very interesting and unorthodox Chilean visual artist, video artist, poet and philosopher. He was the first to introduce the fundamental texts of Walter Benjamin in Chile in the 1970s, and he was very much a Benjamin- or Aby Warburg-like sort of author/artist. His archives are full of notebooks with later-added sticky notes, newspaper clippings, photographs and all sorts of other textual and visual atachments wich have been very consciously added to the pages as a sort of hypertextual composition. These objects shouldn't ever be separated from the notebooks themselves, since they are loaded with meaning; with bits of information in at least five different languages which complement, reveal, transform, displace, or alter in some way the meaning of the contents in the underlying page, establishing the most amazing connections among these fragments and between them and the overall notebook/project.

In the Instagram post from the @wwnuttingvariorum account I read "In this project, I will transform an existing digital edition [...] into a fully linked, version-controlled, interactive website and digital flipbook which calls upon the various versions of the poem as potential paths for users to follow". As soon as I read this I thought this would be an amazing model for the cataloging of Ronald Kay's notebooks. Most notably, I think the very concept of version is a powerful way to describe the various instances of these heterogeneous documents. Instead of treating them as objects which operate as containers, the idea of them being conceived as environments which have different versions seems way more appropriate, or at least more akin to the idiosyncratic nature of these documents/objects. I hope this makes sense. I think of these objects not as the sum of n number of definite parts, but rather as a multi-faceted document in which every instance of the document (every combination/compostition) is a given totality (a version of the totality) of the document. I hope I'm not completely missing the point in your project, and if I am, I hope at least these loose ideas can function as a startpoint for an interesting technical and methodological/epistemological discussion.

I'd love to follow your project and to hopefully make a space for this discussion within the project's forum so that more people can chip into the discussion. Are you in charge of this project? Is there any way I can learn more about this class and stay in touch? How can I learn more and get more involved? Please excuse this very much without-a-point email. I am very much in the margins of this sort of projects but I really wish to learn more and hopefully get more involved in the broader field of Digital Humanities in the near future. Moreover, the Nutting Variorum project seems like exactly the kind of thing I must be paying attention to, re:the Ronald Kay and other "complex" archives. I'll sign up into GitHub so that we can continue the discussion there if it's pertinent.

I hope some of this makes sense and that you'll welcome my probably very foreign interests in your project.

Best regards, Federico

Received January 28, 2019

[taylorcate]

Re: Nutting Variorum: ideas for an application in archival description contexts

Dear Federico,

So glad to receive your email and message on Instagram. I hoped when I began my project that connections of this kind might happen but I never dreamed someone working on such a rich, complex textual corpus would contact me! I am so heartened to hear of your work and I'm glad to be of any assistance. I would like to disclose that I am documenting our conversation on the Issues board for the project which you can access here: https://github.com/taylorcate/NuttingVariorum/issues/12. If you are uncomfortable with our conversation being archived in a public space, I will remove the Issue right away. I feel, however, that these conversations are worth documenting and sharing just as much as the other elements of the edition are.

In answer to your questions: Yes, I am in charge of this project. I am a second year Master's student at Loyola University Chicago studying Textual Studies and Digital Humanities. I am completing Iteration 2.0 of this edition for my capstone project, the defense of which will grant me my degree. I submitted Iteration 1.0 of the edition as my final in ENGL 413 with Dr. Paul Eggert—renown textual scholar and editor of Charles Harpur's works who, unfortunately, just retired from Loyola. The best way to get involved with the project is to keep up with me on Instagram, @wwnuttingvariorum, and on GitHub where all of the project documentation, class assignments, milestones, connections, and witnesses are housed for access from anywhere with internet. I have compiled my editor's commentary in the project Wiki (https://github.com/taylorcate/NuttingVariorum/wiki) and all current documentation is composed on the Projects (https://github.com/taylorcate/NuttingVariorum/projects) and Issues (https://github.com/taylorcate/NuttingVariorum/issues) boards. Most of what I have formally written on the project is in reference to Iteration 1.0 and to my perceived failure in creating a truly linked, open-source edition that adheres to a dynamic digital environment in a real way. I resolved to have another go at it, and that's how we arrived at The Digital Variorum of Wordsworth's "Nutting," 2.0.

