Jump to content


Photo

How best to document evidence of parent relationship?


  • Please log in to reply
25 replies to this topic

#1 David Gorman

David Gorman

    New Member

  • Members
  • Pip
  • 2 posts

Posted 07 January 2017 - 04:28 PM

New to RootsMagic, am former TMG user, and need help doing a minor thing.

 

I am used to mining a source for all the facts/evidence on it, and don't know how best to do this in one particular case with RootsMagic.

 

I have a death certificate which lists the deceased's birth parents.  So, this is evidence of the parental relationship between the deceased and each of the deceased's parents that I want to capture.

 

In particular, I do not want to just document the parent/child relationship as a proven fact, since I anticipate other documents may offer conflicting evidence and I want to weigh the preponderance of the evidence after some more evidence has accumulated.

 

Here's how I approached it (but this approach has left me unsatisfied).

 

I first went into the deceased's "Edit Person" screen.  I then added a source to the Birth fact.  I then shared this fact with both parents.

 

The problem is now when I go to the father's Edit Person screen, although I see the shared child's Birth fact, that fact lists every source for the fact, most of which don't pertain.  In this case the only evidence is the child's death certificate, not the whole stack of source citations that buttress the birth.  I don't want that because it falsely suggests there is a great body of evidence for the relationship, but I only intend to document this one source as evidence.

 

I imagine I'm approaching this all wrong.  How should I approach doing what I'm trying to do?

 

Many thanks in advance.

Dave Gorman

 

 



#2 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6146 posts

Posted 08 January 2017 - 06:39 AM

You could create a custom ChildParent fact type, add it to the child and share with the parents. Tag the citation to it.


Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#3 Jerry Bryan

Jerry Bryan

    Advanced Member

  • Members
  • PipPipPip
  • 3405 posts

Posted 08 January 2017 - 05:12 PM

I think your frustration with this problem is because nearly all genealogy software (including the GEDCOM glue that holds various pieces of genealogy software together) is very deficient in providing a good place to attach evidence of parentage. TMG and SecondSite may be exceptions to my general rule, but I think that RM certainly has the problem. It's interesting that when RM's direct import from TMG imports evidence of parentage, it creates a ChildParent fact in RM to which the evidence of parentage from TMG can be attached. Tom's very excellent suggestion is proposing the same thing except that you would do it from the RM user interface instead of via direct import from TMG.

 

When RM's direct import from TMG creates a ChildParent fact, the evidence is attached to the fact but RM's sentence template for the fact is null. That means that unless you as the user take some sort of additional action at this point, a narrative report will include the evidence but the citation superscripts for the evidence will be sort of dangling in midair and not attached to any sentence in the report. With Tom's suggestion, you could certainly make an actual sentence as a part of the ChildParent fact instead of leaving the sentence template as null. Also, RM's direct import from TMG does not create any roles nor share the fact with the parents. You could certainly add this little nuance yourself.

 

The date for the fact would also be an interesting question. I would probably leave the date blank and include a sort date to force the fact immediately after the birth fact for the  principal. There are other reasonable ways to deal with the date, but you would want to give it some thought.

 

The most interesting part of this puzzle to me would be what would happen with the roles and sharing and what would the data look like for the parents. For example, for the father you might set up the role and the sharing and the sentence something like "John Doe became the father of a daughter Sarah Doe on June 25, 1895". Well, to do that I guess you would have to give the ChildParent fact a date and not just a sort date.

 

However, I have been looking at this problem for a long time and I have not yet decided how best to address it. For example, I have considered introducing a fact called Childparent (or a fact called simply Parent or Parentage, etc.) without actually sharing the fact with the parents. I have long been very reluctant to use RM's shared facts because they export so poorly to third party genealogy software. TMG emigres probably don't have this same reluctance, so you might wish to implement Tom's suggestion more or less as is. But whatever you do, do play with it in a test database and see what the sentences look like and what reports look like etc. before you make final decisions. And by the way, the new GedSite software by the author of SecondSite handles RM's shared facts quite well.

 

But the final straw that has prevented me from adding the ChildParent fact to RM is that RM's narrative reports already include the parents of the spouses. So for the spouses, I would end up with duplicate sentences for the parents. What I call the "spouse sentence" in RM's narrative report is a major pain for me in other ways as well because it is not under complete control of any RM sentence template, so I'm still trying to decide what the best way is to enter evidence of parentage into RM.

 

Jerry

 



#4 zhangrau

zhangrau

    Advanced Member

  • Members
  • PipPipPip
  • 1473 posts

Posted 08 January 2017 - 06:33 PM

I don't know what you actually seeking, but if an "individual" ParentChild fact bothers you with its duplication in the spouse paragraphs, how about trying a ParentChild (family) custom fact?

