Notes on Distributed Archives: 2 of 3

The ongoing or continuous, as opposed to discrete and final, nature of web-based publication in comparison with traditional publication venues (specifically the scholarly edition), as well as the reader expectations aroused by the milieu of the web, point to the need for distributed authorship of web-based archives. The key advantage to distributed authorship is its suitability to a publication process that is ongoing to the point that it can never be considered final. A couple of examples follow.

T. O. Mabbott and Edgar Allan Poe

Thomas Ollive Mabbott’s Collected Works of Edgar Allan Poe is a sterling example of both the excellences and limitations of the traditional scholarly edition. Mabbott, the dean of Poe Studies in the decades following the second World War, spent the majority of his scholarly life working on a monumental edition of Poe’s works. Exhaustively consulting all known variants for each extant Poe text, and uncovering several new works in the process, Mabbott produced THE authoritative edition of Poe most frequently cited in published research. Sadly, such was the extent of Dr. Mabbott’s undertaking that he passed away prior to the completion of his envisioned project.

That being said, Joseph J. Moldenhaur rightly observes Mabbott’s work is not definitive. In addition to the problems noted by Moldenhaur concerning Mabbott’s choice of copy texts, the discovery of new Poe texts, new historical information concerning Poe, and new editorial approaches would have ensured that, 40 years after its initial publication, the Collected Works of Edgar Allan Poe would be in need of at least an addendum if not a more thorough re-editing and republication. Such is the nature of the traditional scholarly edition – the creation of static monuments to an editing process that is, by definition, dynamic.

Of course, writing in the middle decades of the prior century, T. O. Mabbott had only one choice available to him for his editorial project, a traditional scholarly edition. But, even had Dr. Mabbott lived an additional 40 years, he would have faced a similar project due to the fact that his project was the work of a solitary individual with finite resources (time, money) and research interests.

The Web of American Transcendentalism

As a scholar working at Virginia Commonwealth University in the field of pre-Civil War United States literature and culture I routinely receive inquiries concerning The Web of American Transcendentalism. This site was started in 1999 by Ann Woodlief, an associate professor at VCU, as a vehicle conducting student research and disseminating primary texts and background information concerning the authors associated with American transcendentalism. At the height of its popularity, 2002-2004, the site was drawing in the neighborhood of 9000 visitors a month and was the recipient of a positive write-up in the Chronicle of Higher Education.

My usual reply to inquiries regarding the Web of American Transcendentalism is to note that I’m not affiliated with the project and to suggest that the person try to contact Ms. Woodlief directly. I’ve never met Ann Woodlief. My position at VCU, in fact, came open, in part, due to Ms. Woodlief’s retirement. Upon my arrival at VCU I found no files or documentation regarding the site. This puts me in the odd position of being a faculty member in an English department which hosts a digital resource directly related to my field of research about which I know almost nothing.

This is not to criticize Ms. Woodlief or my department for failing to leave me with some sort of owner’s manual for the site. Although, as a digital archive, the Web of American Transcendentalism is capable of continuous updating, neither Ann Woodlief nor my department could assume that their new hire would be interested in continuing the project. In fact, given that The Web of American Transcendentalism was “authored” by someone else, lacking the tradition of co-publication in the humanities, and give the risks associated with digital publication for junior faculty, they could safely assume that I would be weary of contributing my time to the continuation of the project.

Although it is a digital archive, because the Web of American Transcendentalism was organized as the project of a single author – as opposed to an editorial board or some similar institutional authorship arrangement – it is as “dead” as any scholarly edition in VCU’s library. No new content will be added to the archive nor will known errata be corrected.

Moreover, because the Web of American Transcendentalism is an online project, it is subject to special forms of decay to which traditional bound texts are immune. The design of the page is, in 2009, severely dated. This is not just a matter of aesthetics. The gray background with hard-to-read menus makes the side difficult to navigate. Information is hard to fine. And, because the site was laid-out using hidden tables and no Cascading Tile Sheets, there is no easy way to address these issues. More troubling is that fact that, due to various server migrations, portions of the site are no longer operable. While it appears to have been Ms. Woodlief’s intention to allow future visitors to contribute annotations to some of the study texts, the annotation feature no longer works.

Given that many of the texts provided by the Web of American Transcendentalism are available elsewhere on the web, the visual cues provided the presence of a dated design and broken links in all probability discourages likely users from taking advantage of the information provided by the site. While in a physical library an older edition will always find temporary use if the revised edition is checked-out, on the web the revised edition is always available.

An Alternative: Collaborative, Distributed Authorship

The move from paper to digital publication for scholarly editions must be attended by a movement from individual to collaborative, distributed authorship. Such a movement will ensure that new scholarly editions will take advantage of the opportunities for continual publication and revision provided by digital media. Later this week I’ll sketch out what such an authorship model might look life for a Distributed Poe Archive.

