Thu 7 Jun, 9:00-10:30
Session 14: “Done”: Finished Projects in the Digital Humanities
Matthew Kirschenbaum, William A. Kretzschmar, Jr., David Sewell, Susan Brown, Patricia Clements, Isobel Grundy
Some interesting theorizing occurred here.
W. Kretzschmar pointed out that finishedness depends on individual tasks one can call finished, even if a large-scale project itself is never finished. (There’s always more work one could do—but avoid chasing your own tail.) He gave an overview of LAMSASplot (Mac), a linguistic atlas project, whose NEH funding died partway through. The project was ported to the Web, which meant transforming proprietary-format relational files to non-proprietary flat ones, partly because the developer of the relational software, FoxBase, was going out of business; a second iteration (the one currently visible) involved moving to mySQL and Unicode. [Interesting—one rarely sees a move towards flat data files, even if it’s interim.] Closing thought: “you do not have to finish the work, but neither may you desist from it.”
D. Sewell (“It’s for Sale, So It Must Be Finished”) suggested that we’ve had some semantic slippage: open(-ended) vs. closed, and incomplete vs. complete/finished. The tendency has been to merge these two sets of terms, then to valorize open/incomplete. The Talk page for a Wikipedia article titled “Unfinished work” asks whether Wikipedia isn’t itself an unfinished work, to which someone has replied with the question whether something that cannot be completed can really be classed “unfinished.” Landow and Delany 1991 note that if one put a work conventionally considered complete, such as Encyclopaedia Britannica into hypertext form, it’d immediately become incomplete. What if one replaces EB with a published novel?—something not in print form is axiomatically open-ended. (The postmodern project is all about the cult of the perpetual prototype, anyway. [Not that it’s a good thing.]) Customers of UVa’s Rotunda imprint [what Sewell manages] are implicitly bracketing the theoretical implications of “finished” in favor of the economic ones, e.g., the ability to purchase access.
Sewell then walked us through parts of the “fluid-text” Typee edition, which began as Word and Framemaker files before becoming TEI-XML. The parallel presentation of MS facsimile and edited text uses bitmapped PDF for the latter rather than HTML text. He also showed us the Papers of George Washington Digital Edition, which is based upon 52 print volumes. [Yikes.] PGWDE is incomplete at its first release, inevitably. [I missed a bit here.] So—one can ask whether a project is closed / bounded, or complete / static; the hardest one for print culture to comprehend is something that’s neither closed/bounded nor complete/static, yet it’s common on the Web (wikis, blogs, etc.). Meanwhile, in the economic world, if something doesn’t sell or if the projected cost of an upgrade exceeds revenue, it’s “done.” 😛 Competitors’ addition of features can render a site “incomplete” after the fact, as can evolving standards and best practices. Ditto adding a previously standalone publication to a collection, since one generally wants to add things that integrate it cleanly. Failure to migrate as needed is tantamount to “going out of print” (abandonware).
[Not sure who spoke from the Orlando Project.] She gave an overview of how the project had come to be: it developed in stages because doing everything at once was too much. [Sorry, too difficult to hear; not taking notes.]