From the outset of this course, I approached my learning process by experimenting with documentation. The intention, at the outset, was perhaps just simply to continue on from previous experiments with documentation as and of the ontogenesis (Boesch 182) of a project, as described in some of my earlier research on creativity and distributed archives, including a paper on using a cloud-based archive for research-creation, and on distributed version control through an accelerationist lens. (Palumbo 2017a, 2017b) These earlier papers explore the creative potential of leveraging an already established archival system that is used by teams of programmers to manage the changes made to a given project towards expanded uses for research-creation. Through software repository mining, a burgeoning field devoted to understanding collaborative patterns in team software development, version control systems offer researchers insight into the ways in which collaborators communicate, how a project develops, and emergent relationships between artifacts.
I enrolled in the THST6329 Performance Ethnography course because I assumed that sometime in the future I would be studying groups of people who communicate with each other through these systems, and I wanted to have some foundational understanding of a critical reflexivity from the positionality of an interlocutor and the challenges that may arise.
The mechanisms of version control systems concern the way in which the changes to a work are represented. Where Professor Kazubowski-Houston told us at during the first class that
A lot of performance ethnography has been to communicate research findings. a more rare approach is using performance ethnography to use it as a way of doing research, as a form of a research-creation practice. (Kazubowski)
It is in this way that version control systems are well-suited to ethnographic practice. To use a version control system is to engage with, structure, and learn from the history of a process. It also emphasizes that participants contribute contemporaneous sentiments about the people involved, the artifacts they produce, and what is exchanged between them.
I also felt that as performance ethnography positions performance and creativity as an approach to doing research, and as a form of research-creation practice, it would well suit my curiosity with the usage of version control systems as creative tools unto themselves. Moreover to this point was the fact that version control systems provide one with ways of telling stories about the work that was conducted therein. Last, writing code is an improvisational practice, which echoes something that turned out to be oft-repeated in class, where ethnography is less a method, but a methodology, and one that is improvisational. This appealed to me as a musical improvisor, and proved to be tremendously freeing, in the sense that as I developed a particular portion of this paper, I let my imagination ask myself questions about how I might be able to use programming to augment my process, and augment the representations of ideas and research within this very paper. As someone who is quite new to programming websites in HTML, for example, this paper presented many, many occaisions to learn how to program a website, because many of the elements included herein -- such as the portion of the page that presents different versions of the website to visitors -- came to me through this improvisational, playful approach.
Over the past year, I have been interested in documentation, specifically with the ways in which notes taken during class or readings can be used as source material for later work. I experimented with this in the form of marginalia when I conducted my reading of Sarah Pink's Doing Sensory Ethnography. (Pink 2015) It has long been a practice of mine to write marginalia in the books that I own, but in preparation for my in-class presentation on the book, I came across Pink's assertion that "Ethnographers rely on both memory and imagination (and indeed the distinction between the two can become blurred) to create what we might call ethnographic places." (Pink 44) It was this idea of the ethnographic place that helped me to reframe my marginalia as spatiotemporal events that future readers could encounter. Following this, I began to wonder, would the directory on my computer that contained my notes was also an ethnographic place? What of its contents, would they be considered the neighbourhoods of the directory-township?
For my participation in the Imaginitive Ethnographic Walk on November 20, 2017, I decided that I did not want to do an audio recording. As a sound artist, though I have done many field recordings in the past, I was concerned with the reality that a microphone has its own bias, one that is very different from my own. In other words, listening to the audio captured by a microphone would be listening through a very different set of ears. I instead wanted to restrict my 'documentation' to only my own memory.
In our group's workshop the following week, we came up with a the idea of using graphic scores to represent our respective experiences traversing the creek, and rather enjoyed the thought of visitors to our site being able to retrace some aspects of each of our journeys by interpreting our scores. What appealed to me about this was that a graphic score is an abstraction, it pictorially represents a communication between a composer and performer, and as such served as a useful analogy to my intention of not wanting a quantitative transformation of experience into captured data, for I could use creative writing and drawing to express my memory and imagination.
What tended to emerge over the course of the past 4 months of in-class discussions was a sentiment about embracing the messy, unpredictable, and asynchronous tendencies of ethnographic research. [such as?] I was inspired by this very early on. And as I have been increasingly using a web browser as the delivery method and medium of my research papers, for web pages could be considered as parasitic. I grew emboldened to represent both the 'finished' -- albeit probably best described as 'most-recent' -- version, along with all of its precedent versions in some form of non-linear or asynchronous fashion. I wondered if perhaps this would mean that upon first visiting this web page, a visitor would be taken to a random version of the page. However, would I make it possible to then navigate both forward and backward in time through different page states? Would I not provide this option, leaving the reader to have to refresh the page and look at the next version to load? I became interested in the prospect of presenting a 'finished work' in both linear and non-linear forms.