I've been in love with Wordsworth since I was a Sophomore in college. I had a mentor and professor who passed this obsession and much of what I know about textual scholarship on to me and worked with me nearly all of the time I was in school. Under his instruction and funding, I went to England twice in 2016—first to take a short course in manuscript study at the Wordsworth Trust and then to attend the annual Wordsworth Summer Conference. When I was applying for the Master's program at Loyola, I was asked to declare a specific research interest and all I could think about was my love affair with William Wordsworth and the country that keeps my soul, so I said I wanted to make a digital edition of Wordsworth's "Nutting." One that truly shows how much time and energy Wordsworth put into editing that poem—editing himself just as vigorously. I attempted to do this the Spring semester of my first year at Loyola in Dr. Eggert's class but did not succeed in creating the edition I envisioned. Somehow, before I even realized it I think, the perfect plan arose to complete a new iteration in time to do a bit of Spring traveling to Wordsworth country to present the edition at the Wordsworth Trust for curator Jeff Cowton and colleagues. So here I am, attempting to create something very "textual" in an inherently "technical" space in the effort of expanding the dynamic between humanities and computational studies collaboration.

I want to make an edition that is more experiential than readable. I want the edition to illustrate, read aloud, highlight variants, "live" transcribe the poems, and coax the reader down Wordsworth's winding paths of possibility. I plan to achieve this through a sort of choose-your-own-adventure, flipbook which calls upon the various versions of the poems as potential paths. The flipbook will have a companion website, run out of GitHub pages, that houses select portions of the documentation for the project and some sort of call for public participation. As of now, I'm conflicted about marking the versions up in TEI-XML because I'm concerned about the time restraints and my lack of knowledge on how to call that markup in a useful, representational way. I feel a more innovative approach would be to focus my energy on making sure the flipbook is appealing to a greater audience. Ultimately, my hope is that by doing this work in a public, version-controlled environment, I can create a model for other textual and archival projects to follow—one focused on inclusion and knowledge sharing over personal monograph production.

In addition to the "Nutting" project, I'm also the Content Editor for MUDDLE: the digital lit mag that celebrates the messiness of composition. Rebecca Parker, MUDDLE's Technical Editor, fellow DH Master's student, and my friend, should be your next stop on your journey to understanding GitHub and its immense capabilities. Together we created the MUDDLE project as a means to evert the process of composition, in all its forms. I am wondering whether you'd be interested in submitting to the mag? I absolutely love your Instagram account, @fdxnotes, and I think your work would blend really well with some of the other content that's been submitted. We accept everything, for the most part, the only condition is that you need a GitHub account and you need to fork the repo. We are more interested in the artist's process than anything, and I think you may be very interested in how the issue will be "published." We'd be happy to simply stay in touch as well!

If there are any links or accounts for your project please list them! I'd love to follow you on all your platforms. Again, thank you for reaching out to me and I look forward to our continued collaboration!

Sincerely,

Taylor-Cate Brown Loyola University Chicago

Replied January 28, 2019

[taylorcate]

@fdxdatacrafts

Thank you so much for agreeing to document our conversation publicly. I think it is absolutely crucial to preserve these interactions and I'm honored you opened a GitHub account to talk to me here. Now, in response to this absolutely intriguing project you're working on:

Kay's archive sounds like a textual scholar's dream, or nightmare. I'm not totally sure which, however, I am sure this corpus presents a really interesting challenge when it comes to how to represent it. As I mentioned in my Instagram message, Marianne Moore's poetry and ephemera pose a similar challenge. There has been an effort in recent years to digitize Moore's notebooks, however, there are a great deal more extant materials in Moore's estate held at the Rosenbach Museum in Philadelphia which, digitized, could alter the way Moore is edited entirely. Cutting quotes, pictures, lines of verse, drawings, etc. from newsprint, magazines, and books, Moore drew connections between multiple mediums the sum of which contribute to the complex images she describes in her poetry. It is argued that she incorporated other people's fragments into her work as a way of democratizing the poetic medium—to bring in voices not typically associated with high verse. I found that I struggled to piece together her images in my mind without seeing the ephemera in relation to the poems. I proposed an edition/archive that hypertextually links to the digitized extant materials—essentially a hover-over reading text that pulls from an internal database of digitized ephemera. I feel this would create space for less scholarly audiences to engage with Moore's work in an interactive and purpose driven way. In our current political climate, I think it's absolutely crucial to find ways to make the literary and artistic cannons we fight so hard to keep alive relevant and useful for unanticipated audiences.

