Tuesday, November 22, 2011

Tuesday Technological Tirade - Citing and Documenting Sources

I read with interest Randy Seaver's blog post, "What's Bad about Genealogy Software" http://www.geneamusings.com/2011/11/whats-bad-about-genealogy-software.html, which is a discussion of Louis Kessler's Behold blog http://www.beholdgenealogy.com/blog/?p=874 and his list of items he finds disconcerting about the current offerings available to genealogists. 

The one thing that I found most disturbing about this discussion was the shifting of emphasis from using the gold standard for citing sources as published in Elizabeth Shown Mills' work "Evidence", and encouraging genealogists to just document their sources and then draw their conclusions. Perhaps my reading of Louis' comments was too literal, but I do not think so. 

There is a reason that Mills' tome is the gold standard and should be on the desk of every serious genealogist out there. The field needs and will require a continuity of standard for citations. To say that one can document a source and then draw infinite and proven conclusions does nothing to improve the field-wide standards we all seek that will allow us to duplicate and double-check any prior work done by another genealogist. It makes a difference "when" you viewed a certain set of records, "where" you viewed them, and the "context" in which you found yourself analyzing them. Records are lost, transcribed incorrectly, moved, re-ordered, and handled every single day, and it behooves us to track ourselves so that others may ultimately track our research.

To encourage genealogists to just document, rather than carefully create  proper citations is a step backward, in my humble opinion, and does nothing to further the standards so desperately needed in genealogy. We are striving to better our information and documentation and analysis rather than relying on the research practices of yesteryear. Websites are chock-full of "conclusions" that are drawn from poorly cited, rarely documented information that is basically meaningless except to serve as an example of what NOT to do with your family history. There is a proof standard, and it should be applied to every bit of information we gather, especially before it is set out for public consumption. 

Granted, some genealogists are perhaps not as serious as others about their family history, but for the sake of generations to come, instead of making things less cumbersome for the casual family researcher, we should embrace them and help them, and teach them how valuable their information is, and in what fashion it can best be preserved. I do not think this point can be stressed enough. You simply cannot have documentary proof without proper citation. An experiment in any other field must be reproducible. Genealogy should be no different if it is to be worthy of its legacy.

Louis, we need you to keep working on the GEDCOM conversion standards. Your sorting of tags and the things that I see as "gobbledy-gook" most certainly must be improved if we are all to preserve our work in a communicable manner. But let's keep the standards high in all aspects of what we do, and whichever part of the field we're involved in. 

Thank you, Randy, for your wise analysis of Louis' post. 

And yes, I do have a copy of Mills' book on my desk at all times. We all should. 


  1. Kellie,

    Nice post, and I do agree that citations are important if you are planning to publish your genealogy. But they do take valuable time to do right, aren't perfect, and add complexity that can cause confusion and mistakes. Just selecting which template to use and how to fill in the parts of the template is not easy to decide on and is prone to being done differently. Shown Mills has a wonderful guide, but it is not a clearcut set of instructions that will make everyone do it the same way.

    My complaint is not with programs including citation templates, but it is with programs including them in a way that makes sourcing take much longer than it should. Genealogy programmers have to do everything to make adding sources as easy as possible to encourage everyone to do it. To me, single fields that add the specifics about the source are very important, and this could be done with a much simpler, almost GEDCOMish level of source referencing with maybe twenty fields (book, author, publisher, call number, url, where within source, page, etc.) We aren't dumb. We can discern which fields we need to use for which source and don't need a template to do that for us. And at this level, we don't need to worry about the formatting: where to put the punctuation, italics, emphasis, capitalization or what order to put them in. The source information is what's important first and foremost.

    So if the source can be added quickly and efficiently, then I say prior to formatting the source for publication, the important next step is to use the sources as evidence and build your conclusion. Then, document those conclusions. And the programs out there today have no capability to do that.

  2. Louis,

    On this, we may need to agree to disagree.

