Last February, a Senate hearing highlighted the sorry state of the Founding Fathers projects. While no one questioned the extremely high scholarly quality of the published volumes of papers that have been produced to date, the hearing noted the glacial production pace, high cost, and limited access to the finished products (which are expensive and bought by few libraries). The obvious question was whether technology could address the problem, and Congress ordered the Archivist of the US to report on the matter.
NARA’s report, “The Founders Online: Open Access to the Papers of the America’s Founding Era,” was sent to Congress in April, but it has received little notice or discussion. This is unfortunate because it is an important work on an important issue. There is much to admire in the report, but overall it demonstrates a fundamental failure to understand what open access means or how technology can make scholarship more productive. It seems more interested in protecting existing partnerships and editorial practices than in breaking new ground and fostering public access. Let’s hope that Congress recognizes how unsound the recommended approach is and pushes NARA to do more.
Detailed comments follow.
There is much to admire in this report. For example, it recognizes (as do we all) the tremendous scholarly achievement that the Founding Fathers papers represent. It is precisely the high value and scholarly worth of the work done up to now that makes it imperative that more volumes are produced faster, and that access to these valuable volumes be broader.
In addition, while Congress directed the Archivist "to develop a comprehensive plan for the online publication, within a reasonable timeframe, of the papers of the Founding Fathers," the report has elected to speak of "open access" (rather than just online access). There are problems with the report's vision of open access, as I outline below, but NARA is to be commended for in theory at least making open access its goal.
Lastly, one of the recommendations of the report is that NARA commission rough transcriptions of all of the remaining unpublished papers from the Founding Fathers. These rough transcriptions would be made available online immediately, to be replaced by edited and annotated versions as they become available from the editorial projects. This is a clever idea and could do much to speed the production of future volumes (assuming that the transcribers are experienced paleographers and can easily read the various handwritings in the documents).
In spite of these positive steps, the overall report is incredibly disappointing, both for what it says and what it leaves out. Here are just a few examples:
- The report sets up as an option, which it then rejects, scanning published volumes as they are finished. No one would even consider the idea of scanning published volumes that have been produced from electronic files. The real option is to require that the editorial projects produce their volumes according to a standard XML schema, and that those electronic files be transferred to NARA for its use as part of the grant. There is no discussion of how to provide open access to previously published volumes. Why not link, for example, the images of Washington's correspondence found on the American Memory site to the transcripts as published by the Washington papers project?
- What about copyright you might ask (as Allen Weinstein did at the Congressional hearing)? There is no discussion of the extent of copyright ownership in the product produced by the projects. It would be a scandal if over $17 million dollars in NHPRC funding had been used to allow 3rd parties to commercialize and control the words of the Founding Fathers. Fortunately, under the regulations governing grants to higher education institutions and non-profit organizations (OMB Circular No. A-110, section 36), Federal awarding agencies are required to "reserve a royalty-free, nonexclusive and irrevocable right to reproduce, publish, or otherwise use the work for Federal purposes, and to authorize others to do so." NARA can therefore do what it wants with the volumes produced to date. Nevertheless, the absence of any discussion of copyright in the report is appalling.
- There is no discussion as well about the rights or interests of the repositories that provided copies of the papers of the Founding Fathers to the various editorial projects. We can assume that the provision of copies carried with it permission to transcribe, annotate, and publish the works in print. It is far from clear, however, that the repositories - especially those that made copies in the 1950s - gave either the papers projects or the University of Virginia permission to prepare electronic versions of the documents or, in the case of UVA, to commercialize the electronic version. The issue here is identical to the Supreme Court's ruling in the Tasini case, which stated that a publisher does not have the automatic right to prepare an electronic edition of a work first published in print. (The one difference, of course, is that Tasini was based on copyright rights whereas in this instance, the issue would hinge on the contract governing the provisions of copies to the papers projects). Again, there is no mention of the rights of the repositories that own the originals.
- This would be especially important if the report envisaged associating facsimile images of the originals with the transcribed content. This is something hard and expensive to do in print, but easy to do on the Internet. Again, there is no mention of whether images of the originals should have a place in the project - perhaps reflecting the traditional approaches that dominate the report. (I would argue that of course they should be there.)
- Perhaps the most problematic issues in the report surround its use of the term "open access." For some, open access means "digital, online, and free of charge." The report, while saying it wants to provide open access to the material, appears to recommend that all material be given to UVA's Rotunda system for delivery. Rotunda follows a subscription model - not open access - that is remarkably expensive considering that citizens have already paid for all of the editorial work on these volumes. How could this be open access? Apparently Rotunda might be willing to give up its subscription approach if a foundation were willing to pay for all of its costs. Unless such a commitment is in place, I find it disingenuous to describe a Rotunda delivery option as "open access." There is no discussion of other, free, delivery options, such as the willingness expressed by Deanna Marcum of the Library of Congress at the Senate Hearing to make all of the Founding Fathers papers accessible through LC (which already has a good site pointing to currently accessible papers).
- Others argue that for true open access, information must be accessible outside of specific delivery systems (such as Rotunda) and made available in bulk. Open data and open interfaces allow for all sorts of interesting uses of material. For example, someone might want to mashup George Washington's papers to Google Maps in order to be able to easily visual geographically the spread of information. Others might want to mesh manuscript material with published secondary literature. Rather than anticipating the widespread dispersal and re-use of the Founding Fathers papers, however, and hence the need for harvestable data, open APIs, distributed access, etc., the report calls instead for "a single, unified, and sustainable Web site" - apparently the locked-down Rotunda system.
- As far as the slowness of the editorial process, the report has only two recommendations to make: the creation of the common transcription system described above, and the creation of an oversight process to ensure that the projects meet rigorous production benchmarks. There is no consideration given, however, to how technology might improve and speed the editorial and scholarly processes. If these projects are to finish in our lifetimes, it is not enough that the editorial staff work harder or raise more money (as is suggested on p. 12). Instead, the editorial process needs to be done differently. How about common editorial software, for example, and share linked authority files so that a note about an individual in one volume could be used by an editor in a different series? Are there other ways of speeding editorial processes?
As I said at the beginning, the commitment to open access and the willingness to transcribe all the remaining documents is admirable. The report's failure to break from its reliance on the university press publishing model, its failure to discuss copyright and physical ownership issues, its silence on the offer of support from the Library of Congress, and its unwillingness to consider new models of editorial practice to speed production are damning, however. Congress needs to reject this report and tell the NHPRC to get it right.