I must also admit that I am totally naive to the coding requirements of such an immense project. My "expertise" lies with Textual Studies and I've only very recently adopted the title of Digital Humanist. For the "Nutting" project, I'm hoping the only coding will be in the html generated website. I'm in the process of determining whether the Book Creator application will allow me the degree of customization I need to create the variorum. Hopefully we can inspire new methods and ideas in each other through our conversation! Looking forward to your thoughts.

Best, Tay

[fdxdatacrafts]

Dear Taylor: your references are simply awesome, and technically beautiful. Thanks for those. I'll be sure to follow their progress and learn more about their inner workings. Still, they look like a finished product, or at least a very desirable endproduct for a digitization+documentation archival endeavor. But I'll take the liberty of stepping back for a second and focus on the specific subject of archival descriptions themselves. So specific, apparently, that I could not a find a Wikipedia article about them --just articles about description rules, standards and guidelines. Archival description (AD) "serves, as far as possible, as the definitive representation of the archival material and which is required to establish intellectual control over it and promote access to the information which it contains" (archivists.org on the theoretical approach of the International Council on Archives to AD). Archival descriptions are collections of semi-structured data that help navigate archives and retrieve specific files or items within the archives. AD data has indexes and access points which allow for implementation of functional databases, but it also contains big chunks of unstructured data, most frequently natural-language text but also diagrams and images and other resources which may be more fit for representation of complex series, files and documents. The richness and the complexity of these might be lost if they were to be represented simply as data tables in a data set. Most archival description software focuses primarily on the representation of the structure and the ease of retrieval of these chunks of unstructured data (which I'll call "objects" for the moment being, though I reserve my doubts about the accuracy of the concept in this case). Objects include multiple media, but I think text is ultimately the most versatile, complexity-allowing medium when it comes to description. Sometimes the complexity of an object can reside not merely in its arrangement, but instead on "softer" attributes like its history; its singular materiality; the experience of its handling, reading or interpretation, and so on. Text, descriptive and narrative text, is still and might remain the most powerful way to convey these attributes about an archival item or collection. Nonetheless, I think some of these items and collections pose challenges for all "static" media, including text descriptions and audiovisual material. The Kay notebooks, plus myriad other archival items -but let's keep the focus on the former for now- make room for experimental representation media, namely software or computer programs which have the power to represent an experience of the document, provided that the item itself should indeed be judged to be more accurately represented as an experience than as an object. I'll return no this later. For now, let's say this can be a risky approach for many reasons, because it might toy with the instiutional (=> historical, legal) authenticity of the represented object. The authenticity (the authoritave capability/potential) of documents is one of the pillars of modern archives and therefore of archival science. Archives are evidence, like fingerprints, and not just information. Archival items exist because something specific happened, an they are the remainder of this event. All things considered, I think "artistic" archives should necessarily have more wiggle room when it comes to the issue of the authenticity of items within their holdings. Authenticity is a longstanding and unresolved issue in art history, art theory and aesthetic philosophy, and there is no reason to think that this issue should be any simpler or more straightfirward in the AD of this type of items. This is to say AD in an arts archives context should allow for experimentation that reflects [on] the problem of the authenticity of artistic objects themselves. This might in return shine some critical light on classical AD itself, thought that is a different discussion which I won't address now. I'm sure this is nothing new, but for now I want to shift the focus back to the Variorum. I can document this discussion with some bibliography later. I think the Variorum is a very interesting example of a software-type descriptive object which could easily be used as AD proper. Maybe not for an entire archive, but certainly so for complex items whithin it. I'll make sure to compile a decent bibliography on the matter and post it here if it's of any interest to your project (it definitely looks like it would be). I hope I'm not completely missing the point here or trying to reinvent the wheel. There certainly exists software that completely complies with these problems. Mukurtu CMS by Kimberly Christen is a fine example. But once again, I'm for now more curious on a theoretical analysis of "archival descriptions as software" than on specific computer programs. I want to get there soon, but right now I'm happily caught in theoretical issues. We can use these examples as occasions for such a discussion. I'll try to address these in further contributions here. As soon as I have more time I'll edit this and add links to my references, but I rest assured they're easy to google. All the best, F

