* FH V7 Source Citation Options

[ColeValleyGirl]

Paul,

CP have unfortunately not included the ability to import Autotext for Plugins -- they assumed that plugin authors would always generate it via code. I have explained why that doesn't make sense, but a change hasn't appeared yet. If it doesn't appear, I may amend my plugins to allow them to download and install templates as a workaround.

If you start the DEA, you should see a selector labelled Text template (Top left) to select the template you want to use (or none if you don't want to use a template).

[paultt]

Helen, thanks for that.
Made me think on how to start the DEA if we are not supposed to run it as a plugin!! Worked it out by Prepare a Citation to an Existing Source, selected my Source and then ran it at the bottom right, and saw that it showed the Death DEA.
Ran through it in my sandbox project. I could possibly accept it for adding information from a printed or pdf death certificate, but for anything else, the thought of trying to work out what needs to be changed in the DEA's 1700 odd lines of code to make a new DEA that would be functional, would at this stage be way beyond me, and far too time consuming for the value of the output. Not for beginners, unless they are programmers by trade!
As it is now, just by going from v6 to v7 has added so many extra time consuming keystrokes that it actually takes longer manually to input basic generic source and citation info, whereas FH was known for it's ease of use and simplicity.

[ColeValleyGirl]

Paul, I'm curious -- what do you want the DEA to do that would mean you needed to code anything? If there's data capture needed for some form of death document that I haven't covered, I can add that very easily and wouldn't expect anyone to do their own coding. A later version will include a very few options -- definitely "splitter or lumper", but I don't have anything else on the list of options for Death to implement.

If all you need is a different format from the text from source, creating a different autotext template ought not to take very long, if you've got an example document in front of you.

If, on the other hand, you want the DEA to do something other than create facts and relationships from information derived from the source, then it would be a different DEA.

[jmurphy]

There is a step missing from that process flow. Neither what you describe as "source-driven" nor "fact-driven" start just with the source. We don't pick an arbitrary document and see if it relates to our tree. We are interested in the source because we believe it supports a new fact. Usually we find it as a result of looking for sources to support postulated facts (finding a family in a census, baptisms for a new family, etc).

IMO, whether we enter that fact first, then the source, and finally other facts supported by it and cited to it, or enter the source first and then all facts at the same time, is neither here nor there - it is personal preference, and one is no more "correct" that the other.

I agree that what you describe as source-driven is easier to automate than fact-driven, but the decision on whether to use AS or DEAs is about whether users prefer to make the link from source to fact manually or use some form of automation (with varying degrees of flexibility). That is not the same as the difference between fact-driven and source-driven entry. AS/DEAs probably require entering the source first, but using a source-driven approach does not require AS/DEAs.

This may muddy the waters somewhat, but I'd like to describe the workflow I use and make some comments about why I work that way.

The FamilySearch Wiki describes the Research Process as a cycle of five steps:
1. Identify what you know
2. Decide what you want to learn
3. Select records to search
4. Obtain and search the records
5. Evaluate and Use the information

The graphic shown the Wiki article has an arrow from step 5 to step 1, closing the circle. https://www.familysearch.org/wiki/en/Research_Process

Another way of looking at the big picture is to use a flowchart like Genealogy Explained's Genealogical Research Process: How to Conduct Research using the Genealogical Proof Standard.

The research process described here is a person-centric or person-focused process. Entering the data into a lineage-linked program such as FH is part of step 5, evaluating and using the information. We have concluded the records we've collected are likely to belong to a person of interest who is in our database before we enter them into FH.

Pardon me if the following seems really obvious, but if you are just starting out, and are new to genealogical research, it isn't obvious at all. I have yet to find a truly satisfactory way of keeping track of the information inside the "maybe" records where you aren't sure if they belong to your person or not (e.g. while sorting out cases where you have multiple people with the same name). Lineage-linked databases, for the most part, aren't designed to organize the "multiple maybes".

The other assumption here which may not be said out loud is that we are often trying to solve a particular research question, such as finding the marriage date and place for a particular couple. It is tempting, once we have such a record, to enter the date and place and make sure we've cited our source, and call it a day.

