Tuesday, March 06, 2012

Lock, stock and barrel

Prompted by cousin Russ Worthington's When to enter data into your genealogy program and Randy Seaver's Are You an Evidence-Based or a Conclusion-Based Genealogist? two concepts emerge in Ol' Myrt's mind.
  • When to enter info
  • What to do with disproved info
In Oct 2010, I reported on a Second Life discussion of the  Genealogical Proof Standard. "What developed was a serious discussion of WHERE to keep a digital copy of research and analysis during the research process that can take many months or years before concluding. One participant explained she has to flip through her paper pages wasting time looking for an obscure notation. We all agreed that paper note-keeping isn't productive since it doesn't permit a computer to do an every-word search of the contents." See: Note-keeping during the research process.

It was in that blog post Ol' Myrt here stated: "The necessity of a note-keeping program assumes we're not ready to place the information into our genealogy management program. (Ol' Myrt here doesn't like to enter names and info until a family relationship has been established.)"

My reasoning? I wanted to keep my genealogy database "clean", meaning free from error.


What has evolved is my understanding that conflicting evidence and disproved family relationships must remain part of my database, quite simply so I can discuss my reasoning with the next researcher who continues to prefer the erroneous material.

If I delete disproved parents from my database, I also remove my notes about documents attached and inferences drawn. Where else should that information be stored except right with the ancestor in question in my database?

The software I use provides a top level view of conflicting evidence in the "Edit Person" screen, without having everything buried in notes. If I enter three different birth dates, I can see them listed in the "Edit Person" screen. Clicking "Notes", "Citation" and "Media Album" provide access to details.

Like Randy, I grew up on PAF, which forced a conclusion. In the case of multiple birth dates, PAF made me decide which was "correct" and type that date in the only available birth field.

Fortunately the genealogy software Ol' Myrt now uses, permits multiple parents (birth, adopted, step, foster, related, guardian, sealed and unknown) and multiple facts for events (birth, marriage, death, among many others). I can also assign a proof factor (proven, proven false or disputed) for each document, though not for each bit of info provided in that document, unless I am careful to add a "fact" entry for each bit of info. For instance:


Typically a beginning genealogist takes pictures of ancestor's tombstones, thereby acquiring abbreviated information about the decedent. Here we "learn" that Alma O (partially obliterated by the sun spot in the image below) Player was born 1862 and died 1929.

Having visited the Salt Lake City Cemetery with family members, I know this to be my paternal great-grandfather, Alma Oades Player. If I were entering "facts" provided by this tombstone there would be two entries regarding Alma -- his birth year and his death year. Both "facts" will show up nicely in my genealogy program's "Edit Person" screen in date order.

Death Certificate
State of Utah - Death Certificate, State Board of Health File No 2047  indicates the full name of the decedent as Alma Odds Player. Many other documents, and my father's personal knowledge of his paternal grandfather indicate "Oades" is his true middle name, derived from his mother's maiden name.


Oddly enough, I located a 100% typed version of this record today at FamilySearch.org, but I digress.

Using the old PAF genealogy program, I'd have to draw a conclusion about Alma's middle name Oades, though this document doesn't show it. I typically placed conflicting evidence in "notes" for the individual. With my modern genealogy program, I can put in an alternate name, providing "notes", "citation" and "media" (the image of the certificate shown above) for each statement of "fact".

Death Date
Fields 16-18 constitute the "Medical Certificate of Death" portion of the document. Here the physician J. J. Galligan, M. D. certifies the death occurred "Nov 26th 1929". Cause of death was "unknown but natural, no physician attending". No operation preceded death, there was no autopsy. I'd enter the death date and place "Salt Lake City, Salt Lake, Utah" providing
"notes", "citation" and attaching the same certificate image under "media" for Alma's death fact. I can also indicate the death info is "primary" (first hand) information since the physician signed the certificate during the normal course of his work, signed the day after the death occurred.

Birth Date
The death certificate indicates the birth date of Dec 12th 1861 is provided by "informant" Acel Richardson. Now I knew Acel to be a responsible person. He married one of Alma's daughters. However, the 1861 year of birth does not agree with the birth year 1862 from the tombstone previously entered in my genealogy program. Do I change the birth date? No, I merely add a second "birth fact" based on the info from the death certificate. I do not describe this birth info from the death certificate "primary", because the informant, Acel, was most surely not present at the time of his father-in-law's birth, and probably only knows the 1862 date from family tradition. Acel may have had access to the family bible but I've never seen it.

Indeed, Ol' Myrt here needs to review the birth records for Salt Lake County, Utah for the years 1861 and 1862 to determine which is the precise birth date. I'd then enter the information as a "fact" about my great-grandfather, indicating the birth record as "primary" in the genealogy program, since the county birth register is an "official document" created close to the time of the event by someone who has no other motivation than to perform his regular function as recorder in the county clerk's office.

If I find Alma in a "birth index" I'll enter that information in my genealogy program if I am not within the hour also retrieving a copy of his official birth record. The reasoning here is that I may not get to the birth record for a few months or years, and the index is all I'd have to go on at that point.

However, there maybe differences between the index and the actual entry. I've seen this in marriage record indexes compiled by local genealogy societies. I am of the belief that if the information differs, I should also list the info from the index in my records about a ancestor. If the index agrees with the final record, I consider it nothing more than an indexed entry at the end of a book, and don't include the info as a "fact" in my genealogy program's "Edit Person" screen.