Notes on Distributed Archives: 1 of 3

I’ve been thinking a lot about “cloud computing” and what it might mean for the future of the traditional lab-centric model of digital humanities research. In what follows, I’d like to make a few observations about the historical function of lab-centric digital humanities projects en route to sketching an outline of what a distributed digital humanities archive might look like.

The Lab-Based Archive

Unsurprisingly, most of the pioneering works of digital humanities scholarship were and are associated with pioneering digital humanities labs: The University of Virginia’s Institute for Advanced Technology in the Humanities (IATH), The University of Maryland and the Maryland Institute for Technology in the Humanities (MITH), Brown University’s Scholarly Technology Group (STG), and the University of Nebraska’s Center for Digital Research in the Humanities (CDRH). The close relationship between projects such as the Walt Whitman Archive, The Dickinson Electronic Archives, and the Woman Writers Project and these laboratory environments is only natural given the needed conjunction of scholarly vision, technological expertise, and institutional funding required to bring these projects to fruition.

Fortunately, and in large-part due to the success of these pioneer projects, the technological expertise required by such projects is now much more broadly distributed throughout humanities departments (in many cases by the hiring of faculty who contributed to these students as graduate students). Moreover, the last few years have seen steep declines in the cost of scanning, imaging, and digital retrieval technologies associated with these projects.

And while the centralization of resources and aspiring young digital humanists around these centers of digital humanities development was crucial in helping digital humanities achieve the necessary critical mass for its larger acceptance in the humanities departments, it seems as though a broader distribution of digital humanities training and participation is desirable for the health of the digital humanities relative to both the larger higher education community and the culture as a whole. The more universities, scholars, and students there are participating in the digital humanities, the more likely that students, administrators, and assorted funding agencies will come to understand the work accomplished by digital humanities scholars and projects.

The Distributed Archive

As its name implies, a distributed archive would be less strongly affiliated with a particular lab or digital humanities initiative. As a “born digital” project, such an archive relies on the presence of a robust digital infrastructure (including pervasive high-speed internet access, ready access to production equipment, and the availability of digitally-oriented humanities scholars) to allow for allow for scholars from a wide variety of institutions (including independent scholars) to collaborate on digital humanities projects. Such a model underlies John Bryant’s plans for the Melville Electronic Library as well as the Virtual Humanities Lab at Brown University.

The benefits of such a distributed archive are several:

• Greater diversity of contributors. The contributor base of such an archive is not limited to the faculty or graduate students associated with a particular department. In particular, faculty, students, and individuals who are not currently associated with one of the prominent digital humanities labs or working groups, but who possess valuable content or methodological insights, will be able to contribute to the archive in significant and meaningful ways.
• Decreased project cost. Associating an archive with a number of universities, libraries, and cultural institutions defrays the costs of material and equipment procurement, course reductions, and graduate funding, across several institutions. Moreover, smaller departments may be more inclined to meaningfully encourage faculty participation in such archives as the burden of course reductions, leave time, and graduate funding will be spread over multiple institutions. For individual scholars, the ability to point to other departments’ contributions to the archive may be helpful in petitioning for leave time, special topics courses, or travel monies with which contribute to the archive.
• Increased sources of project funding. Such a funding model lessens the dependence of digital humanities projects on NEH funding – and we all know how many worthwhile projects go unfunded each year – while, if the contributing institutions are located in different states, opening the doors of multiple state and regional funding agencies (state humanities councils, etc.).
• Improved Project Flexibility. To the extent such an archive is less strongly associated with a single institution or scholar, it will be able to more rapidly deploy new features and to respond to the needs of its user community. It is the latter of these two flexibilities, the meeting of changing user needs, that is the more important. Unlike traditional scholarly editions, whose user community consists of a very small cross-section of the population who has read a given text (Ex. Users of the Centenary Hawthorne as opposed to readers of The Scarlet Letter), online digital archives serve a larger and more heterogeneous user community. In a truly distributed archive, the needs of the archive and its users are, by definition, privileged over the needs of any individual institution or scholar. As a result, the distributed archive should be able more rapidly respond to evolving user needs and unforeseen technological developments.

Such an archive would also seem to meet some of the criteria outlined by the NEH and CLR’s recent report, Working Together or Apart: Promoting the Next Generation of Digital Scholarship. As Amy Friedlander notes in her opening essay: “The evolving cyberinfrastructure must support collaboration across traditional boundaries and allocate resources efficiently, writes Friedlander, “to enable research at a scale that takes into account the wealth of heterogeneous digital source material as well as computational and analytical power.” Increased institutional collaboration, improved resource allocation, and scalability: all features of the distributed archive.