 

I dpn't use such a custom fact, but I do paste a citation wherever and as often as it makes sense to me. If parents names are documented in a birth notice (newspaper usually), birth certificate, or baptismal record then you'll find the citation for that event pasted to the birth, baptism and alternate name for the child; and also pasted to both parents' alternate name (and when recorded: age, birth place, occupation, residence, etc.) I feel this does a suffiecient job of documenting the relationship.



#5 SomebodySmart

SomebodySmart

    Advanced Member

  • Members
  • PipPipPip
  • 112 posts

Posted 08 January 2017 - 07:33 PM

Are you trying to distinguish natural versus adopted?



#6 David Gorman

David Gorman

    New Member

  • Members
  • Pip
  • 2 posts

Posted 08 January 2017 - 08:21 PM

Thanks all, for your time helping me to think this through.

 

What I'm struggling with a bit is the fact that TMG and RM just implement facts differently, and my habits were set by TMG.

 

c24m48, I agree with you that RM is "very deficient in providing a good place to attach evidence of parentage."  That's a hoot!  Genealogy is, after all, the establishment of parentage!  Parentage seems to be the main thing, so that's an odd thing to be deficient in.  That just sounds crazy to me.

 

SomebodySmart, no, am not trying to distinguish natural vs. adopted.  I'm just trying to document all evidence I run across, and when I have good evidence of parentage, the place I'm used to documenting that doesn't seem to exist.  I don't want to throw away the information or create custom tags so that I can enter a second time information which has already been entered.

 

You can see what I'm after in the Timeline tab, which is similar to the Person Edit view except that it includes birth, marriage, and death facts belonging to immediate family members.  So "Birth-Child" is depicted automatically, for example.  That's what I want, only I expected it on the Person Edit view.  Also, I expected the fact to be editable independently of the birth fact, which might have many citations/evidence of the birth, but very few of which might also provide evidence of the parentage.  The parentage and the birth are separate things!  They should be citied separately, because one might prove the birth but not have proved the parentage.

 

TomH, your idea of creating a custom tag for ChildParent seems to have an issue: it loses the automation that should be there...the software should make that tag automatically when I tell the software that A is a child of B.  I don't want to manually (laboriously) create all these tags, and then have the issue of getting the narratives squared away, etc.  The automation is present in the Timeline view.

 

At the root of what I'm struggling with is that RM has a category of facts called "family facts", whereas TMG only has "facts", they are all individual facts.  In TMG the "family group sheet" is just a view of the data model and not it's own data construct as it seems to be in RM.  I'm not sure why RM even has family facts, it seems that all facts should belong to individuals.  

 

What in RM appears in the Timeline view as "somebody else's fact" (e.g., "Birth-Child") really belongs to the individual, not the immediate family member (child), because the individual sired or birthed someone, which is a significant event that in my view should be automatically reflected in the parents set of facts (as it is done in TMG).

 

I suppose I will move on to more productive things, now that I see there isn't a simple way of doing what I was trying to do.  I'm old enough to know better than to try and cram square pegs into round holes.  But I'm rather impressed that software intended for serious genealogy cannot capture easily partial, conflicting, or tentative evidence of parentage (which is the main thing in genealogy, isn't it?).  I hope RM will fix that in the future.

 

zhangrau could you please clarify what you meant when you say you "paste a citation"?

 

Thanks again,

Dave Gorman



#7 Jerry Bryan

Jerry Bryan

    Advanced Member

  • Members
  • PipPipPip
  • 3405 posts

Posted 08 January 2017 - 10:18 PM

At the root of what I'm struggling with is that RM has a category of facts called "family facts", whereas TMG only has "facts", they are all individual facts.

 

Indeed, this is the crux of the matter. But it seems to me that the RM data model is like virtually every other genealogy software in the world including GEDCOM, except for TMG and SecondSite. Never mind that the TMG data model may be better, it is the outlier.

 

In the GEDCOM model, there are individuals and there are families. Each individual has an INDI record. Each family has a FAM record and/or FAMS/FAMC records. Parentage consists of children with INDI records standing in FAMC relationship to a family and of parents with INDI records standing in FAMS relationship to the same family. But somehow or other that doesn't seem to me to provide a place to hang the parentage evidence.

