My comment on Les Harrison’s post, Fluid Text as Textual Deformation, is too long, so I decided to put my comment here. I would quibble with your claim that “digital replication of the print book is not properly the province of the digital humanities scholar.” And I wrote a blog post about this during the Miami of Ohio 9s seminar. But perhaps I can say more with respect to how this concern has functioned in my own scholarship.
Digital humanities scholars may choose not to be interested in digital reproduction, but to suggest that it is not their “proper province” is bewildering. Some students of the history of the book find analytical bibliography not properly in their field, but, as Tanselle has suggested, they fail to consider some primary evidence. I disagree with Tanselle’s assertion that books are “primary” and publisher records are “secondary,” because the distinction depends on what type of study one undertakes. 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.
For my dissertation, I worked with a physical object. A collector assembled an edition of Uncle Tom’s Cabin from the National Era numbers (weekly issues were numbered, and the term “numbers” avoids confusion with the bibliographical use of “issue” for another purpose), and he or she decided that every page of each newspaper number was part of this privately constructed book, and maybe even part of the work defined as Uncle Tom’s Cabin, though that principal would play havoc with a traditional notion of a literary work (as perhaps it should). He or she also decided that three of the numbers (no. 242, no. 252, no. 259) were part of Uncle Tom’s Cabin even though these individual newspaper numbers did not include the text of Stowe’s work. I decided to respect that choice, as well as to respect the attentions suggested by binding the volume within blue boards (making a book out of it) and inserting endpapers. So I photographed the cover, the endpaper, and every page of each number of the newspaper, including those numbers that lacked an installment of Stowe’s novel
The choice to navigate between four options, “Quasi-Facsimile Text (with Line Breaks),” “Normalized Text (Prose Re-Set),” “Zoomable Image,” and “Plain JPEG (100 dpi),” is informationally dense. But it is not meaningless, and it’s a line that highlights conscious decisions and failures both of the meaningful and incidental variety. A conscious decision is to transcribe Uncle Tom’s Cabin (an editorially constructed version of the work that happens to coincide closely [except in some important spots] with the linguistic text of the two-volume Jewett book version) while providing images of each newspaper installment. Thus, each navigation option recognizes different ways of conceiving of the work: as a faithful documentary reproduction of Stowe’s text with editorial notes on copy (quasi-facsimile), as a traditional transcription with very slight emendation and less emphasis on accidents like line-end hyphenation and no notes (normalized), as a low-resolution page image that preserves much about the “book” as documentary object (plain JPEG), and as a high-resolution version of the same. In the last, image, like text, is sliced, diced, and re-constituted in parts according to user navigation. Thus, I offer three versions of reproductions that I consider respectful, according to different standards of scholarly interest. I offer a more rigorous and theoretically grounded version of this explanation in chapter I. When I invited Peter Shillingsburg to look at my edition, he objected to my decision to not just open the edition with the image, by which he meant the first page of the first newspaper issue. I chose for the opening page to foreground the transcription, because I want readers to know that that access to the text has been mediated. But that’s a conscious decision based on the belief that images, too, are mediated. One of the edition’s failures will be immediately apparent to any reader with Internet Explorer who uses your link to access my site. They are presented with nothing but menu options. Readers who use Firefox will also see a text. But that’s why I prefer readers to see the front page and its software requirements on this page: http://www.iath.virginia.edu/~wnr4c/index.htm
My failure is a matter of digital workmanship, which makes demands on users because of items they are unlikely to care about. I insisted that ligatures and digraphs be reproducible. IE (at time I was doing the edition) would fail or substitute. Firefox on PC reproduced the ligatures and digraphs beautifully. The use of divs to create a divided frame works beautifully in FireFox but results in no text in IE. A better manner of proceeding would be to warn IE users, but neither I nor my technology guru knew how to fix it in time to finish dissertation.
Maybe I’m just touchy because my technology implementation includes stumbles while I claim also to be a digital scholar. The part that I’m doing more conscientiously, textual scholarship, is lost in interface concerns. My quibbles and explanations aside, it is an honor for my edition to be compared and contrasted in your blog post to John Bryant’s monumental project. I hope to create a site that will deserve that comparison in the next few years.
My site is free, and maybe another truth holds. Some of what you get is what you pay for.