Replicating Lab Culture in Twitter

I was following an interesting conversation on Twitter this past Friday in which a couple of Digital Humanities scholars attempted to quantify the difference between Facebook and Twitter (Amy Earhart’s distinction of Dorm vs. Office is pretty close to my own use). This was immediately on the heels of one my graduate students reacting with incredulity when I told her that I used my Twitter account primarily for academic exchange. I’ve been using my Twitter account actively for about two weeks now (following more than posting) and I thought I’d do a short write-up of how I’m using Twitter and the ways in which these uses are different from Facebook. But first, a caveat:

The Characteristics of Social Media are User-Defined

The ways in which I use Facebook and Twitter are, to a large part, not the product of constraints within these social media. A quick glance at my wall reveals that probably 2/3 of my status updates fit within the “draconian” 140 character limit of a Twitter update. Had I wanted to, most of my FB status updates could have been posted as Tweets instead. And if you’re using a program such as Digbsy to consolidate your IM and social network feeds, your Twitter and FB feeds are really only distinguishable by the size of the icons to the left of the updated text. Take pictures and the various applications out of the game (most of which, to be honest, I tell FB to ignore) and you’re looking at two very similar services.

So, given that FB and Twitter are, from a software standpoint, more alike than different, it descends upon the user to distinguish between the two services. In my case, I seem to have used a spatial as well as a work/play metaphor in differentiating the two services.

I Use Facebook to Eliminate Space

Having moved from Cleveland, OH, to Oxford, OH, to College Station, TX, before settling in Richmond, VA, I have friends located in all of these cities with whom I’d like to remain in contact. When I started using FB this summer, it was a delight to reconnect with old friends with whom I’d lost touch due to geography. Given that many of the friends I’d made in College Station had, themselves, moved on to jobs in other cities, my set of FB friends is even more scattered than my string of moves would indicate. FB allows me to reassemble the various communities to which I have belonged over the course of my adult life. The reassembling of my various graduate school communities is particularly interesting since these communities are, by definition, temporary and particularly geographically dependent. The graduate community to which I belonged at Texas A&M University no longer exists in College Station, TX. It exists on Facebook.

I Use Twitter to Create Space

In contrast to FB, I have only actually been in the same room with 3 of the 24 people I am currently following. All of the people I’m following are actively involved in the field of Digital Humanities. Unlike FB, which I access primarily through the web (which, I admit, a result of the multi-media content available on FB), I use Twhirl to follow my tweets. The result is an intermittent stream of conversation in the background of my computer, all centered around Digital Humanities issues: MITH announcements, links to recent DH articles, outrage over the FB TOS, as well as various fun and social tweets.

The net effect of this having this background digital humanities conversation on my laptop has been to create a virtual DH lab on my computer. Here, I’m concerned with the social and shared workspace aspects of a lab environment as opposed to the access to specialized equipment. Through Twitter/Twhirl, my work space is surrounded by a low-level hum of DH news and concerns. Moreover, as with a lab environment, Twitter provides me with a space where I can query a body of similarly – oriented scholars as I attempt to solve various DH-related problems related to my research.

I like the Lab metaphor as opposed to Amy’s office metaphor as the conversations drifting through my actual office tend to be only tangentially related to my research (but very relevant to my teaching).

Again, while I’m intrigued by the spatial component that I observed in my uses of FB and Twitter respectively, these are much more a product of the ways I have chosen to use these social media – as well as the tools through which I access their feeds – than they are due to inherent differences in the services themselves.

Feel free to comment with your own observations on how you distinguish between the various social media you use.

 

NINES offers digital exhibition space as alternative to newly evil Facebook

As almost everyone now knows, Facebook changed their Terms of Service last weekend (Happy Valentine’s Day FB users!!) in ways which seemed to imply that FB would own anything you posted on their site in perpetuity (even after you left FB). As news of the new TOS spread around the net, I was dismayed to see a couple of poet and creative writer friends announce that they would no longer be posting draft work on FB. Although FB, faced with a growing tsunami of criticism, announced a return to their previous TOS, many users, understandably, remain skeptical.

Fortunately, in wake of Facebook’s TOS fiasco, Dana Wheeles has offered the use of NINES collaborative exhibit builder as a creative-commons protected workshop space for poets and creative writers, in addition to scholars. You’ll have to create a free user account, but the process is painless. NINES, nineteenth-century studies online, is headquarted at the University of Virginia. Even if the workshop space doesn’t suit your needs, the site has a ton of resources available for the teaching and research of poetry (Rossetti, Whitman, Dickinson, and more).

Fall Course Proposal — So It’s Digitized: Now What?

Over the weekend, I started noodling around with the skeleton of a course I’d like to teach in VCU’s Media, Art, and Text (MATX) Ph.D. program. The program itself is an interdisciplinary Ph.D. run jointly by VCU’s School of the Arts, School of Mass Communications, and the Department of English. As such, the class would be comprised of students drawn from a variety of disciplines, all of whom have a digital orientation to their work.