In another case, it took me 9 years to act on information provided in an index. My notes for the burial of my Union Veteran William Henry Phillips lists the following info:

"28 Nov 1999. Orlando Public Library. Today I ran across Marion County Genealogical Society's bicentennial project for 1976, Cemeteries of Marion County, Iowa.  Buried at Graceland Cemetery, Knoxville, Marion, Iowa (Old Section) Block 14 lot 6 is Estelle Mae Goering 1876-1921. I was very surprised to see that buried next to her are those I know, from family bible entries, to be her parents. It was also surprising to see that all three died in 1921. Perhaps there was an epidemic in the area that affected them?"

It wasn't until 2008 that I visited the cemetery in question and was able to take pictures of the tombstones. If I hadn't typed the index entry in my genealogy program, the info about the burial site would have been relegated to the obscure reaches of the 116 research binders in my collection. 

Because I entered the burial data for Estelle and her parents from the index entries in the book Cemeteries of Marion County, Iowa, I was able to retrieve the data when creating a place-sorted list in preparation for my on-site research trip. Allowing my computer to locate data for me is exactly why I use a genealogy program.

Back to the example of Alma Oades Player's conflicting info about his birth date. According to the Genealogical Proof Standard, one must resolve conflicting evidence. I may conclude the tombstone carver made a mistake, or that information provided by Alma's son-in-law was skewed owing to the stress of the grief felt by family over Alma's death. I may find additional info that agrees or disagrees with the dates in question. All pieces of the puzzle must be analyzed.

If the carving on the Alma's tombstone turns out to be incorrect, I certainly don't want to remove the tombstone picture from my compiled genealogy. I'll merely include my analysis, sometimes called a proof argument, about why I believe date XYZ is accurate, in the notes field for that fact for Alma.

Everything's in there, lock, stock and barrel.

Happy family tree climbing!
Myrt     :)
Your friend in genealogy.










  1. Dear MYRTLE,

    Great blog post.

    Thank you,


  2. Pat, interesting synopsis of your changing perspective on maintaining unproven or disproven data. Your comments, along with Randy's and others, casts a darkening shadow over the limitations of today's crop of most commonly used genealogy software.

    Although most genealogy software today is not as inflexible as PAF for documenting conflicting data, virtually all of it is still oriented toward documenting conclusions rather than organizing research. For instance, the current version of Family Tree Maker (2012), continues that application's small-minded behavior of recognizing only marriages and marriage-like relationships as the only type that can be created and documented in any meaningful way. You want to document neighbors in an intriguing census entry? Sorry, can't do it. You suspect Agatha is the daughter of John and Mary Smith and would love to create a tentative relationship in which to store your research? Can't be done. In fact, it does not provide a decent way of actually documenting even parent-child relationships! How 1990s!

    Makers of genealogy software, please take note: increasingly, serious genealogists consider themselves researchers, not data-input specialists. They want to track their research, document their evidence, and maintain their conflicting findings along the way. Genealogy software of the future should accommodate and enable these aspirations, empowering research rather than hindering it.

  3. So then, do you upload the gedcom, with all the unfinished research to genealogy websites? And if everything isn't worded exactly right, won't they assume your gedcom is your conclusion?

  4. James, I know that Family Historian supports the GEDCOM ASSOciation tag to document relationships because I use it extensively. My website AncestorsNow.com supports these relationships tag as well.

  5. This was a great post! I maintain my data simlar to your description in this post. I got a few tips for improving it from your post and I thank you very much. I am always looking to do it better. May I ask what software you use. I am using FTM 2012 and can do most of what you described but not all. I am looking for something that will do what you described.

  6. I agree with James. Here's what I wrote on another site:

    I think the problem with a lot of the gen software assumes that the user will be making some sort of end product with it and tries to provide those outputs. My problem is that I use my database as a work-in-progress, and I’d like it to support me as I research both online and offline.

    For example, when I’m on site, I take photos of [or scan] docs and related materials with my iPhone and an app. Then another app [integrated with the first] allows me to make a quick copy of the scan, mark it up with notes and citation info [if I want to do this at that time], and then the 2nd app allows me to upload it to an online storage service that syncs with all of my electronic devices.

    Additionally, I’ve come up with alternative methods while I’m at home and online searching with my laptop for capturing information, annotating, transcribing, etc.

    In both these processes, I think gen software has not met up to my particular workflows. However, I’ve been looking at FTM 2012, and it seems to mirror what I do pretty well. [But it has other drawbacks.] I want my database to be full of my clues, conclusions, evidence [both in note form and image form] simply because it’s where I’m working from, and my research is ongoing. I don’t look at my database as an finished product.

    If I buy gen software, I shouldn’t have to use all these other technologies to augment the software.

    However, my goals are to be completely digital; to have my offline database sync to my online database; to be able to digitally attach notes, annotated images and ‘clean’ images to both facts and people; to be able to digitally share my info with others [but not in a GEDCOM]; and to have my data available on all my devices at all times. And to do it all with ease and efficiency. [Meaning less steps, not more.]

    Therefore, I think it depends on what the researcher’s goals are with their database, and what they want it to be able to do for them. Also, I think the goal of being completely digital is much easier for those who are newer to researching.

    And, in answer to Sheryl's question: I use Ancestry.com's trees, but keep it private. In this way, other researchers can contact me. However, I don't think one can control what another person does with the info that you provide them. When I'm providing data, I present my notes and supporting evidence to help explain my conclusions.