DigitalTexts2

This site documents the Digital Texts 2.0 project funded by SSHRC through the Research Development Initiatives program. This project is the companion to the Mashing Texts project.

Researchers

• Stéfan Sinclair, principal investigator
• Geoffrey Rockwell, co-investigator
• Stan Ruecker, co-investigator
• Amanda Etches-Johnson, collaborator
• Krista Godfrey, collaborator

Developers

• Stéfan Sinclair
• Johnny Rodgers

Development Notes

Development Notes & Screenshots (last updated Feb 27th, 2008)

Description

[This sections is from the grant application - additional edits are still required for the wiki formatting]

Context

The term “Web 2.0” has been used since 2004 to broadly describe a second generation of web sites with two distinguishing features: 1) they usually function more like desktop applications than a simple medium for delivering regular hypertexual content, and 2) they often emphasize collaboration and networking over individual browsing (O'Reilly 2005). Although there is considerable disagreement about the exact nature of “Web 2.0” (Bray 2005, Berners-Lee 2005), most recognize that in the past five years there has been an explosion of sophisticated online resources for collaborating and exchanging digital content (Boyd 2007, Levy et al. 2007). Common examples of “Web 2.0” social networking applications include del.icio.us (for sharing bookmarks) Flickr (for digital images), YouTube^? (for video clips), and Facebook (for personal profiles). The way in which many people find, manage and exchange their informational assets has been fundamentally transformed.

Despite this transformation in digital society as a whole, common practices in digital humanities scholarship have remained relatively unchanged, especially with respect to electronic texts. As several recent reports have identified (Huxley 2007, ACLS 2006, Brockman et al. 2001), the adoption of digital texts by humanities scholars has been relatively limited. This situation is somewhat puzzling given that, since the early 1990's, governments and funding agencies have invested significantly in the creation of high-quality digital text collections, many of which are based on best practices and standards such as the Text Encoding Initiative. Digital texts are being created but they are not being extensively used.

Though this situation can be explained in part by the mechanics of academic training and research (shifts in methodology take time to work their way through generations of professors and students, Graff 1987), we believe there is another determining factor involved: systems for searching, browsing, reading, managing, and analyzing digital texts are largely predicated on first generation ways of working with digital media, and even print content.

Opportunity

Digital Texts 2.0 is an initiative to experiment with applying the principles of Web 2.0 to the realm of electronic texts. We intend to preserve and expose all of the existing qualities of digital texts (rich hypertextual associations, refined encoding practices, analytic affordances, etc.), while enhancing them with additional characteristics provided by Web 2.0 and social networking.

For instance, rather than maintain associations exclusively through embedded hypertextual links (as many HTML and XML documents do), Digital Texts 2.0 will enable innumerable simultaneous associations through social tagging and “folksonomies” (users who label their texts using a common vocabulary). Similarly, rather than merely exist as autonomous, free-floating objects in cyberspace (at best as part of some digital collection), Digital Texts 2.0 will always be associated with other humans, and thus inherit all of the interest and knowledge that those human associations enable. As an example, a reader may be interested in viewing a list of texts associated with another reader, or perhaps conduct a hybrid search that combines the traits of authors and of their readers (e.g. all of the female authors tagged by male readers between 40 and 45 years old). This is an innovative way of conceptualizing texts: as inherently associated to “consumers” and “producers” of those texts.

Finally, whereas much attention has been focused on analytic procedures for digital texts, we hypothesize that creating a social context for electronic documents will encourage different uses, more oriented towards the permutability, remediability, and recombinant nature of digital texts (McCullough^? 1998, Bolter 2000, Ryan 2005). Not surprisingly, the “Web 2.0” has a specific term for some of these kinds of metamorphoses – “mashups” (Hof 2005) – which describe resources that incorporate and adapt components from other resources. The ability to mix and match multiple digital texts from different collections would surely open up new forms of textual scholarship.

Though deliberately simplified, the following table is useful to understand the contrast between existing conceptions of digital texts and the innovative conceptions that we are proposing:

These characteristics are not binaries: we envision Digital Texts 2.0 as representing a cumulative, evolutionary change that maintains the traditional strengths of digital texts. It is important to emphasize that we are not proposing essential changes to digital texts themselves, but rather, to ways in which we as humanities scholars interact with them.

Our objective is to experiment with social networking within the culture of humanities research and teaching. Some of the specific questions that we will have of our colleagues and students are as follows:

• what types of texts do you feel comfortable sharing?
• what types of comments would you make on texts from other users?
• what might be the benefits of hybrid searches between texts and their readers?
• what other media would be useful to include in this type of social networking environment?
• what analytic functionality would be useful to include?
• how could you reuse other people's texts?

Our current conception of Digital Texts 2.0 is heavily influenced by several successful social networking sites, including del.icio.us, Flickr, YouTube^?, and Facebook. In fact, there already exists a website for uploading and sharing documents: Scribd.com, though its functionality is relatively limited and the site's design is not particularly conducive to humanities scholarship (the commoditization and relative thematic chaos may not be appealing to many academics).

The following sketch provides a schematic visual representation of what the Digital Texts 2.0 interface might resemble (this is a simple adaptation of the Facebook user interface – it is unlikely that the Digital Texts 2.0 interfaces would so closely resemble an existing resource):

Figure 1: Sketch of Digital Texts 2.0 (as a simple adaptation of the Facebook Interface)

Note in particular the existence of digital texts, collections and tools along side of other digital assets in the left column. The right column suggests some of the flexible searching criteria that might be included, such as traits of the author and reader, format and copyright of the digital text, etc.

