As expected, research this summer has been a near constant encounter with the unexpected. I apologize for not blogging sooner or more often. I would attribute this lack of blogging to the large and seemingly not ending collection of records I’m working with. Most days I would rather get through a shelf of records than collect my thoughts, but I suppose I should start and breathe a little and pause to reflect on what I have done and what I am doing.
As I began work on the Underground Archive Project in June, I decided to first build the website and eventual home of my archive database. In doing this I ran up against the age old adage of research, (I can’t say I wasn’t warned) : “this is more work than anticipated”.
With my archive I continually try to pay close attention to the medium I am working with: record sleeves and text, and with my website tried to design something most reflexive to these medium. So, I bought a domain, I bought a server, I installed the Omeka application, as advised by Josh Honn, and began building. Along the way I learned how to transfer files with an FTP client connection, how to use cpanel, when to throw a website theme away and start fresh, how to code (very clumsily) in php, how a website really functions, what UTF-8 encoding is and why it matters when importing a .csv file, the list goes on. While learning all of this took me longer than expected and held me up for a week or so from getting to the archive itself, I think anyone could agree the home I created for WNUR’s Rock Show collection looks and works superbly: www.undergroundrchiveproject.org.
The challenges and setbacks of creating that website led to learning I in no way foresaw when I started this archive-minded project, and has given me tools and knowledge I may very well use down the road. To bring this back to the records, my experience creating the underground archive website, made me think critically about the interaction between media and medium. Specifically how websites interact with he media they are presenting, and how, as the web evolves and we move into a new world of digital humanities, we must go beyond purely archiving to create flexible and intuitive online houses/interfaces for the specific medium of our studies. Because my project focuses heavily on the visual and tactile attributes of the WNUR collection, I wanted to create a website and interface that would present the the record covers in a uniform and eye catching way. I was able to achieve this by modifying the Big Picture theme for Omeka, which in turn affected my decisions regarding the metadata I would collect from each record cover.
Though in my original conception the website was a glance on the route towards days in the record stacks, I found considering the media-medium relationship one of the most important questions I have asked myself in my research thus far. The choices I made creating the eventual home for the WNUR records, shaped the future of my research and continue to direct my focus as the archiving continues. A lesson in research: you never know where learning, answers and rewarding work may arise, as in life in general, the unexpected will shape the future.
With my website in existence and implementation, I was able to focus my day to day work on the task of filling the empty hollow of a database I had created. This work continues to this day, and in its current form reflects the lessons I have learned and the new goals I am working towards.
As with the website, the actual archiving has been a continual learning process. Using the same process of medium consideration I used in creating the website, I decided what information to include as item metadata. I decided to only include information that would be readily available to a WNUR DJ pulling the record from the stacks for their show — only information I could gather from the record sleeve itself. By this rule, I decided to include title, creator, date added to the stacks, music director notes, record label, producer, where the album was recorded and style. Of all these, I acknowledge style is the most unclear from the physicality of the sleeve and the most subjective from DJ to DJ assuming they are informed and discerning rock music listeners. That being said I realized how important it would be to collect information on the style make up of the rock show collection in order to create a cloud of subgenre as a representation of the collection.
*To make a brief aside here, I must remark on how frightening it can be to select the metadata you will collect when archiving objects, there is a certain feeling of inflexibility that comes with the choice, to know whatever data manipulation I could eventually do would be limited to the data fields I had decided to collect at the beginning of my research. In the end it was helpful to apply a meta rule like “only knowledge the DJ could gather” and to cast a wide net to accommodate for the most flexibility possible.*
It was not practical nor scientific for me to listen to every record and make the judgement as to what rock sub-genre the music belonged to. And so, in a search for an objective, succinctly collected and accesible collection of these sorts of judgements, I came across Discogs, a general record database I already use as a music listening enthusiast whose metadata includes a style field for nearly every record in the database. This classification gave me the objectivity and unity I desired for the style field, and further than that provided me with a helpful tool in Discogs to speed the actual process of databasing.
Getting to that actual process, I began by using Omeka’s built in “add item” feature. I should add here Omeka is a website application designed and supported by librarian and media information specialists specifically for archiving processes like mine, you can find out more at Omeka.org. I quickly found this add item feature to be cumbersome and slow, and sought out an alternative. After spending some time, and failing to design a program that would use Discogs as an API resource to mine the basic metadata and limit the information I would enter date added and music director notes (I do hope to one day design something like this if I expand the website and archive to other college radio stations) I cam across a plug-in for Omeka that takes in .csv formatted files a.k.a. excel spreadsheet and converts the rows to items. With this, I found an easy and healthy medium between the cumbersome add item features of Omeka and the more involved, off-track and unrealistic-at-the-moment aspirations of creating an automated program to do much of the work for me. With this in place, I was able to begin the day to day work of filling my database, working through a stack of records at a time, spreadsheet open aside Discogs filing line after line with record information.
The rest of what I do on a typical archiving day has not changed since I began: after entering a portion of records into my spreadsheet I bus them on my skateboard one carton at a time to the library where I scan them and upload the compressed photos to a Dropbox folder for eventual integration on the Omeka platform as I comb through each item uploading a photo and adding a geolocation tag for each. This process of entering on my computer, scanning and combining, makes up the bulk of my days now and has contained its own healthy amount of challenges and setbacks.
The first realization I came to while working through my initial piles of records went something like: wow, there are more records that I thought here! In my project proposal, I had something like 2500 records in mind as the complete rock show vinyl collection, though as a I made it to the letter C and saw my total record count at 600, I quickly realized the total number of records added to the stack in the 80s and 90s in the collection is most likely closer to 6000, which for a single person summer project is nearly impossible, if I am trying to sleep and maintain my sanity. Here I ran up against the distortion field between expectations and reality, and had to adjust my expectations that I would complete the archive of the entire collection, to a more realistic proposition of getting to that 2000 record mark I had initially set out to reach. This adjustment helped me psychologically more than anything, as the pressure I had placed on myself dissipated and I was able to settle into the comfortable pace I have been keeping up the last few weeks. Speaking of psychology, I quickly realized in the first few weeks of archiving that working in the windowless and solitary backroom of the radio station was not setting me up for success in maintaining a positive outlook in the thick of excel spreadsheets. Since then I have transitioned to carrying the record to sunnier open places with more foot traffic and now work faster and with a greater amount of positive energy. In the same vein, I remind myself throughout the day of the larger picture of the archive, the fact that most of the music I am handling is not accounted for on Spotify or other major streaming services. I also look at the website at least once a day to stay focused and energized about the final product I am working to create.
With that, I think you should have pretty good idea of what I have done and the unexpected challenges I have encountered over the past five or so weeks. While some of the things I have learned have been specific to my digital archiving project, I would like to think the more general research bumps in the road are those shared or crossed by researchers everywhere and in that I find comfort. I think I will stop this post here, and leave my future plans and preliminary observations on the records themselves for the future, in shorter, more focused blog posts. Thanks for reading, check back soon here and at undergroundarchiveproject.org for updates!