However, I was trained when doing linguistics to capture ALL of the data in a source, because we were studying ALL of the language we were collecting. We may have an immediate research question, but we're also going to want access to the other information in that source when we have a different research question later on.

Because of my training, once I have a record in hand, I take a source-centric approach. I learn about the records so I can evaluate the information inside more effectively, using the Evidence Analysis Process Map and other tools such as the information inside Evidence Explained.

TL;dr: ESM's view is that a Source is a container that holds information. It is useful to analyze both the container and the information separately in order to get the most thorough analysis; information becomes Evidence when we apply it to solve a specific research question.

In Family Historian (I'm referring to FH3 through FH6 here since I haven't installed FH7 yet), if I'm working manually, I create the multimedia object first, then the Source to point to it. Once that is done, I turn on Auto-Source Citation and extract the data from the source, associating all the data with the appropriate people and events. Ancestral Sources has, of course, been a huge help with data entry for census and other records that have multiple people in them. If I'm doing a census record, I'm likely to use Ancestral Sources to do the basic data entry, then use Auto-Source citation within Family Historian itself to extract the information in US census records that don't correspond to the basic household data. US Census Data varies much more than data from the UK (see links at the bottom of the post).

Before I used FH, which accomodated my source-centric approach, I found it was all too easy to slam data into the program first, then worry about how I should cite it. In a question on Genealogy Stack Exchange, one of our community members asked What tools exist for collecting and managing evidence?

Are there any good systems that take an evidence based view, where I can enter all the information I have and then draw my conclusions to produce a pedigree rather than producing a pedigree with source citations hanging off it?

Another person on Genealogy SE asked How many sources/citations is too many?, a question which also arises from wondering what information in a document is 'worth making a citation for', instead of the view of capturing the most possible information from your sources and recording where you found it.

I am greedy. Once I have a record in hand and I've concluded it belongs to my subject, I want ALL of the information I can get out of that source into my database so I can use it for other analysis. If I record only the information that I wanted to answer a specific research question and nothing else, I won't remember that I have that information when I need to to answer a different question.

So I extract everything, with FH's Auto-Source Citation being my dutiful assistant, noting which source contained the information I'm extracting.

Further reading:

Elizabeth Shown Mills' post EAM & GPS: Newsflash! Siblings, not Twins
Michael Hait's Why we don’t always need source citation templates … and … but we do need Evidence Explained.
Enumerator instructions for US Census records from 1850 onwards, via IPUMS-USA.

[paultt]

Helen, to satisfy your curiosity, and what I mean about having to modify or create DEAs. The one you have will most likely satisfy an English Death certificate, but I would probably tweak it just a little bit to satisfy my needs, as I suspect others may do aswell for anyone who died in England. That is only one instance of a Death document, that now has a workable DEA.

I maintain a website www.1820settlers.com as part of my hobby, which is trying to trace all of the 3900+ individuals who were registered as being members of the 1820 Settlers to South Africa, what happened to them and their descendants. I began this challenging hobby in 1968 when my wife's grandfather told me that she was a descendant of one of the settler families, Miles and Anna Maria Bowker. They came from Northumberland, and sailed from Portsmouth January 1820 to settle in the Eastern Cape of South Africa. To date I have managed to trace just over 3000 of the original settlers and their descendants and have approx 175000 individuals in the project gedcom.

In South Africa, they have 4 major provinces, and each one has a different variant of a Form of Information of Death, and a Death Notice, and each of them have changed their format over the past 200 years at least 3 or 4 times, and can be produced in English or Afrikaans (Dutch) depending on the Province and years, which gives me a maximum of 64 variants of the forms for South Africa. The original settlers descendants have migrated to neighbouring countries as well - Lesotho, Namibia, Botswana, Swaziland, Zimbabwe and counties further north, and each of those have their own Death Notices in some format or other.