– VCU MATX, MFA, and MFA Students — Please leave feedback and/or suggestions in the comments section of this post –

For this course I’d like to explore new ways in which digital archives can be used to build new knowledges, methodologies, and communities that are unavailable to their printed analogues. As a discipline, English has done a good job seeing the potential benefits involved in digitizing and disseminating primary texts and research documents. Aside from the copyright concerns plaguing texts written after 1923, most of the texts which are not now digitized will be in the next 10 years (feel free to quibble with these broad assertions).

The question, as many have noted, then becomes one of how to move beyond the constraints of the textual model in using and interacting with these digitized materials. While it is certainly nice to have access to extensive scholarly archives via your laptop/iPhone/neural shunt, and these new levels of access are, to be sure, important, access is – or should be – only one part of the transformation wrought by the digitization of primary works. Having digitized these materials, having freed them from the confines of the codex and the library, what new ways can we envision to interact with these materials?

In “Scholarly Primitives” (2000), John Unsworth identified some basic functions common to scholarly activities across disciplines. According to Unsworth, these scholarly primitives include:

• Discovering
• Annotating
• Comparing
• Referring
• Sampling
• Illustrating
• Representing

How, then, does the transformation of primary sources in English into digital medias transform our approach to these scholarly primitives. For example, given a textual corpus that is freely available to anyone via the internet, how has our understanding of scholarly “discovery” been changed? Are there new primitives that, in the wake of pervasive digitization, might be added to Unsworth’s list? As Lisa Spiro’s write-up of the 2008 Research Methods Session at THAT Camp demonstrates, such questions are central to the field of digital humanities (and, by extension, the humanities in general) .

The Example of Poe

It seems to me that the Edgar Allan Poe Society of Baltimore (EAPSB) would serve as an ideal starting point for thinking about some of these questions. As it currently stands, the EAPSB provides its visitors with access to digitized versions of the vast majority of Poe’s published writings, along with selected introductory and scholarly materials. Established in the late 1990s, the site is an archetypical web 1.0 archive. Texts are presented in html with minimal to no metadata. For the purposes of this class, these texts would provide student with a blank canvas upon which to build additional learning and research structures. Students enrolled in the class could work to transform the information in the archive in ways that revealed new knowledge about Poe or which would serve new and unanticipated user communities. While students would be encouraged to reconfigure the material in the archives in ways congruent with the practices of the social, remixable web.

Moreover, Poe is uniquely suited for such a class due to his location at the confluence of a number of disparate knowledge communities. As Shawn Rosenheim and Steven Rachman have noted, “unlike the works of Melville or Hawthorne, Poe’s texts have not been primarily transmitted through the schools” (The American Face of Edgar Allan Poe). Poe belongs to the people in ways other authors simply do not. As I mentioned in a previous post, while Hawthorne and Melville garner 1.7 and 2.3 million hits respectively via Google, a search for Edgar Allan Poe yields 5.6 million hits pointing to a variety of serious and not-so-serious engagements with the Poe corpus. So while it’s pretty clear who is going to be visiting Melville’s Marginalia Online, a comprehensive Poe site is going to have a more heterogeneous user community, one which will probably be comprised of a number of separate but overlapping user communities.

Digital Humanities

On a more theoretical level, the course would look at the problems inherent in digitally representing physical, textual objects in ways that fully take advantage of digital mediums and the tensions between these methods and the physical presence of the literary artifact or book.  Most of the readings would center draw upon recent digital humanities scholarship dealing with the promise and limits of new media for humanities research projects.  Potential readings might include some of the following:

• Hayes.  Poe and the Printed Word.
• Situates Poe in the context of a c. 19 in which the definition of textuality is rapidly transforming.
• Bryant.  The Fluid Text: A Theory of Revision and Editing for Book and Screen.
• As an added bonus, Bryant provides a lucid overview of the last 40 years of textual editing theory.
• Hall.  Digitize this Book: The Politics of New Media, or Why We Need Open Access Now.
• Kirshenbaum.  Mechanisms: New Media and the Forensic Imagination.
• Hansen.  New Philosophy for New Media.
• Jenkins.  Convergence Culture: Where Old and New Media Collide.
• Keen.  The Cult of the Amateur: How Blogs, My Space, You Tube, and the Rest of Today’s User-Generated Media are Destroying Our Economy, Our Culture, and Our Values.
• Liu.  The Laws of Cool: Knowledge Work and the Culture of Information.
• Moretti.  Graphs, Maps, Trees: Abstract Models for Literary History.
• Jeanneney.  Google and the Myth of Universal Knowledge: A View from Europe.

These are just my preliminary thoughts on the course. I welcome and encourage suggestions in the comment thread – especially from VCU MATX, MA, and MFA students who might be interested in such a course.

