Oct 03, 2011 — Compatible Data Initiative Highlights Workflows between Archives, Linked Data, and Authors
Jan 15, 2011 — Call for Proposals: NEH Summer 2011 Vectors-CTS Summer Institute on Digital Approaches to American Studies
This document describing a tool developed by the Vectors team was first released in July, 2006.Â The text has been updated to reflect recent changes to the software.
First launched in 2005, Vectors is an international electronic journal dedicated to expanding the potentials of academic publication via emergent and transitional media. Moving well beyond the text-with-pictures format of much electronic scholarly publishing, Vectors brings together visionary scholars with cutting-edge designers and technologists to propose a thorough rethinking of the dynamic relationship of form to content in academic research, focusing on the ways technology shapes, transforms and reconfigures social and cultural relations.Â Vectorsâ€™ fellows are afforded the opportunity to work closely with our design and development team in order to realize new media instantiations of their scholarly projects.Â Projects published in Vectors push the interface of scholarly publishing in exciting new directions, but these â€œfront-endâ€ innovations are largely possible because of the database structures we have also been developing.
One of the most promising and unexpected outcomes of the Vectors project has been the emergence of the database as a critical tool for the future of scholarship, not just as a repository for information but as an intellectual instrument in its own right.Â In an effort to scale the Vectors process and to make the successes of our collaborative endeavors more widely available, we have developed the Dynamic Backend Generator (DBG), an open-source middleware tool that has the potential to reconfigure scholarly endeavors in powerful new ways.Â This tool functions as a kind of middleware â€˜engine,â€™ a robust and flexible database construction kit well suited to scholarly endeavors.Â Databases, of course, are nothing new, having long proved useful in their ability to separate the structure of information from its presentation, allowing that information to be organized and manipulated while still in a state of abstraction.Â The uniqueness of the DBG is that it begins to harness the power of the database for next-generation scholarship in the humanities and qualitative social sciences.
In the course of their research, scholars in humanities fields are constantly experimenting with different ways of organizing knowledge and understanding.Â This process occurs both within the scholar’s mind and through the various kinds of writing undertaken as part of the activities of a career.Â Scholarly writing, however, must necessarily take the form of descriptions of ideas and concepts elaborated with enough skill and attention to detail to successfully communicate them to colleagues.Â A database does not need to describe ideas, however, because it can embody or enact them, much as a sketch embodies or enacts the ideation of a visual artist.
A sketch is where visual ideas are worked out in concrete form; the light touch of its technique allows for flexibility, iteration and constructive revision.Â It is the “reality check” for the artist, and a realm where visual ideas are refined until they are ready to be executed in more permanent forms.Â Similarly, an easy-to-use database authoring tool could function as a kind of intellectual sketch pad, allowing scholars to easily experiment with various ways of organizing their ideas until arriving at a structure that could serve as the foundation for future work.
Many of the Vectors fellows have found that the fusion of scholarly writing with database practice has resulted not only in a deeper understanding of or new approaches to their own research, but also in a scholarly endeavor that proved to be one of the most intellectually satisfying of their careers.Â They have discovered new contours and nuances in their work through the restructuring afforded by a database environment and have also experimented with new forms of argument and expression.Â The ability to deploy new experiential, emotional, and even tactile aspects of argument and expression has opened up fresh avenues of inquiry and research.Â These reasons alone make it important to pursue the development of tools that can bring their experiences into the world of academia at large.Â Currently, there are few easy-to-use, open-source database design tools widely available to humanities scholars.Â The DGB has the potential to demystify information architecture for a diverse array of scholars.
The obstacles are significant. Databases have traditionally been the domain of the computer specialist, the programmer or IT expert, and so the interfaces by which they are created and maintained have remained generally impenetrable to those lacking in such skills.Â The problem is more than one of interface design, howeverâ€”even if we were to redesign the typical administrative interface to an Apple computer level of usability, we still would not have addressed the central fact: that scholars will and should approach database development in significantly different ways from their computer scientist counterparts.
Computer scientists approach the database authoring tool as a means to the end of implementation.Â Â It is a site where pre-existing plans are executed according to well-defined methodologies that determine how a database should be organized and structured.Â What we are proposing, however, and what the scholars we have worked with have found most enlightening, is re-imagining the database authoring tool as a site where writing, exploration, and ideation take place.
An authoring tool conceived according to this principle will aid the scholar in translating their conceptual formations into database architecture.Â The scholar will be queried via the tool about the basic intellectual components of their research, be they specific works or authors, schools of thought, metaphors, citations, or other concepts.Â Each one of these is then created as its own category of information, initially independent of the others.Â The scholar can then embellish each of these categories with as much or as little detail as required.Â For example, a work might require basic information like title, author, date, publisher, etc., while a metaphor might only require a title and a brief description.Â The process is deceptively simple, amounting to the scholar deciding what kinds of information are important to them at this point in his research and using the tool to make space in the database for that information to reside.Â Any of these choices can be revisited at any point should the scholar’s understanding of the structure of their research change.