As an example pair, I have picked one of the settlers, Septimus Bourchier Bowker,( yes he was the seventh son ), and these are links to his documents pertaining to his death:
Form of Info of death : https://www.familysearch.org/ark:/61903 ... cc=1779109
Death Notice: https://www.familysearch.org/ark:/61903 ... cat=331262
Currently in my workflow, I attach an image copy of these forms to the Citations made for the Sources applicable, giving the links and making them available if I have found them, for other researchers to view the documents and make their own minds up on what the want to capture for their own systems and family trees.

If you would like to see a more complex death Notice, have a look at this one: https://www.familysearch.org/ark:/61903 ... cat=331262
Yes, Daniel did have 21 children from 2 wives, but note that they have given the birth dates of all the children. Sometimes if a Child has married, had children and died before her parent(s), they would list all the grandchildren and their birth dates of that deceased child.

So, correct me if I am wrong, but I foresee that I would have to seriously modify your Death DEA, or create multiple variants to cope with just the Form of Info of Death and Death Notice documents outlined above, before even considering the multitude of variants per Church religion, per Province, per Language of Birth, Baptism, Marriage and Burial records available in Southern Africa.

Not being a Lua programmer, I think this task would be beyond me, and hence I think I will stick to my Generic sources and current workflows, thank you, and I wish you the best of luck in your efforts.

[ColeValleyGirl]

Paul, thanks for that. I'm grateful for your input, even though you're probably not going to use the results.

Yes, if you wanted to use the DEA for the first example you give, it would need to be altered to collect the information about the children's ages and count, and the duration of the last illness, but it couldn't do anything useful with those except include them in the Text from Source (which you would have to create the template for). The changes wouldn't be massive -- and they'd all be in the table-driven section of the Lua that controls the data collected -- but even that sounds as if it's a step too far. (If you needed to make changes to collect extra information and create additional facts, that would be a bigger task with some proper programming).

You wouldn't need to edit the plugin at all if all you wanted was a different template for Text from Source, either creating a new one or customising an existing one. That would all be handled in FH using Tools > Manage Autotext ? Autotext for plugins.

(An alternative for handling additional content for the Text from Source would be to leave the plugin as it is, and edit the additional content in manually).

The other two documents are Probate documents (I think?), so it wouldn't make sense to use a DEA designed around Death Certificates for them. (I don't think a DEA for Probate Records would make sense, except for somebody doing an awful lot of work with Probate records from a single jurisdiction).

That is only one instance of a Death document, that now has a workable DEA.

The same DEA works for death certificates from England & Wales, America (various states), Canada (various states), New Zealand, Scotland Isle of Man, Channel Islands, Ireland, Northern Ireland (I haven't got any other examples in my personal stash). All that changes is the template selected for Text from Source, and the fields which the user leaves blank because they're not included in the source they're working from.

[paultt]

Helen,
Extra piece of South African complexity. The first doc Form of Information of Death only came into action after 1894. Prior to that the Death Notice was the only death notice/certificate issued, and it has a question about the value of the Estate being more than £300 and whether there was a will or not. That was the limit to decide whether a Probate was necessary or not. The Death Notice is a valuable document for family historians as it has so much information on it.

PaulTT

[AdrianBruce]

... The Death Notice is a valuable document for family historians as it has so much information on it. ...

Thanks for that Paul - one of my families went out in the 1820 lot, so I've used your site extensively (the Pickstock family, if any asks) but up until now I've misunderstood the phrase "Death Notice", equating it with the totally informal report of a death in UK newspapers. Clearly a bit more to it than that.

[paultt]

Adrian,
Just done a quick search for you and found Mary Ann Gravett's Death Notice:
https://www.familysearch.org/ark:/61903 ... cat=331262 so this should make a bit of sense to you. Unfortunately, the person who filled in the Death Notice was a son-in-law from Thomas Berrington's first marriage and he may not of known that Mary Ann's mother was actually Charlotte Pickstock, and just took a guess that Mary Ann was named after her mother. Luckily this d/n is typed, some of the handwritten one's are shocking!
Glad someone is getting use out of my website

[AdrianBruce]

Thanks for that Paul - my image(s) of Death Notices so far (only a couple) are from the eGSSA Library - I'd not worked through to the FS films yet so that connection / demo might be useful.