 

Well, following that thought very literally, I suppose there actually is a place in RM to hang parentage evidence. In RM's Edit Person screen for a person with parents, there is a line towards the top of the screen labelled Parents which lists the individual's parents. That line includes a box for sources and if you click on that box it will open a screen to add sources for the family of the individual's parents. The problem is that in any practical sense, such sources are really associated only with the parents and there is really nothing about such sources that tells you as an RM user or that tells any reader of any of your reports that such sources are evidence for parentage. That's why RM's direct import from TMG creates a ChildParent fact and why Tom was suggesting that you might think about creating a ChildParent fact. That's perhaps a harsh assessment, but I think you would run into the same issue with FH or FTM or Legacy or Reunions or Gramps or any other similar package except TMG. I could certainly be wrong, but that's the way it seems to me.

 

Jerry



#8 KFN

KFN

    Advanced Member

  • Members
  • PipPipPip
  • 206 posts

Posted 08 January 2017 - 10:44 PM

Dave,

Most genealogical software is not used for collecting information, rather it is used for displaying conclusions. This was one of the unique features of TMG. Most programs are based on a loose interpretation of one schema, while TMG created its own schema.

I'm not saying this is good or bad or that TMG is superior, or that any other program is inferior, (no flames please!). You will need to adapt.

Personally I have two sets of records and databases 1) my database that I share with my conclusions based on the facts I have, 2) my documents that back up those conclusions and any suppositions that are not yet in the #1 database.

What you should have is a attribute associated with the child to parent/family indicating a status code that allows passing on the users opinion of the status of a child to family link, proven, disproven, or challenged.

#9 KFN

KFN

    Advanced Member

  • Members
  • PipPipPip
  • 206 posts

Posted 08 January 2017 - 10:53 PM

c24m48,

Just to set the record straight, GEDCOM does support proven, disproven, challenged in the child to family link. It can also contain a NOTE to explain the findings. No mainstream software that know of supports this.

#10 zhangrau

zhangrau

    Advanced Member

  • Members
  • PipPipPip
  • 1473 posts

Posted 09 January 2017 - 12:51 AM

From the RM Help for "Citation Manager"
RM_citation_manager_memorize_paste.png

 

Additional thoughts on documenting the Parent-Child relationship

RM_edit_person_parent_relationship.png



#11 Jerry Bryan

Jerry Bryan

    Advanced Member

  • Members
  • PipPipPip
  • 3405 posts

Posted 09 January 2017 - 09:42 AM

With his wonderful picture, zhangrau did a vastly better job than I did of explaining the place that RM makes available for entering parentage information. But I would still suggest that as a practical matter, entering the data in the location provided by RM is essentially worthless.

 

I've done the following experiment in the past, but I repeated it just now to be sure I wasn't forgetting something obvious. I created a new and empty database and added three people to it: Samuel Doe born in 1870, his wife Jane Smith born in 1870, and their son John Doe born in 1900. Initially, I did not enter a marriage fact for Samuel and Jane. I entered a parentage source for John on his Edit Person screen, adding the source as a family source for John's parents Samuel and Jane. Indicative of the problem with RM's handling of the situation is that the exact same sourcing information can be entered on the Edit Person screen for either Samuel or Jane in their Spouse line on the Edit Person screen, and the same sourcing information can be entered even before you enter John into the database as their son. So you can't tell if the source is really a "spouse source" for Samuel and Jane or if it's a "parentage source" for John. By "spouse source", I mean some sort of evidence that Samuel and Jane were spouses, and they could have been spouses without having children. Indeed, if you enter John into the database after the having entered a spouse source for Samuel and Jane, then John will inherit the same sourcing information that you might have intended as a "spouse source" for Samuel and Jane and John will default to a birth son, although you can change the nature of the relationship.

 

In any case, if you now run an NEHGS report for the family starting with Samuel Doe, the report will not include any sourcing information at all. As I will describe in a second, I think that's a bug in RM's code rather than a defect in the data model. But more importantly, no sourcing information will appear  for John Doe's parentage even when you get past RM's bug, and I think that's a defect in the data model rather than a bug in RM's code.

 

Next I added a marriage fact for Samuel Doe and Jane Smith without a source. The parentage source for John Doe now appears in the report. This effect is what I think is a bug in RM's code. The parentage source should appear whether or not Samuel and Jane have a marriage fact. But the parentage source for  John Doe appears as a superscript for Samuel Doe and Jane Smith, viz. "Samuel Doe and Jane Smith1 were married in 1895.". The parentage source for John Doe is not associated with John Doe at all. This is what I consider to be a defect in the data model.

 

Finally, if you add a marriage source, for Samuel and Jane, their sentence is "Samuel Smith and Jane Doe1 were married in 18952", where 1 is the superscript for John Doe's parentage source and 2 is the superscript for Samuel and Jane's marriage. Once again, the sourcing for John's parentage is not associated with John at all.