Assignments

I haven’t worked through the assignments portion of the course yet (and I want to get this posted sooner, rather than later). Potential assignments could branch outside of the Poe archive if students desired and might include the use of social media, mapping, and tagging projects, as well as book or archive reviews and – if desired – a traditional seminar paper.

Again, comments not only welcome, but needed. If there is enough student interest I’ll probably launch a wiki for the course to solicit more student input for the development of the class readings and assignments.

Goodbye Google Chrome

I gave up on Google Chrome this week and switched back to Firefox 3.x. I was a reasonably early adopter of Google Chrome and thought that the Google team got a lot right:

• Speed: Chrome was noticeably faster than Firefox on my p.c. I know Firefox’s advocates assure me that the speed and memory issues have all been addressed, but Firefox still feels pretty slow most days – especially if the computer hibernated overnight and I didn’t bother to reboot. Chrome really seemed to fly with Google as well.
• Interface: I liked the minimal interface. This is unsurprising as I tend to like Google’s minimalist aesthetic. While odd at first, the collapsing of the search field into the address bar was particularly smart. I found myself repeatedly typing search queries into the address bar when I switched back to Firefox.
• Draggable Tabs: The ability to drag a tab into a new window was a real winner. Having lost a couple of documents in Google docs due to having a gazillion tabs open and then navigating away from my work, I really appreciated the ability to drag a Google docs or other important tab out of the main browser session.

Here were the deal-breakers in order of importance:

• “Confirm Form Resubmission” – a number of people have commented on this problem. I get that this feature is intended as a protection in commercial environments – it guards against double purchases or billings when submitting commercial forms. Unfortunately, it also required me to reload my searches every time I used worldcat via firstsearch, as well as a number of other databases.
• Unitary Search Bar – while I like the omni-bar from a design standpoint, I really missed the ability to narrow my search to a specific site or search engine as I do via Firefox’s search bar. In particular, I missed the ability to easily limit my search to the major bookseller sites (Amazon.com, BN.com, Bookfinder.com)

It’s worth noting here that these two shortcoming seem to be the result of the relationship between Google Chrome and Google adsense, or the programming and the commerce sides of Google. Obviously, Google wants Chrome users to use the omnibar to search Google, supporting Google Adsense in the process. Similarly, the “confirm reload” problem seems to result from an emphasis on the commercial as opposed to research uses of form data. These problems, then, aren’t so much bugs as they are the natural outgrowth of Google’s philosophical orientation – as opposed to the that of Firefox.

My other two major problems:

• RSS: This one seems to be a bug. Unlike Firefox or IE which have convenient buttons for adding an RSS feed to your reader, Google Chrome currently lacks this feature (I find Firefox to have the best implementation with the RSS button in the address bar). This is slightly ironic as I use Google Reader for my RSS reader and like it very much.
• Applications Running in Chrome: At some point in my early usage of Chrome, I think I must have given Chrome permission to run applications such as Windows Media Player form within the browser. As a result, every time I launched a streaming WMP feed, it showed-up as a new Chrome tab. Unfortunately for me, I prefer to minimize WMP to the system try when streaming in order to free-up space on my miniscule laptop monitor (I think this is also a left-over habit from the small-monitor days). I never could figure out what setting I needed to poke to stop this behavior, and in the end I simply used IE7 to launch Radio Paradise or SomaFM while I was working.

While there’s a lot to like to Chrome, I wonder if the commercial aspects of the software (and I wonder to what extent the windowing of applications such as WMP is part of this orientation) will allow it to be the browser so many early adopters want it to be (basically a faster version of Firefox).

Why a Contextual Archive?

In-between the MLA and other holiday travels, I had the opportunity to discuss the possibility of a revised or extended with a number of Poe and Digital Humanities scholars. While many of my DH colleagues were very supportive of my proposed idea to publish the entirety of each journal issue in which a Poe story appeared, a fellow Poe scholar raised some good questions that I need to answer as I move forward. The question is as follows:

“I am also not sure that I understand the utility of reproducing the full issue of Graham’s, including the items not related to Poe. What is the purpose of reproducing the text of “The Bride,” “Procrastination,” or “Perditi”? . . . I hate to think of the additional work necessary to reproduce the full texts of the surrounding material.”

These questions boil down to 1) Why publish non-Poe materials; and, 2) Will it be worth the effort.

Although I’m interested in the writings of Edgar Allan Poe, this interest is derived from a broader interest in antebellum American literature. My previous research centered on Hawthorne, Melville, Stowe, and Whitman, in the context of a developing American museum culture (characterized by P.T. Barnum’s American museum). While I can honestly state that I love all of the authors I teach and research, my efforts to understand these authors center on connecting or re-connecting them to their historical context. As I look at the list of authors covered in my previous research project, it’s interesting to note how each of these authors explicitly invites it readers to understand their work through a specific historical context. It’s unthinkable to teach Hawthorne without discussing New England History, Stowe without a discussion of c. 19 millennial and abolitionist thought, and both Whitman and Melville (in both “Song of Myself” and Moby-Dick’s wonderful Cetology chapters) go so far as to import the relevant historical and social material into the bodies of their texts. The problem is aggravated by Poe’s geographic mobility, resulting in a situation where 5 major US cities can all make some sort of claim on Poe (Richmond, Philadelphia, New York City , Boston, Baltimore).

