Dear editor;
I'm reviewing some of my Research Reports and Notes and see that I'm constantly struggling with recording forms-based items (eg. a 1926 Ontario Birth certificate) in a complete and consistent manner. While paragraph-based items (eg. wills and letters) readily lend themselves to transcription, form-based items do not seem to do so. (And; the formatting restrictions of the editors in many genealogical programs makes it nearly impossible to lay out the transcription of a form.) I "think" that I should likely be using an abstract for these situations. Abstracting paragraph-based items, by dropping boilerplate, is a bit more straight-forward. But; abstracting form-based data is something that I'm not sure how to tackle, due to its lack of "flow" to the information presented.
I read over the QuickTips article, "Abstracts vs. Abstracts, with a Bit of an Extract." I suspect that I should likely use something that follows the example for a, "notetaking context," rather than an "academic context," in order to retain as much information and original context as possible.
If abstracts are the correct solution; would you have any short examples of a form-based record and the resulting abstract? Seeing how someone else has tackled this would help me to see the path forward.
The term "abstract," as used…
The term "abstract," as used in academic papers, is not at all appropriate.
There is, of course, no standard format for extracting or transcribing data from a form. It depends upon the structure of what we're working from and, as you say, the formatting limitations of databases not designed to be word processors. That is one of the reasons why I do most of my data collection within a word processor.
I'll paste in several examples of formats from reports that I've posted online at my website Historic Pathways under the "Research" tab. (This is the website I use to archive online my published articles and unpublished research reports).
Most often, I put the column header at left, add a colon and a tab, then transcribe the data in that column. When I do a cut-and-paste into the "notes" section of a relational database, the line breaks usually survive but the tab space between columns 1 and 2 may disappear. The colon after column 1 will remain to indicate the break between (a) column headers or pre-printed data; and (b) the data that was penned into the blank space on the form.
Sample 1. Data extracted from a death certificate:
Depending upon what my reporting needs are, I may put all items in Column 1 in boldface, with column two in normal face. This also helps to make a visual break between data when using a database that strips out the tab. (In the example above, I needed to use bold face to emphasize a set of date problems. Thus, column 1 is in normal face.)
Sample 2. Data extracted from a tax roll with columns:
Sample 3. Data extracted from a more-complex tax roll whose entire slate of column headings needs to be preserved to identify what the person-of-interest did not own, as well as what he did own:
Sample 4. Data extracted from a U.S. military enlistment register with columnar headings:
Sample 5. Transcription from a U.S. patent whose information is in paragraph form rather than columnar form:
Like you, I have absolutely…
Like you, I have absolutely all of my raw research (transcriptions, abstracts, citations, notes etc.) in word processor files. Bringing that information over into a genealogical database has always presented issues. Your examples and explanation of your methodology are very much appreciated and should help alleviate my concerns with transcribing and abstracting form-based data.
I reviewed my old notes on one of your presentations; "Info Overload? Effective Project Management, Research, Data Management & Analysis," NGS, Raleigh (2017). (Excellent presentation, by the way.) In it there was an example of generating the Individual Research Notes reports via the genealogical database by cutting and pasting from the word-processor-based Research Report. Now I understand that you also did some "massaging" while inputting the data. That was what confused me, since I seldom have something I can simply just cut-and-paste verbatim. I will assume you also "massaged" the report output just a bit more, prior to putting it up on the Pathways site, since the example reports are much better structured than one might expect directly from the program.
(As a side-note; I have, on occasion, used "Multi-Markdown" notation for simple tables in my Research Notes and Comments. I saw this specified in the Smithonian's transcription guidelines and recognized the format they had adopted. It allows one to capture simple tables in pure text format; with column headers that can still be easily recognized. I've also had some decent success in post-processing the resulting reports in order to render the tables in the way that one would see them in a word processor. Your method is likely easier for more complex situations and I'll likely go with that as a default approach.)