For instance, during the latter few classes of the course, in-class notes were taken using Sangwon Lee's Live Writing interface (Lee 2015), as can be seen here.
It is worthwhile to point out that from each experiment in form and representation came at least one desire to try and use this form somewhat differently the next time. This is a practice which has been described by Götz Bachmann as that of radical engineering, where:
Radical engineers fundamentally challenge existing notions of (here, digital media) technologies: their basic features, purposes, and possible futures. Their radicality is not to be confused with political radicality, or the radicality of “disruption”, or the radicality of some of engineering’s outcomes. Theirs is a radicality that puts them outside of assumptions in the wider engineering field of what is obvious, self-evident, time-tested or desirable. Their positions are so heterodox that they often stop calling themselves “engineers.” (Bachmann 2017)
For example, after using the Live Writing website for a few note-taking sessions, I began to want to embed the playback in a web page. This was motivated by wanting a visitor to remain on the page instead of linking outward. However, due to a security restriction placed by Lee into that website's code [what was the name of that problem? something to do with allowing embedding in iframes? search notes for this] this simply was not going to be possible.
I had been introduced to Sangwon just one week previously, and so in response to my request for assistance in embedding the playback of my writing in a page, he recommended that I try using a similar tool called Scrimba. Lee and the developers of Scrimba have different goals in mind for their work, but what their projects share is the ability to capture the input of text in real-time, and allow anyone to play it back. Where they differ, is that Lee's project is better suited towards creative and/or limited research writing, while Scrimba is designed for live programming, which is an educational tool. With Scrimba, I was suddenly able to write again in HTML, and render it to a webpage all within the same recording. I used it thusly in the Black Creek Scores project to create a short text and image-based graphic score of my path through the creek and some of the thoughts I had along the way.
Another affordance of the Scrimba tool, was that a visitor may interact with the text replay and in fact modify its contents. This appealed to me along the same lines as how Sarah Pink depicted the ethnographic place as a locus of participation, and that they are not in stasis (Pink 49). I pursued this idea a little further in one of my contributions to the web site, where I published the chat logs between each of my fellow group members. In one of these chat logs was the login and password information for the website, which would mean that anyone could come in and change or delete anything within the site, and even delete the site itself. This appealed to me, as I reflected on issues of permanence, and of challenging the notion of authorship and singular vision. It reminded me of overgrowth and urban decay. Situated as an ethnographic place, a website is prone to the same reauthorship that occurs in other human-centric spaces: graffiti. (Miller 142) With this in mind, I very much hope to return to this site in several years to find it 'contaminated' with new content, ads, misinformation, and more.
So I had intended to return to Scrimba.com to write the entirety of this paper. However, there were a few concerns. What if the Scrimba website produced an error during the recording, would I lose my work? With no information available related to the maximum length of a recording, would it be worth it to risk losing potentially entire pages of work by pushing this neat tool beyond its yet unkown limit? While some of my recent other works have each been crafted in the name of embracing and exploring failure, the fact that this ethnography carried a looming deadline pursuaded me to find another solution.
Following a critical reflexivity, I also questioned the intention and effect of framing the entire writing process in one singular, linear representation -- an endurance performance, if you will. Who would this serve? Through this I became conscious of a desire to perform a level of intelligence and 'put-together-ness'. Along this analysis of my own anxiety surrounding the presentation of myself in my work and process, I asked to whom and for what purpose would a very long video serve? What might one conclude about myself as a performer that I, at the outset at least, expected that a viewer would watch the entirety of a paper's coming into being?
My concern over the length of the video was inspired in part by the reaction to my contribution for our group's project presentation. We had decided that we would forego a formal presentation, and instead provide the class with the link to our site with the intention of permitting them roughly two thirds of our alotted time to browse. This, of course, would not nearly be enough time to view the content that all five of us had produced, but we wondered if even that might provide a space for reflection. However, the score that I contributed, Soundwalking Through Black Creek, was eighteen minutes and twenty-nine seconds in length, which meant that even it in its entirety would not be watched by anyone in the class.
Additionally, along a concern for the length of the video, gien that the video would purport to represent my entire mental process in the form of contemporaneously written text, it might give rise to a false narrative of an author possessing a consistent, singularly focused mind, rather than one that paused to do tasks such as laundry, devote time to serious procrastination, and ponder about what, at times, I might rather be doing.
In the discussion period during our group's class presentation on the imaginitive ethnography of Black Creek, some of my fellow students provided great criticism over our choice to represent our collective effort using a web site. They offered that such a choice privileges those who have acces to a computer: that in order to access it, one would require either a computer or a smart phone, and even the latter case would not provide the same quality of experience that laptop, with its larger screen, would. This point about screen size was directed at my score in particular, because the contents in the video would become partially obscured on screens as small as a those found on smart phones.
