Collaboratively written by Mithun Shetty, Kim Arena, and Sheel Patel
The efficacy of a digital humanities project can be vastly improved if the delivery and interface is thoughtfully designed and skillfully executed. The following two websites, “Speech Accent Archive”, and “10 PRINT eBooks”, both utilize non-traditional forms of displaying content that alter the experience of their internet audience. Both projects will be critically assessed according to the guidelines described by Shannon Mattern’s “Evaluating Multimodal Work, Revisited” and Brian Croxall and Ryan Cordell’s “Collaborative Digital Humanities Project.”
The Speech Accent Archive is an online archive of audio recordings that contains multiple speakers from a variety of regional and ethnic backgrounds dictating a specific set of sentences. These recordings are submitted by the public and reviewed by the project administrators before being added to the site. The purpose of this media element is to expose users to the phonetic particularities of different global accents. This website is useful because it provides insight regarding the various factors that affect the way people talk and how these factors interconnect – from their ethnic background to their proximity to other countries. This site proves to be a useful tool for actors, academics, speech recognition software designers, people with general appreciation for the cultural connections in languages, and/or anyone studying linguistics, phonetics, or global accents.
The website’s layout is ideal for accomplishing this purpose: users can browse the archive by language/speakers, atlas/regions, or can browse a native phonetic inventory. This allows users to explore accents on a regional basis which makes it easier to see similarities between local dialects. The audio recordings are all accompanied by a phonological transcription, showing the breakdown of consonantal and vowel changes as well as syllable structure of the passage. Each user submission is accompanied by personal data, including the speakers’ background, ethnicity, native language, age, and experience with the English language. The site also has a very comprehensive search feature which has many demographical search options, ranging from ethnic and personal background to speaking and linguistic generalization data. This level of detail is an invaluable resource for those who study cultural anthropology, phonetics, languages, and other areas, as it allows for a specific manipulation of the data presented. Also, the quality of user contributions is consistently high – it is very easy to follow the playback of the recordings.
However, the project does have its limitations as well. The passage being read by the contributors is in English, no matter the speaker’s fluency or familiarity with the language. Pronunciations of this passage may not reflect the natural sound of the languages represented. Further, because the audio samples are user-contributed, it is hard to maintain a constant of English fluency among contributors. Another limitation to the site is that many of the sections of the site have little to no recordings or data; this is merely due to a lack of user contributions, but could be resolved by website promotion. The project is still ongoing, thus the database will continue to grow as time goes on. Another limitation of the site is that it lacks any sort of comparison algorithm. The accents are all stored on their own specific web pages and load via individual Quicktime audio scripts; consequently, it is very difficult to perform a side by side comparison of accent recordings. As a result, the project is not really making any conclusions or arguments with this data. This could be improved by allowing users to stream two separate files at the same time, or by allowing a statistical comparison of the demographical information accompanying each recording. It would also be interesting if an algorithm or visualization could be created that could recognize the slight differences in the voices and arrange them based on similarity along with the demographical data that accompanies the voice sample. Further, the project could establish a tree-like comparison of regions and accents, visually representing the divergences and connections between where people live or have lived and the way that they speak.
With these additions, it would be easier to aurally understand the effects of background or ethnicity on speech accents. Still, this website shines albeit these setbacks. The project offers a tremendous amount of data in an organized manner, presenting many opportunities for further research and applications of the information. This level of detail is an invaluable resource for those who study cultural anthropology, phonetics, languages, and much more.
The book titled 10 PRINT CHR$(205.5+RND(1)); : GOTO 10 is a collaboratively written book that describes the discovery and deeper meaning behind the eponymous maze building program created for the Commodore 64. The book can be seen as a way to look at code, not just as a functional working line of characters, but also as a medium of holding culture. 10 PRINT CHR$(205.5+RND(1)); : GOTO 10 uses the code as a jumping point to talk about computer programming in modern culture and the randomness/ unpredictability of computer programming and art. The book explores how computation and digital media have transformed culture. Along with this book, one of the authors, Mark Sample, created a “twitterbot” that uses a Markov Chain algorithm to produce and send tweets. The @10print_ebooks twitterbot takes the probability that one word will follow another, scans the entire book, and tweets the random phrases it generates.
