At first glance, publicly crowdsourced projects, whether in the humanities or sciences, may not seem that novel in concept – especially when one considers the inherently collaborative & social nature of so many corners of the internet. But as initiatives removed from the hierarchy and controlled access of formal collections in the library, archive, & museum world, they do represent a radical departure from conventional practice and, to some of the more entrenched members of the LAM establishment, “best practices”. While the realities of 2020 may have thrown out many of these collections access best practices, crowdsourcing represents a potential public history platform (in one of the most literal invocations of the phrase) for institutions seeking to develop new paths for community engagement and collections use. In particular, my focus is on the opportunities afforded by volunteer-driven digital transcription projects, with the anticipated end result of a crowdsourced transcription site for the Judd Papers, which are currently housed by the Forbes Library in Northampton, Massachusetts.

With the goal of articulating a set of the most, well, practical of best practices, I examined the transcription interfaces of the Papers of the War Department, 1784-1800 (at the Roy Rosenzweig Center for History & New Media, George Mason University), the Library of Congress By the People initiative, and the Smithsonian Digital Volunteers Transcription Center, three established (and presumably well-executed) crowdsourcing projects.

My approach to each began with a rather simple question: How long will it take me, starting from the project home page, to create an account and begin transcribing a record? As it turns out, it took me less than three minutes to get started on all three occasions! That time would have been reduced to “under two minutes”, but the Smithsonian Transcription Center Browse Projects landing page is a bit more difficult to navigate than the others. Here’s some of the best features I noticed, in a roughly chronological manner:

The LoC’s By the People has a shortcut that immediately takes you to a random page in need of either transcription or review, a neat feature that streamlines the process if volunteers are interested more in the process of transcription, rather than the subject matter (currently, there are twelve different “Campaigns” active on the site, with the majority of them focused on some aspect of women’s history). Clicking “jump right in” immediately brought me to “The Second Battle of Concord,” a record found in the National American Woman Suffrage Association Records.

Following the path at the top left brings up a Subject File for Edward H. James, the record’s author, which features a overview layout of the record’s contents, including a page-by-page status breakdown (“Transcribe”, “Review”, & “Complete”) and a progress bar for the full record, as well as a handy little progress color chart as well. I do have to wonder, though, is this layout approach useful for records that have dozens or hundreds of pages?

One thing that is useful, regardless, is the ability for multiple contributors to collaborate on records through both transcription and review.

Beyond the record pages, I thought the ways that these services provided guidelines and resources for their transcription volunteers to be particularly intriguing. The Papers of the War Department has a 3-path approach to their Guides page, providing instructions for use of the collection, historical context and paleography resources, and detailed guidelines for completing the transcription process. I find this to be a comprehensive approach, if through a somewhat physically underwhelming interface. However, as an Early Americanist particularly focused on the “Late Eighteenth Century” who also has undertaken a good deal of transcription across various academic and museum collections projects, I do have to acknowledge that I am evaluating these resources from very much an experienced perspective. I do think, regardless, they will be a useful reference point as this semester’s project curates our own set of resources.

The Smithsonian takes a much more direct approach to their guidelines; immediately upon opening a transcription record, you are immediately prompted to consult the project-specific (i.e. content-specific) tutorial, though you do have the option to just continue through to transcription.

The Freedmen’s Bureau Papers instructions page is organized, comprehensive, and includes examples of common transcription elements and links to outside resources, as well as the general instructions page for the Transcription Center as a whole.

The Frequently Asked Questions section of the page is also downloadable as a PDF, which I thought was a nice touch.

While these three projects certainly have their own strengths and drawbacks, I found them all to be quite user-friendly, with comprehensive resources and a peer-edited revision & review step, through which volunteers can check each other’s work and collaboratively produce a transcription that will then be reviewed by project staff before being marked “complete.” This feels like a built-in QA process, and also ensure that no single volunteer’s quirks (especially if they’re detrimental ones) will mark the project’s entire tenor.

The Papers of the War Department and My (Great-Great-Great-Great-Great-Great-)Grandpa

Most of my discussion of transcription best practices so far has not been focused on the Papers of the War Department, but that experience ended up being my most memorable, and personal. This past winter, I finally started tackling a family archival project that had been put off since my great-grandmother’s death in 2011 – sorting through her numerous photos and papers, which thankfully had been hanging out in a sturdy box in my parents’ guest bedroom closet, which thankfully provided a very environmentally-controlled space in the interim.

In an attempt to figure out who, exactly, was who in the hundreds of images she either took herself or accumulated over the years, (many of which included captions in her neat cursive on the reverse, which my own advocacy for responsible collections metadata practices appreciated so, so much) I did what any reasonable research-happy historian would do – I turned to Ancestry.com. And thanks to the convenient facts that 1. my family has been, literally, in New England for as long as there have been white people in the region, and 2. if there’s a people who love a meticulously-maintained public record structure, it’s New Englanders, I’ve been able to piece together a document record for the Robinsons that goes back 10 generations (while my preference has always been to discover more about the women in this family record, unfortunately traditional Western familial conventions mean that I’m stuck with the dudes, so to speak), or roughly to the 1630s.

One of the more interesting ancestors I’ve been able to turn up has been Elijah Robinson (1750-1826), of which I know:

• Both Elijah and his wife Mary Dike (1748-1822) were born in Dudley, MA, but had moved to Killingly, CT (right on the state’s northeastern border with Massachusetts) sometime prior to 1775.
• After “the shot heard ’round the world” was fired at Lexington, MA on April 19th, 1775, Elijah and the Killingly town militia were some of the first Connecticut minutemen to arrive in support of the Massachusetts rebels. They quickly organized under their neighbor, Major General Israel Putnam (who lived about 8 miles west in Brooklyn, CT) and Elijah spent 18 days “mustered out” during this time, according to the muster rolls in the Record of Service of Connecticut Men in the War of the Revolution.
• I also have enlistment records for various periods and various ranks in 1777, 1778, and 1780, all based out of Connecticut, and what I thought was probably a red herring record from 1799 in Vermont.

Turns out it was my Elijah Robinson after all! Searching him turned up two records, both related to a review of mustered recruits in Windsor, Vermont in January of 1799. Since neither of them had been transcribed yet, I figured I might as well just go ahead and do it! The whole thing took maybe fifteen minutes total (a chunk of which was me reading through supported HTML codes to double-check I wasn’t neglecting to format some of the text properly), and I really liked how smoothly the document viewer pane functioned; it has a crisp zoom & move feature that responds to a mouse scroll wheel, a “reset” button to restore the original display, and rotational arrows, which are immensely helpful for deciphering text written on a slant or perpendicular to the image’s default orientation.

One thing I will especially note from this experience is the fact that the ability of volunteers to produce accurate transcriptions greatly depends on the quality of the original image they are working from. While I had hardly any issue with the Orders for Elijah Robinson, I did have a more difficult time deciphering the faint text visible in the Certification of payments documenting the disbursal of funds by the War Department Accountants Office, which suffered from a grainy black & white scanned image (which itself may have been a photocopy), and also undoubtedly more natural fading of ink than the Orders, which have been digitized at a very high DPI (dots per inch) and in a full greyscale, which (at least in this case) allows for much greater image clarity.