The same cannot be said of Poe. Quick . . . Where is Roderick Usher’s house located? It’s never mentioned. Even when Poe does provide a specific location for his tales and Poems, the location is often incidental to the action and/or characterization of the story. Moreover, as Terrence Whalen has noted, Poe’s politics fuel his exclusion from historicist accounts of c. 19 American literature as, “the very scholars committed to analyzing the social determinants of writing are also the scholars most likely to be repulsed by Poe’s ideology, by his Whig politics and his habitual denunciation of that mysterious collectivity known only as ‘the mob’” (Whalen 1992, 383). To clarify, the fault here lies not with Poe, but with the inability of many historicist scholars to find a place for Poe in their historicist projects (though work by Whalen, Kevin Hayes, and Meredith McGill, among others, is rapidly rectifying this omission).

The under-historicizing of Poe is aggravated by additional critical and digital factors. Critically, the combination Poe’s embrace by nineteenth-century French symbolists and twentieth-century U.S. Freudian critics. The result is a long tradition of Poe criticism that, while often insightful, is frequently a historic and even a temporal.

Such readings abound on the internet which, as a medium, itself contributes to the further dislocation of Poe from any sense of historical context. As I have mentioned elsewhere, at 5.5 million pages, the number of sites dedicated to the presentation and interpretation of Poe’s tales and Poems dwarfs that of any other nineteenth-century American author (Whitman comes in second at 3.5 million). This digital dissemination of Poe further serves to sever him and his works from their historical context. Setting aside the various reappropriations and presentations of Poe’s works offered by goth and horror sites, even those sites which purport to present Poe as a species of Literature (such as literature.org), do so by lifting Poe out of his c. 19 US context and into a generic, timeless and placeless literary realm.

The net, then, aggravates an already existing tendency among the general population to view Poe as an archetype as opposed to an author, universalizing his works at the expense of their ability to speak to the central concerns of nineteenth-century America: each disembodied digital presentation of “The Tell-Tale Heart” or “Ligeia” further removing Poe from his culture.

Moreover, publishing Poe’s tales in their full journalistic context also becomes a way in which a revised Poe archive can go beyond simply providing the Poe corpus with the addition of metadata and keyword searching. The Poe corpus is available online in numerous – often unreliable – forms. Google Books is making more and more c. 19 texts available online (often with faulty metadata). What’s missing is a way to bring together a reliable set of Poe texts with an appropriately edited, tagged, and indexed selection of the nineteenth-century literary productions which formed a significant background against which Poe’s texts were read and understood.

While I cannot unequivocally state that reading M.H. Parson’s “Procrastination” or William Wallace’s “Perditi” along with “The Mask of the Red Death, A Fantasy” will necessarily lead to a major critical re-evaluation of Poe’s tale, it does help us to a fuller realization of how Poe’s work spoke both to and against the literary and commercial conventions of his culture. While the amount of work involved can seem daunting, it seems to me that there is at least the potential for rich dividends in the future.

“Slow Blogging” and the rise of RSS readers

There was an interesting article in the New York Times prior to Thanksgiving on the rise of “Slow Blogging.” As defined by the article, the term slow blogging describes blog authors who post long, thought-out articles on an infrequent basis, in contrast to rapid-fire blogs with lots of, “hey, I’m buying sushi” entries.

In tying this phenomenon into a general drop-off in the pace of postings at some high traffic blogs (excluding aggregators such as Boing Boing and the Daily Kos), the author of the article notes the role played by the rise of Twitter and Facebook.  Whereas previously an individual might use their blog to post short observations or updates, now these purposes are served more capably by “tweets” or FB status updates.

While all this rings true, I think an overlooked factor in the drop-off in posting rates is the rise of RSS reader usage.  In discussing slow bloggers, article cites authors who deliberately see how long they can go between posts before their readers desert them.  But with an RSS reader, I use Google Reader, a blogger could go months between posts and, as long as the posts are interesting or useful, I’d still read them via my RSS Reader.

Aside from Boing Boing, Life Hacker, and a couple of other high-volume blogs, I almost never visit any of the blogs of which I consider myself a regular reader.  Rather than reading individual blogs, every day I run through the unread stories in the “Digital Humanities” folder in my Google Reader Account.  While I’m not checking Dan Cohen or Lisa Spiro’s blogs every day, I also never miss a new post.