In addition to the social networking component of Digital Text 2.0, we have the capacity to readily integrate a range of computer-based analytic tools, through services provided by HyperPo, TaporWare^?, and the TAPoR Portal. These analytic tools can help realize the full potential of digital texts by framing them within a web application (Web 2.0) and by enabling operations that are not possible with print texts. The following is a sampling of some of the analytic “recipes” that can have been described on the Text Analysis Developers' Alliance wiki (created and managed by Sinclair):

• identify simple themes in a text
• explore dynamically aggregated text
• compare texts to verify authorship
• build a social network map from texts
• explore diachronic characteristics of an author's corpus

An interesting by-product of integrating tools within the Digital Texts 2.0 framework is that many kinds of analytic results can be considered texts to be shared and annotated in their own right. For instance, a tool that dynamically aggregates several disparate texts together based on a search term is in fact creating a new digital text, which can enter into the social networking ecology.

Methodology and Timeframe

The research life cycle for Digital Texts 2.0 will include the following stages:

• establishing theoretical foundations
• theorizing the innovative components (social networking of digital texts)
• sketching static interface designs (in Photoshop or Illustrator)
• dynamic sketches (likely animations in Flash)
• environmental scan of implementation technologies (PHP, Ruby, Java, etc.)
• implementation of functional prototypes
• preliminary usability assessment

It is worth noting that in the condensed timeframe of the RDI grant, most of these stages will be shorter than normal (with the exception of the complete literature review). In particular, a full-scale proposal would include more formal testing and usability components, as well as additional iterations of design and implementation. However, in the context of this RDI, we will focus on producing a small number of functional prototypes that will allow us to assess the viability of a more extensive Digital Texts 2.0 platform.

One of the primary objectives of Digital Texts 2.0 is to conduct a thorough literature review that will provide a foundation for subsequent phases of the project (beyond this RDI). This research component of the project will touch on historical, theoretical, popular, and technical aspects, all of which will be made more widely available through a project wiki. Content will include:

• research into social networking
  • social networking prior to and outside of a web-based context
  • social networking as a popular trend
  • the effectiveness of social networking for various tasks
• implementation solutions
  • appropriateness and usability of different interface designs
  • back-end (server-side) solutions (PHP+MySQL, Ruby on Rails, Java, etc.)
  • user interface (client-side) solutions (AJAX, Flash, etc.)
• potential for integration with other resources
  • other social networking sites (del.icio.us and Facebook, etc.)
  • library catalogues and information systems
  • text analysis tools (HyperPo, TAPoR Portal, etc.)
• users
  • identifying user communities (especially in the humanities)
  • identifying the needs of communities
  • user feedback on different prototypes

In addition to the technical research, the Digital Texts 2.0 project will deliver a small number of actual prototypes. The development of these prototypes has two main purposes: 1) to provide concrete interfaces for preliminary user feedback; 2) to identify the most promising technologies for a full implementation. The prototypes will not be meant for general use.

Although much of the work will happen concurrently throughout the twelve months of the project, the following provides a more detailed timeline:

• Month 0
  • initial project planning
• Month 1-4
  • preliminary research in all areas
  • preliminary interface sketches
  • theorizing innovative approach
• Month 5-8
  • initial prototype implementations
  • refinements of interface designs
  • viability of social networking in the humanities
• Month 9-11
  • preliminary user feedback
  • prototype refinements
  • integration with other resources
  • complete wiki documentation
• Month 12
  • final prototype revisions
  • complete wiki documentation
  • prepare for follow-up grant

Research Development

The Digital Texts 2.0 project is a preliminary attempt to better understand the phenomenon of social networking and how it might be adapted to benefit the ways in which humanities scholars interact with electronic texts. A key first step in the process will be an extensive literature review of social networking, both in an academic and non-academic context, as well as a survey of current uses and needs of humanities scholars and students. This preliminary work will directly inform the nature of the prototypes that we create, and will be of immense value for a subsequent phase of the research project (beyond this RDI), when development of a more complete system is undertaken.

This project is situated at a critical crossroads between two research agendas. The first research orientation is past work on developing web-based text analysis tools, exemplified by existing resources like HyperPo, TAPoR, and TaporWare^?. While these tools provide extremely useful functionality to a certain subset of more experienced users, the underlying conception of both texts and tools is relatively traditional: resources exist independently of users and there are few affordances for allowing users to share and comment on work being accomplished.

The second research orientation is what is envisioned for future work: the creation of environments that foster collaboration and networking through the meaningful exchange and annotation (commenting) of digital objects. Humanities scholars are accustomed to sharing ideas through traditional modes of scholarship, including conferences and publications. Digital Texts 2.0 is about experimenting with using primary materials (such as digital texts) directly as a vehicle for increased collaboration and knowledge production.

The research described in this proposal is both preliminary and experimental: it would be premature to advance this research agenda in the context of a regular grant proposal, without first acquiring a better sense of this new terrain (of social networking) and without first experimenting with possible technical approaches. We anticipate that Digital Texts 2.0 will help us establish the foundations for future grant proposals, including a SSHRC Standard Research Grant, and possibly a successor to the TAPoR project with a much greater emphasis on social networking.

Suggestions

Feel free to delete/repurpose this section, but I didn't see a better place to comment and have suggestions for further development maintained.

• Would it be useful to include some sort of quick personal stats to the app? I am imagining something down the line of ... you have X number of texts, possibly a quick chart showing number added and when (a timeline), comments contributed, comments received, others...?

-- ShawnDay - 23 Mar 2008