I began to wonder what, when freed from preoccupations over the macrotemporal (Roads 3) accounting and subsequent representation of this paper's ontogenesis, what specific action could define the shorter temporal segmentations of the process? This particular problem offered the potential of a new way of approaching the work.
For example, the following is how the ethnographic paper actually began: as a previously abandoned attempt at writing a web page for my in-class presentation on Sarah Pink's Sensory Ethnography:
I had been looking at it and several other of my notes from that class for an entry point into starting my writing process and began, rather absent-mindedly, to begin writing this paper inside this earlier document. This carries forward a practice that I experimented with leading up to that presentation, where I wrote marginalia in the pages of my copy of Sensory Ethnography with the full intention of sharing the book with the class during my presentation. So it seems fitting that the point of articulation that segments this ethnographic text's beginning as a Document is that very same act carried out in recursion.
I discovered that this actually speaks a little to my process, as I tend to find it is helpful to begin with a page that has something on it, regardless of whether that content will remain in a later version.
While the many affordances of the version control system used in the aforementioned papers written in the spring of last year were sufficient for the contemporaneous documentation of changes to a project and the analysis of data contained therein, there was one affordance in particular that actually posed a hinderance on what I intended to do with this paper. The system uses revision tracking algorithm that stores only whatever lines of a document have changed. Referred to as delta compression, (McCullough 2012) this method saves potentially tremendous storage space because it does not store the full contents of the file at each state save, but instead collects the differences between the current version and its immediate precedent version, and stores them in a type of file called .diff. While this method is certainly beneficial for reducing the amount of data transmitted between collaborators, and is suitable for retrieving an older version, the diff's representation of a file, when taken out of the context of its nearest versions (the version before and after it), may not provide a viewer with enough information of the document's role in an overall piece of work.
Note: conversion of the .diff into HTML formatting handles by the diff2html library, by rtfpessoa
I wanted to have a copy of each saved state of this ethnographic document, paired with its corresponding diff, which would include the full context of everything I had already typed in the document. It would also be very helpful to have the program convert the latter into HTML so that there could be more options for how to make use of it in the final webpage, and so that I would not need to do this step for every single diff generated, which, at the time of this writing, is up to 133. I set out to write a small computer program using node.js, which would run on the command line in the background while I worked.
The first step that the program would need to complete would be to watch this document I am writing -- whose filename is index.html -- for modifications. In other words, for each time that I saved index.html, the program would notice this and begin to create the files I wanted.
Once a save had been noticed, the program would then need to request a diff from the version control system, collate its contents alongside the rest of their full context within the document, and convert it to html. After a few tries on my own, I came across a javascript library called diff2html (Fernandez DATE) that converts .diff files to .html.
The last step would be to make sure that the version control system was now also tracking the two new files that were generated by the program, so that eventually, when the imaginitive ethnographic paper was finished, all of its previously saved states and diffs would be archived on the same remote server and thus accessible to visitors of either the repository or the website.
Featured below is the code for the program that produces both a copy of the index.html document and its full-context diff in html for each time the file was saved:
Works Cited Bachmann, Götz. “Utopian Hacks.” Limn, 8 May 2017, https://limn.it/utopian-hacks/
Fernandes (rtfpessoa), Rodrigo. “Diff2html.” Diff2html, https://diff2html.xyz. Accessed 26 Dec. 2017.
Kazubowski-Houston, Magdalena. (2017) "THST6329: Performance Ethnography, Week 1" York University, Toronto, Ontario. September 11, 2017
Lee, Sang Won, and Georg Essl. “Live Writing: Asynchronous Playback of Live Coding and Writing.” Proceedings of the First International Conference on Live Coding, ICSRiM, University of Leeds, 2015, pp. 74–82, doi:10.5281/zenodo.19322.
McCullough, Matthew. “Git, Compression, and Deltas - An Explanation.” Gist, 14 May 2012, https://gist.github.com/matthewmccullough/2695758.
“Musecore Player to Embed in Webpage.” MuseScore.Org, https://musescore.org/en/node/139506. Accessed 27 Nov. 2017.
Node.Js. https://nodejs.org/en/. Accessed 27 Dec. 2017.
Palumbo, Michael. Version Control Systems as Tools for Research-Creation: (Or) Using Git to Write a Paper About Using Git. 11 Apr. 2017, https://michaelpalumbo.github.io/self-referent/
Palumbo, Michael. A Technicity of Distributed Version Control Through Accelerationism. 6 July 2017, https://michaelpalumbo.github.io/spth6155/
Pink, Sarah. Doing Sensory Ethnography. 2015. /z-wcorg/.
Roads, Curtis. Microsound. MIT Press, 2001.