[fdxdatacrafts]

I think I hardly made a point in the previous comment; it seems like I was really just throwing around vague definitions and beating around the bushes instead of posing a question which I think is probably fundamental for us to have a productive conversation in the theoretical discussion I'm trying to push here (I'll put further efforts in pursuing the technical discussion as well), but which I also fear may be too dumb or naive. The question is: Does Nutting become software through your representation? If so, what does this mean, what does it imply, what does it entail? The latter I wonder about more emphatically. Likewise, the question I'm asking myself is can archival descriptions made in this fashion be considered software? What does this entail? What definition of software are we using here? Does this definition come from the discourse of engineering, of science, of philosophy?. For some reason I think this are like first-grader questions, but then again, I've shuffled through the bibliography coming from the archival sciences and I've rarely seen the discussion posed in terms like these. Maybe it's because the terms are somehow invalid, meaning it's a false dilemma, maybe an ill-peddled one? Is any of this relevant? Maybe I'm stating this equation wrong, but in some given order the terms "representation"; "software"; "materiality", plus some other whatchamacallits, must be able to constitute a valid, relevant and fundamental question. I hope this makes sense. Best, f

[taylorcate]

Dear Federico,

I am absolutely appalled that is has taken me so long to reply to your wonderful comments. Any patience you still have for the project, though unwarranted, is extremely appreciated!

You have given me so much to think about with archival descriptions and the methods we use to represent material items in our collections. I completely agree with your assertion that text is, perhaps, the most valuable descriptive method we have for conveying the complexity of objects, and text augmented by digital representation takes on an entirely new life.

In response to your questions: Does "Nutting" become software through your representation? & Can archival descriptions made in this fashion be considered software? What does this entail? What definition of software are we using here? Does this definition come from the discourse of engineering, of science, of philosophy? I believe that the "Nutting" text itself is what's known as a "fluid text"—coming from John Bryant's 2002 book. By definition, a fluid text is "...any literary work that exists in more than one version. It is 'fluid' because the versions flow from one to another" (Bryant 1). As we collect these texts in a digital environment, we create new works in the transmission in much the same way the work's previous editors and collaborators did. There is no definitive text, just the fact of a multiplicity of texts existing in an intricate web of sociological interactions. So, does the digital text become software and, if so, what definition of software are we using? Defined by Google, software is "the programs and other operating information used by a computer." As I collect and edit "Nutting" digitally, I impose a machine readable structure on an otherwise completely material corpus, thereby imbuing it with software but not necessarily transforming it entirely. I believe that the edition that results from this project will exist as yet another "text" in the transmission of the "work" but one that works with the medium it is published in instead of against it like many digital editions that adhere to print conventions do.

My "Nutting" edition will be a truly dynamic, multimedia experience that invites users/readers to choose their own path through the text's transmission. The edition will feature audio recordings of the poem, illustrations of the actors in the transmission, time-lapse transcriptions of the manuscripts and print versions, and hyper-textual linking to extended commentary and network connections. The edition will be a modern intervention on the way we "read" texts, accommodating users of all abilities both physical and intellectual. It will highlight the potential for our digital projects to connect unanticipated audiences with the archival materials in our private institutions by both augmenting their existence through interactive media and by literally linking the archives that house them.

I hope that makes some kind of sense. I apologize if any of that was just convoluted rambling. Please send along any resources you think are applicable to the project! I am always interested in reading about methods in DH adjacent fields. :relaxed: Thank you again for your patience!

Best, Tay

[nadinejl]

This is all absolutely magical. All of it! My favourite little nugget of gold:

'...all I could think about was my love affair with William Wordsworth and the country that keeps my soul, so I said I wanted to make a digital edition of Wordsworth's "Nutting." One that truly shows how much time and energy Wordsworth put into editing that poem—editing himself just as vigorously.'