Spoken Web 2.0
Research Notes and Observations
Research compiled by Annie Murray, Jared Wiercinski and Celyn Harding-Jones
Research has reinforced our sense that there is a hole in the market in terms of audio poetry-related sites that are useful for scholarly researchers. Most sites provide the audio file on a very rudimentary web interface that doesn’t take into account the ways that the researcher would use audio for their research or the new technologies available online.
- Most sites don’t tailor their functions and interface to the types of audio it contains. For example, it is often hard to find out if the audio is part of a specific event in history or part of a larger collection because audio files are treated as singular files, or each recording is broken down into so many sections that even listening to the whole recording becomes unwieldy.
- Most sites don’t tailor to their target audience. Many sites assume to be able to listen and browse audio files is enough, but they haven’t considered that users need or want context, indexes, the whole file or clips of the whole. Options are often limited, and audio files are often treated as if they appeared out of nowhere, and are left to float in cyberspace.
- Sometimes new technologies added to the site don’t contribute to the audio experience, or worse, they complicate with useless information. For example, Slate.com’s audio poetry site is full of sharing, rss, print, recommend, reprint and discuss forums that aren’t tailored to the audio, but to the magazine’s general site. Facebook ‘like’s may not be necessary for research where a note-taking plug-in may be invaluable.
- One issue that persisted during my research was how does each site paint a picture of the reading series or the provenance of the audio file? How does each audio recording as a single contribute to context of the whole? Kootenay School of Writing (KSW Audio), Kelly Writer’s House, Aspen Magazine, Faulkner at Virginia, 28th Maori Battalion, University of Chicago’s Poem Present, and UC Berkeley’s sites all use context and the series of audio files to create a coherent and informative research tool.
- Bigger and more general websites may be useful for issues pertaining to layout, plug-ins, new technologies and sharing issues. See TedTalks, National Film Board, UbuWed and Youtube.
- Transcripts: There are only a few websites that contain transcriptions to their audio files. While a handful of sites provide the poem’s text along with the recording, these are only the published version of the poem and do not take into account any changes that the author made to the poem during the reading. Nor does it include any transcribed extra-poetic speech, comments or unpublished versions of poems that make the recording unique are impossible to track and are not indexable. The issue of indexing—an invaluable tool for researchers in print resources—is all-important in my mind. As an undergrad and now graduate student, I have been hesitant to listen to hours of recordings for research as I am unsure that I will find the information I am seeking. The Faulkner at Virginia website has transcriptions of extra-poetic speech that provide very important insights into his writing. Not only can we find out how he pronounced his fictional county ‘Yoknapatawpha’, but during question and answer periods, he provides advice for young writers and his own interpretations of his works.
- The ways in which the audio file is cut (or not cut) into individual poems makes a big difference to the way listeners and researchers approach the recording. Sometimes audio files are too big or too long to listen to completely, and need to be cut into sections. However, when audio files are cut, there is a danger of loosing extra-poetic commentary that is made before or after the poem—often a poet will speak about an older poem in an introduction to a new poem. A researcher or listener might not want to have to download or stream 23 different files to listen to one whole recording. Perhaps a large range of listening options is best (individual poems, complete recordings, parts 1, etc.).
- Useful features of other sites:
- Description of archive and our mission statement: what are you looking at, listening to and why.
- Registered users can contribute to the site in some way—way to reduce too many random comments (YouTube often suffers from too many off-topic comments)
- Links to stable sources for bibliographies and other sources of information are useful for context.
- Files organized (or tagged) by date, historical contextual relevance, geographic locations as well as author and indexable subjects
- Discovery features such as Recommended or Related audio files and archives
- Layout: Minimalist vs. Multi-Optioned. UbuWeb has a very minimalist aesthetic, which often comes across as an endless wall of archived material, yet it suits the nature of UbuWeb as forever expanding. The inverse is a website that has too many searching and browsing options in an attempt to make discoverability easier can become harder to navigate. For example, Off the Page has several browsing options by categories—however not all of the recordings are included in each category. This site is a good example of how a project similar to ours could fall short of its potential.
- Play-back options. Players that open in a black or blank screen that don’t open a new tab or window are frustrating because you leave the metadata and the main page. Pop-ups are okay, but when listening to many audio files, they can get lost behind windows.
- The Sir George Williams University Poetry Reading Series doesn’t have much visual or photographic data—something to keep in mind when creating our site.
- Pedagogical guides are fun and encourage teachers to use sound recordings in their courses.
- Perspective: Distribution or Archiving function of the site vs. Interactive research-based site.
- Audience: Researchers, scholars/ students/ casual listeners
- Playlists: by theme, subject, historical groupings can help contextualize (or decontextualize)
- Important to link up site to other similar sites: EPC, Penn, UbuWeb, etc.