 

Therefore, I have a hard time seeing how you can store source information for parentage in a way that makes sense for reports without having some sort of parentage fact to which the source information can be attached.

 

Jerry



#12 zhangrau

zhangrau

    Advanced Member

  • Members
  • PipPipPip
  • 1473 posts

Posted 09 January 2017 - 10:19 AM

Curious. I did not do any test, as Jerry did. And as I described above, I've long been accustomed to documenting the parent-child relationship implicitly, by overlapping source citations, rather than explicitly as Mr. Gorman seeks to do. I can agree that a more direct and specific way to document the parent-child relationship would be a good idea.



#13 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6146 posts

Posted 09 January 2017 - 10:40 AM

RM uses custom GEDCOM tags to transport the child-parent relationship defined in the Parents line of the Edit Person screen.

0 @F14@ FAM
1 HUSB @I12@
1 CHIL @I13@
1 CHIL @I14@
2 _FREL adopted
2 _MREL adopted
2 _FPROOF proven
2 _MPROOF disputed

CHIL @I13@ has no specified parental relation. CHIL @I14@ has both father and mother specidied as "adoptive". Both _FREL and _MREL are exported even if RM Extra details is unchecked whereas the two PROOF tags are exported only if it is checked.


Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#14 Jerry Bryan

Jerry Bryan

    Advanced Member

  • Members
  • PipPipPip
  • 3405 posts

Posted 09 January 2017 - 03:21 PM

In the same GEDCOM, there should also be standard GEDCOM tags of 1 FAMS @F14@ for the parents and 1 FAMC @F14@ for the children. This is where I point to "the whole world gets it wrong" when it comes to documenting evidence of parentage because the standard FAMS and FAMC tags are what the whole world is using and the FAMS/FAMC model really does not seem to  provide a place to attach evidence of parentage. The RM extensions would at least provide a potential of a place to attach the evidence. But once again, RM doesn't seem to have a good place to display the evidence in a narrative report, short of adding a custom fact such as ChildParent or Parents or Parentage or something like that.

 

I'm sure I've mentioned this before, but I ran into this problem even before I became so heavily involved with genealogy software. I was looking at an application to join First Families of Tennessee. To qualify for membership, you have to be descended from someone who was living in Tennessee before Tennessee became a state in 1796 and  you have to prove every generation. I have since seen applications for membership in other organizations, such as the D.A.R., and the applications all work substantially the same. Well, I had the data but I couldn't figure out how to fill out the application. I'm sure I was and still am overthinking it, but here follows more details of my problem. The form was laid out something like the following.

 

First Tennessee Ancestor:  ____________________    Evidence: _________________

second generation:           _____________________   Evidence: _________________

third generation:               _____________________   Evidence: _________________

 

and you continue in this fashion until you get down to yourself. Suppose for example that I have the will of Thomas Bryan that mentions his beloved son Peter H. Bryan. Suppose I have the death certificate of Will Bryan that says that he was the son of Peter H. Bryan. Which evidence goes on which line? I don't know, and I'm a math/computer guy who both uses and teaches such things as data structures, including trees.

 

I think the form is totally wrong. I think it should be set up as follows.

 

First Tennessee Ancestor:  ____________________    

                                                                                         Evidence: _________________

second generation:           _____________________  

                                                                                         Evidence: _________________

third generation:               _____________________  

                                                                                    

 
With this model of the form, I know how to fill it out. If I have the first generation's will mentioning the second generation as a son, the will is evidence on the line between the first generation and the second generation. And if I have the second generation's death certificate mentioning the first generation as father, the death certificate is evidence on the same line between the first generation and the second generation. In other words, the evidence should be attached to the linkage between the generations, not to either of the principals.
 
With something like a ChildParent fact, the evidence is getting attached to one of the principals rather than to the linkage because there is no "there" there when it comes to the linkage in RM. But at least it's clear which principal the evidence should be attached to. And the evidence could still be attached to both principals. It's just that the will might be attached to the Will fact for the father and to the ChildParent fact for the son.
 
Jerry


#15 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6146 posts

Posted 09 January 2017 - 09:26 PM

Yes, each child has under its 0 @I##@ INDI block a 1 FAMC @F14@ tag pointing to the family block.

 

The GEDCOM 5.5.1 draft says that a FAMC pointer under an Adoption event "indicates a relationship to family by adoption. Biological parents can be shown by a FAMC pointer subordinate to the birth event(optional)." This seems to be an alternative to posting a PEDI tag subordinate to a FAMC tag that is subordinate to INDI which would be required in the absence of the Adopted or Birth event. The PEDI tag can have one of four values [ adopted | birth | foster | sealing ]. A further tag STAT, subordinate to the FAMC tag, is defined as CHILD_LINKAGE_STATUS:= [challenged | disproven | proven]. No explicit provision is made for tagging a source to either FAMC or STAT.