In fact, my only reason for using Google Reader is to aggregate a bunch of slower-paced academic and digital humanities blogs into a daily news feed.  While I initially had some more fequently updated blogs in my RSS feeds, the pressure of 10, 20, 60, unread articles every time I logged-in made me quickly remove them from my list of feeds.  I’ve even removed a couple of digital humanities blogs from my feed list for the “sing” of too frequent postings (USC IMD, I’m looking at you).

While FB and Twitter have surely siphoned off a fair share of blog posts, RSS readers have made it possible for bloggers to post less frequently without frustrating their readers.  I definitely read a wider variety of blogs now than before I started using one.

Replication, Deformation, Transformation – A Dialogue with Wesley Raabe

About a week ago, Wesley Raabe offered up some interesting commentary on my recent post, “The Fluid Text as Textual Deformation.” Wesley’s point of departure with my post lies with my claim that, “digital replication of the print book is not properly the province of the digital humanities scholar.” Having read through Wesley’s comments, I feel some clarifications on my part are in order.

The problem here, I think, lies in my use of two terms: “replication” and “book.” In response to my use of the terms in my previous post, Wesley notes that, “dismissing books as physical objects and dismissing the intellectual rigor necessary to reproduce them as someone else’s field (or mere workmanship) is to narrow the scope of digital humanities.”

What’s interesting to me, here, is the question of the extent to which Wesley’s project can be accurately described as a “reproduction” of the National Era text of Uncle Tom’s Cabin. By clicking on the “zoomable image” or “plain JPEG” options on the site, the reader of this text is able to view high-resolution photographs of each and every page of every issue of The National Era which contained Stowe’s novel. Here, Raabe’s project moves well beyond the limitations of microfilm or even the physical text itself as readers are able to see the not only the hue and texture of the original National Era pages, but also the binding boards and end pages of the individual who bound the individual numbers of the Era into a single volume.

Yet, it is very attention to detail, the hyper-reality of the object at the center of Raabe’s project, that moves it beyond the category of “reproduction” or “replication.” As a reasonably young man, I do not tend to read with a magnifying glass in hand (yet). But with the aid of Raabe’s zoom took, I can focus in on the lightest of pencilings on this particular issue of the Era. In those places where the text in the zoomable image is obscured by the fold of the newspaper, I can call for a normalized transcription of the passage from a drop-down menu as an aid to my reading.

A fold in the National Era text

Moreover, the provision of the entire contents of each issue of the National Era provides an important explanatory context for Stowe’s work. Here, the electronic edition of the National Era text does not limit itself to narrow replication of the original physical object (as it would if it were part of Google’s book digitization project). Rather, it offers the reader a host of new ways to interact with and to examine the text.

A larger problem here is the false dichotomy I may have set up in my previous post between “replication” and “transformation.” While my intent was both to echo McGann in my use of the word “deformation” and to use the word in a complimentary fashion, it’s clear that the word has too many negative connotations – especially in the world of textual scholarship. I think a more useful term going forward, and borrowing from digital imaging, would be “transformation.”

Complimenting this ambiguous wording is the slippery nature of the very term “book.” Where book is used interchangeably with the word text, I will more cautiously assert that the replication of the book (text) should not be the primary goal of the digital humanities scholar. Raabe’s case, in particular, is interesting here. While his edition is labeled as, Harriet Beecher Stowe’s Uncle Tom’s Cabin: An Electronic Edition of the National Era Text, the edition reproduces three numbers of the National Era which do not contain the Uncle Tom’s Cabin text. Raabe’s reasoning for this editorial decision is sound (the original collector included these three numbers in his or her ‘edition’ of the text), but it also highlights the tension between the often bodiless form of the text and the physical reality of the book as object – a bodilessness which is doubly aggravated when dealing with canonical texts and their digital incarnations.

Raabe’s project, however, is engaged in a more complicated transformation of Stowe’s text insofar as it offers not a “reproduction” of Uncle Tom’s Cabin, but a transform-enabled presentation of a specific edition of the novel based upon its serialization in the National Era – a presentation which, I might add, is thoroughly aware of its mediated status. Because the project is transformative (both of the novel Uncle Tom’s Cabin and of the physical text which it represents), because is aware of the mediated nature of its project, its labor is easily visible as digital humanities scholarship.

User-Designed Digital Project: Notes from MELCamp

In a forthcoming article for Digital Humanities Quarterly, Kenneth M. Price asks an important question for digital humanities projects: “What’s in a Name?” Using his experience with the development of the Walt Whitman Archive as a spring board, Price examines the implications of a wide variety of suitable names for online projects dedicated to the digital presentation, annotation, and transformation of literary texts: names such as edition, project, database, archive, and thematic research collection. This question seems particularly relevant to me of late not only because of my own work with the Digital Poe project, but due to the discussions in which I participated at John Bryant’s MELCamp during the final weekend in October.