    Accurately stated and standardly formatted source citations are not just important for publication of a family history. Many genealogists never publish short of printed summaries or ancestral charts that are passed out at family reunions. That does not mean, however, that the proof that is said to exist should be scrutinized by other researchers with any less precision. Proof is just that. It matters not what “level” of work is being done, or at what “level” the genealogist has experience. Uniformity and complete analysis should not be discarded in favor of making things easy, because genealogy often is not.

    When we gather information, or evidence, we wish to use that information to prove or disprove a theory or to fill in missing information. Our analysis begins with the information itself - the document, microfilm image, cemetery transcription - whatever it may be. Though that document may appear on its face to be "proof" that fits our genealogical need, it may or may not be the best proof or evidentiary item to confirm or dismiss the fact in question. Our analysis and consideration of the item must be careful and deliberate, and our efforts exhaustive in order to obtain a sound conclusion.

    This is where the citation for the source comes in. It is important to note the provenance of the document, the date upon which we viewed or obtained the document, the condition of the document, and its reason for existing in the first place. We may obtain a microfilm image of a document that may differ from an original written 100 years earlier. We may then inherit the original of that same document and find that the clerk erred in transcribing the original court copy retained by him. It matters when, where, from whom, and in what state or condition and under what circumstances these bits of information are gathered. Without those elements, we do not meet the Genealogical Proof Standard, and in essence, have really “proved” nothing.

    I agree that we are not dumb, that we basically know what fields we need, but we still can benefit from standardization, time-consuming or not. It is becoming more and more important to hone our accuracy and our interpretations through the use of the GPS, and thus, our citations of same. Again, I would suggest that rather than ignoring these basic research tenets, that we would all be better served by increasing and sharing our knowledge with those who are unfamiliar with the current gold standard. Let’s forge ahead and educate one another to continue to create a uniformity from which all will benefit, rather than making things faster and easier.

    Just as you have decried the differences in codes for GEDCOM files, genealogists decry the unsourced, uncited, and nonduplicitous nature of the information they see. Neither situation is convenient, nor of recent development, but both can be improved upon by us as we strive to share the standardization we seek. The Genealogical Proof Standard is our experimentational guide to the often unholy experiment that is genealogy, and we absolutely cannot accurately reproduce a colleague’s work without proper citation and analysis.

    Best of luck to you on your launch of Behold 1.0! :)

  3. Fair enough. I'll continue in my goal to make documentation of evidence and conclusions easier for genealogists, and you can continue your goal to formalize proper citations.

    If you could help figure out a way to make that formalization easy, without providing 150 different structures that no two programmers will ever implement the same way, then I'll be very happy to help take it forward.


  4. Hi Kelly,

    I, too, am a fan of Elizabeth Shown Mills and Evidence Explained. I also really like the work produced by folks over at the Register, who publish well-developed citations using a style that is a little different. To my knowledge, both styles have been influenced by Chicago (CMOS).

    I can point to cases in my family file that have been "solved" by applying Mills' principles. I use citations to document my work. If we are talking genealogical software, that means I develop a programmed citation. Unfortunately, my hard work is barely recognizable when transfer my file to a generic Internet service or share a GEDCOM. The transfer is so bad that I usually limit what I share to birth, marriage and death events and the related citations ... but that often leaves out tags or facts that are important to the "proof."

    For almost a year, I've been working with the folks at BetterGEDCOM on citations. As you probably know, that's an open, international group. (You probably also know that Mills is a US Centric style, but oddly enough IMO, not many US techs have chimed in at BetterGEDCOM to support the technology work on Mills style.)

    Some of the points Louis makes are spot on--no two vendors made the same implementation of Mills, and that holds true functionally and technically. (=My findings, knowing that each vendor invested substantially to develop the functionality.)

    I have my own opinions about why some developers find Mills complicated to implement in genealogy software, and probably even more opinions on what should happen to make the style more usable in a world-wide sharing and syncing environment. I'm not the only person with opinions, though.

    We are closing in on decision making about the BetterGEDCOM framework for "Sources and Citations." I hope you'll drop in on our wiki to at least comment to show your support for the citation. --GeneJ