In the next step, the scholar will be prompted to start defining the kinds of links that can exist between the various categories of information just established.Â For example, a concept might need to be linked to one or more citations, but each citation can reference one and only one work.Â Once the scholar is happy with the conceptual structure that’s been established in the database, she can begin writing and entering data.Â As in a traditional writing project, some of the information will be indexicalâ€”identifying works, sources, dates and suchâ€”while other information will be more expansive and conceptual, speaking to the themes and topics of the research itself.Â Much of the information will likely be text, but the tool supports the inclusion of images, links, video, and audio as well.Â The structure of the database and its contents will be displayed and can even be edited graphically to enhance ease of use.
During their research projects, scholars generally accumulate a large collection of notes, media, and writing articulating a point of view within their area of expertise.Â With the tool we are proposing, those materials, and most importantly, the structures within which the scholar chooses to organize them, acquire a machine-readable form.Â This affords the scholar portability and flexibility, not to mention the reports from Vectorsâ€™ scholars that database practice actually enhances the depth, reach, and rigor of their work.Â The process allows a rethinking of the relationship of form to content in scholarly production and opens up the possibility for new genres of research and publication.
The kinds of databases created by the DBG are already the lingua franca of the Internet, information resources that can be deployed in any number of waysâ€”as news feeds, blogs, monographs, or works like those published in Vectors. A variety of tools and methodologies are already in existence to enable these outward, front-end manifestationsâ€”but they all depend on database engines which to a large extent have become the engines of culture at large.Â Our intent is to make sure that the needs and demands of humanistic scholarly research are accounted for in the design and creation of those engines, while also continuing our exploration of the potential for various â€˜front-endâ€™ innovations in scholarly publishing.
History, Technical Aspects, and Potential Development of the DBG
In the first two issues of Vectors, each fellowâ€™s project was developed on a one-off basis of sustained iteration between the scholar and our design team.Â While this process provided the scholar with an intense and meaningful collaboration process, we soon came to realize shared database needs across a variety of collaborations and also desired a more scalable and general-purpose development tool.Â In order to be cost-effective and to allow for maximum re-use, a tool was required that met the needs of many different project architectures with little or no customization by the Vectors development team. We began production of the Dynamic Backend Generator tool in Summer, 2005 and have since successfully tested and refined it on over twenty of Vectorsâ€™ database-driven web sites.
The DBG is a dynamically-generated middleware interface to a database, designed to help academic users conceptualize data and database structures in relation to the nature of their content and a variety of possible front-end display(s).Â The software is driven by PHP, HTML, and Ajax.Â Â Based on a set of relational table matching rules, the software generates a formatted, visual representation of a MySQL database and an interactive, dynamically-generated tool for entering and managing rich database content.
In its current form, the DBG could be considered a complement to and an extension of the already existing PHPMyAdmin (PMA) software.Â Â PMA, an open-sourced MySQL management tool, allows a user to administer a MySQL database using the web browser.Â There are many administration features of PMA not presently including in the DBG.Â However, the DBG goes further than PMA in regard to relational table matching, creating relationships between tables based on a set of architectural rules.Â Users add, edit, or delete content that acts on multiple database tables, providing a quick way to enter and traverse data from multiple tables.Â When viewing the table content, for instance, data from relational tables is combined and displayed as one html table (figure 1).Â Furthermore, when editing a table row, relational tables are formed into HTML select boxes and other interfaces.Â This eliminates the need to move back and forth between tables to â€œpick outâ€ content and add to the root table (figure 2 & 3).
As noted above, we have discovered an unanticipated benefit of the tool for scholarly production in testing on several Vectorsâ€™ projects.Â New media projects require ideas or core concepts to be structured in a relational model (data stored is mathematically related to other data), and the database tool in its current version helps scholars re-conceptualize their rich content into this model.Â The DBG has drastically improved the development time of Vectorsâ€™ projects, but it has also allowed Vectorsâ€™ fellows and developers to re-imagine and recreate their scholarship in new expressive modes by using the tool as a canvas for idea management.Â Put differently, we have learned that backend architectures, properly managed and accessed, can become a content-design or theory-making tool.
To heighten its role as an intellectual sketch pad and tool kit, future versions of the software will include new or improved functionalities.Â These include creating advanced graphical representations and interactions for the user; further developing the methods for placing information into the relational model; moving the output to the point where the tool could act as an external â€œread-onlyâ€ web site; and also working in concert with the semantic information research being performed in the â€œblogsphereâ€ and â€œWeb 2â€ communities. The DBG, partnered with a â€œWeb 2â€ and semantic model, could in the future support work with â€œkey word cloudsâ€, semantic databases, and other new forms of information storage and retrieval research currently in production at major research universities.
The DBG was principally architected by Vectors Info Design Director Craig Dietrich with Creative Directors Erik Loyer and Reaegan Kelley, editorial overview by Tara McPherson and Steve Anderson, and with the collaboration of Vectors Fellows.
— Vectors Journal, May 19th, 2009
Leave a Reply