MEL is shorthand for the Melville Electronic Library, a proposed online project for the study of the works of Herman Melville. MELCamp itself was a one-day conference dedicated to fleshing out some of the ideas constellated around the Melville Electronic Library and to demonstrating some of the philosophies and technologies (TextLab) which will be central to the project. During the introductory presentation, John briefly discussed some of the strengths and weaknesses of referring to the project as an “electronic library.” Aside from the charming MEL acronym, referring to this digital project as an electronic library would seem to foreground the textual core of the project while allowing sufficient wiggle room for the project to develop in new, unforeseen ways.

At the same time, at this early stage of the project the very capaciousness of the term “library” can lead to a bewildering array of not always complementary conceptions of the future Melville Electronic Library. Following the morning panels spent discussing where Mel “is,” eight conference panelists shared their thoughts on “Where We Want Mel To Be.” Their comments are as follows:

1. Wyn Kelley: MEL has potential to address international audiences, and it should include translation.
2. Samuel Otter: MEL should address Undergraduates and the wider reading public as well as scholars; it should enable users to gain deeper access to “resistant texts.” How then might they “navigate horizontally” through MEL’s content rooms?
3. Robert Levine: How can sites like MEL use creative annotation and links to stress intertextuality as well as traditional source study.
4. Robert Madison: Development in Optical Character Recognition (OCR) programs can facilitate transcription of texts. How can MEL negotiate proprietary texts; not only secondary critical works but modern copyrighted editions of HM’s work, such as the Chicago Billy Budd or NN [Northwestern-Newberry] Moby-Dick. MEL can facilitate the study of HM’s “plagiarisms” or borrowings from sources.
5. Robert Sandberg: hopes for a synergy between MEL and the NN [Northwestern-Newberry] project.
6. Tony McGowan: hopes for a MEL “mission statement”; wonders how MEL can make new scholarship more accessible to young scholars.
7. Wesley Raabe: MEL is right to work out a clear conception of itself early on. In particular MEL needs to determine the pathways between the textual core and the rest of the rooms, or what may be called a “thematic research collection.”
8. M. K. Bercaw Edwards: How can MEL make oral sources available and visualize whaling knowledge?

While everyone at the table seemed to be in agreement with regards to the importance of the textual core of MEL, once this core has been assembled, the possibilities for MEL start to multiply – as do the questions. Who is to be the primary audience for this project? What is the appropriate balance between features for professional Melville scholars and “enthusiasts”? To what degree will user-generated tags, emendations, and content be permitted or even encouraged? Is MEL to be a clearing house for Melville scholarship, or should it limit itself to fostering engagement with primary texts?

To a certain extent, it seems unavoidable that digital projects of this nature tend to become blank canvasses against which interested scholars will project a variety of competing “ideal” projects. Freed from the constraints of the codex and the finite production schedule of traditional scholarly editions, the digital horizon seems limitless. Given enough funding and interest, there’s almost no feature that a humanities scholar can imagine that cannot be added to a digital project. Here, the problem presented by the digital edition/archive/library is analogous to that of Carroll’s and Borges’ maps “with a scale of a mile to a mile.” At what point does the digital representation of the author and his or her historical and critical contexts become unworkably large, both for the producers of the project and for its users.

MEL, of course, is in its earliest stages and has yet to work out a final mission statement. As this mission statement is hammered out over the next year, certain potential MELs will drop by the wayside while certain other MELs are realized. It seems pretty safe to assume that, initially, this mission statement will revolve around the textual center of the project: the need to mount a new, digital, fluid-text edition of the complete works of Herman Melville.

Once this initial project is completed, however, how to determine where the project goes next? Or, more to the point, who is to determine the next phase of the project.

Here, it seems as though the “feature request” and “bug fix” models of open source software development might be particularly useful. In many software projects, many of which are commercial, authors maintain feature request logs as a means of planning future developments in their product. That is, user demand drives the future development of the project. This idea seems antithetical to the traditional editorial practices of humanities scholars – and critical editors in particular – but it seems to me to have much promise.

For a project such as Digital Poe, where, at least on the prose side, there are more primary texts and the texts are shorter in length, mounting an initial, limited sampling of the corpus and then allowing user feedback to determine the priority of the remaining texts seems like a smart way to modularize development. This can also be carried over into the archive’s feature set: is there a demand for the ability to visualize the geographical distribution of Poe’s publications? Fine, build it into the project. Is there an interest in presenting facsimiles of all the contemporary reprints of Poe’s works, add it to the project? Obviously, there are limits to this approach. To follow every whim of the user-base in terms of planning future projects is to return to the initial project of the limitless horizon of the digital project. But to the extent that the ultimate goal of all DH projects is the provision of projects which are widely used and which promote a deeper understanding of their subject, user input should be considered when planning project growth.