The clear goal of this book is to demonstrate that reading and programming coding does not have to always be looked at in the two-dimensional, functional sense that many people see it as. The authors argue that code can be read and analyzed just like a book. They do so by delving into how this 10 Print program was created, its history, its porting to other languages, its assumptions, and why that all matters. They also talk about the randomness of both computing and art, and use this 10 Print program as a lens through which to view these broader topics. The purpose of this book is stated very clearly by one of the co-authors, Mark Sample: “Undertaking a close study of 10 Print as a cultural artifact can be as fruitful as close readings of other telling cultural artifacts have been.”
The implementation and format of this book and twitterbot is a little difficult to understand and doesn’t necessarily help them portray and establish their goals, especially when talking about the twitterbot. The book itself is coauthored by 10 professors who are literary/cultural/media theorists whose main research topics are gaming, platform studies, and code studies, which gives a broad range of perspectives regarding the topics. It also dives into the fact that code, just like a book, can be co-authored and can incorporate the views and ideas of more than one person. This idea draws on the parallels that the authors are trying to draw between coding and literary elements. Code is not just one-dimensional; it can incorporate the creative and artistic ideas of many people and can achieve many different forms that often have very similar functions in the end. In this sense, the co-authoring of this book inherently showcases their main message regarding code and how it should be viewed. The book also progressively talks about the history of this Basic program, and how it coincided with cultural changes due to the advent of the personal computer. Sample’s twitterbot, on the other hand, leaves the user more often confused than educated, but that may be point. Using the algorithm it spits out random, syntactically correct sentences that sometimes mean absolutely nothing, but also occasionally it creates coherent thoughts from the words in the book. The occasional coherent sentence that the bot spits out may be a demonstration of code itself. The user may see that within jumbles of code or in the case of the book, words, and meanings can be pulled if put in the correct syntax. Also, the form definitely fits. The randomness of the twitterbot allows people to see that even by coincidence there can be substance to code. If it was done having people point out specific parts of the code then we would be limited to their interpretations. Having a machine randomly spew out phrases allows for many different interpretations.
This tool, although abstractly useful could be implemented much better. If the twitter bots are “occasionally” genius, then the website would be more efficient if it had implemented some sort of ranking system for the most interesting or coincidental tweets. If they had some sort of sorting mechanism, then the project may be more convincing in saying that code can be made to have a creative license or brand.
Regardless of the various limitations both projects may have shown, it is abundantly clear that their media elements vastly improve their ability to illustrate their ideas and accomplish their purposes. It would be practically impossible to illustrate these projects with text alone. The Speech Accent Archives’ audio recordings give concrete examples to an entirely aural concept, which is infinitely more useful than simply listing the phonetic transcriptions. The 10print Ebooks’ twitterBot, while difficult to understand, is an interesting concept that also generates concrete examples of what the project is trying to illustrate – that code is multidimensional in its structure and can be interpreted and analyzed similar to a complex literary work.
Sources:
@10PRINT_ebooks, “10 PRINT ebooks”. Twitter.com. Web. https://twitter.com/10PRINT_ebooks
Baudoin, Patsy; Bell, John; Bogost, Ian; Douglass, Jeremy; Marino, Mark C.; Mateas, Michael; Montfort, Nick; Reas, Casey; Sample, Mark; Vawter, Noah. 10 PRINT CHR$(205.5+RND(1)); : GOTO 10. November 2012. Web. http://10print.org/
Cordell, Ryan, and Brian Croxall. “Technologies of Text (S12).” Technologies of Text S12. N.p., n.d. Web. 15 Sept. 2013. http://ryan.cordells.us/s12tot/assignments/
Mattern, Shannon C. “Evaluating Multimodal Work, Revisited.” » Journal of Digital Humanities. N.p., 28 Aug. 2012. Web. 15 Sept. 2013. http://journalofdigitalhumanities.org/1-4/evaluating-multimodal-work-revisited-by-shannon-mattern/
Weinberg, Stephen H. The Speech Accent Archive. 2012. Web. http://accent.gmu.edu/about.php