 

You would think that RM could|should support either or both of these relationship models in GEDCOM, given that it declares its GEDCOM files to be 5.5.1. Why did they resort to custom GEDCOM? Perhaps, it is a holdover from earlier GEDCOM specs, if such did not specify the PEDI tag or the use of FAMC subordinate to the child's parental event. More likely, it is because GEDCOM spec seems to make the relationship to both parents identical, i.e., I do not see an obvious  way that you could define one parent as the birth parent  and the other an adoptive parent (maybe  an adoptive parent could be another family of one parent separate from that parent's spousal family). And the RM solution supports more than 4 possible relationships, adding [ step | guardian | related | unknown ]. Lots of room for interpretation. 


Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#16 KFN

KFN

    Advanced Member

  • Members
  • PipPipPip
  • 206 posts

Posted 09 January 2017 - 10:18 PM

All,

 

Because it was asked here is the GEDCOM way to indicate adoption, by whom and the strength of the proof to the family relationship of the child.

 

Adoption:   ... = You know the data

 

0 @IPersonOfInterest@ INDI   <= NOTE this is valid does not have to be a number.

1 ADOP

2 DATE ...

2 PLAC ...

2 FAMC @FMyFamily@

3 ADOP HUSB | WIFE | BOTH <= pick um

1 FAMC @FMyFamily@

2 PEDI adopted

...

 

Child to Family Proof

 

0 @IPersonOfInterest@ INDI

...

1 FAMC @FMyFamily@

2 STAT challenged | disproven | proven <=you pick um

2 NOTE @NSharedNote@  <=A shared note can have a source_citation.

...

 

 

EDIT: TomH you are correct the STAT tag is new to v5.5.1, PEDI was present however in 5.4 and maybe 5.3 I don't recall.



#17 Don Newcomb

Don Newcomb

    Advanced Member

  • Members
  • PipPipPip
  • 1033 posts

Posted 12 January 2017 - 06:10 PM

If it helps, I've asked for the same feature in the past, on several occasions.



#18 OlivePink

OlivePink

    New Member

  • Members
  • Pip
  • 2 posts

Posted 10 April 2017 - 10:55 PM

One part of my job is to create genealogies that are used for legal proceedings. Understandably everything must be cited. Not being able to cite the source for the parent child relationship is crazy and extremely frustrating! I have also written to RM hoping they can add the feature. I miss TMG....



#19 KFN

KFN

    Advanced Member

  • Members
  • PipPipPip
  • 206 posts

Posted 11 April 2017 - 07:16 AM

Olive,

GEDCOM has a way to support a person to person (INDI<->INDI) relationship, I'm not sure if RM provides this structure, maybe an RM expert can help here.

I use this relationship all the time to provide additional documentation and linkage between multiple non-child/parent relationships, for example godparent/godchild, business partnerships, in one case slave/owner, neighbor/next farm, and church pastor, to name a few I can remember.

It acts similar to a fact without date and place but does have a Source_Citation and NOTE structure.

ASSOCIATION_STRUCTURE
1 ASSO <XREF>
2 RELA <relationship indicator>
2 SOUR
2 NOTE

The relationship indicator is the key. It says the associated person is "relationship indicator" to the person the tag is associated with. You would need an ASSO tag in each individual to cement the relationship. Normally you would not need it for a child/parent relationship since most people would hang the documention of birth off the birth tag which in general is good enough to show parentage, then use the FAMC.STAT tag to indicate that the relationship is proven, but the ASSO tag could be an additional connection if needed.

Again, not sure if RM supports this tag, but it could be supported and used in your case.

#20 OlivePink

OlivePink

    New Member

  • Members
  • Pip
  • 2 posts

Posted 11 April 2017 - 06:06 PM

Thanks very much KFN,      

 

Your detailed description is a bit out of my skill set!  I agree that there is so much that can be captured with non-parent/child relationships. It is very applicable for my work.

 

But to solve my current problem I think I am going to have to add a mother/child source fact AND a father/child source fact.

 

The clumsy sentence template I came up with is [person]'s mother's source comes from [Desc]. In the description field I put in an abbreviated title for the source then tag the full citation to the description field. It hopefully reads "Joe's mother's source comes from George Field notes".(superscript)

 

I figure I can change the template sentence when I figure something a little more eloquent.