Monday, September 2, 2013

The Problem with FamilySearch Family Tree - For Example: Thomas Dudley (1576-1653)

I've been adding, or correcting and standardizing, several of my ancestral families each week, in the FamilySearch Family Tree using RootsMagic 6 to match my persons with the FSFT person(s), adding or standardizing vital and other information, etc.

After my bout last week with Thomas Dudley and the link to Charlemagne, I decided to look at him in the FamilySearch Family Tree (FSFT).

This exercise is intended to show the problem with many of the early U.S. colonial ancestors in the FSFT.  The problem is too much erroneous data, and too many uncombined persons who are the same historical person, and not nearly enough sources to support any of the information in the FSFT.

1)  When I used RootsMagic 6 to try to Match the Thomas Dudley (1576-1653) in my database with Thomas Dudley in the FSFT, here is the list of possible matches:

As you can see, there are 15 different potential matches on the list.  All have the same name, all have the same birth year, all but one have the same birth county, 7 have the same death date, three have a death place, three have a father, only one has am other's name, and all 15 list a spouse's name (but it is the same name!).  All 15 are separate entities in the FamilySearch Family Tree.

Since these 15 persons are all the same historical person, logically I should match to all of them.  I'm leery of doing that...but it is what should be done by somebody at some time.  I'm surprised that nobody else has done it all during the 5 years of access by LDS church members and the encouragement to combine historical persons into one profile.  Obviously, that was not done in the case of Thomas Dudley (hmmm, maybe it was done and the combining ended up in 15 profiles?).

Why am I leery of combining all 15 Thomas Dudley's using RootsMagic?  Because I know that cleaning up an FSFT profile using RootsMagic is very time consuming and not everything can be done from within RootsMagic.  For instance, you cannot delete persons, or delete spousal or parent-child relationships from within RootsMagic.  Those tasks need to be done in Family Tree, and they can be done faster.  I've found it's better to do that before I match a Family Tree person to my RootsMagic person.

I picked the one that listed the spouse's name to match my RootsMagic person to, since it also had everything else correct, except the birth place:

All of the information on that FamilySearch Family Tree person is similar, although that profile does not have Thomas Dudley's second wife, Katherine Deighton, and the three children they had together.  Perhaps other Thomas Dudley profiles have them.

2)  I decided to go look at some of the other profiles for Thomas Dudley in the FamilySearch Family Tree just to see what the situation was.  I did a "Find" for Thomas Dudley and saw a long list (the same ones that are on the Match list above), and I picked one of them (it was the second one on my Match list above) - the "Vital Information" list::

The "Other Information" in the profile:

That's just the top of the category - there are 37 different variations on Thomas Dudley's name on the list, including one "Samuel Dudley."  There are also 7 Titles of nobility (including one for 9GG), and a baptism, confirmation and first communion event in the 1890s (which may be LDS ordinances).

Further down, there are the two family lists - one column for the "Spouses and Children," a second for "Parents and Siblings:"

Again, this is only the top of this section.  There is only one set of children (but only three children!) with Dudley's first wife, Dorothy Yorke (there were five children by her, I think).  However, there are 31 entries with Dudley's second wife, Katherine Deighton (many different spellings, and all 31 have a different FSFT ID profile number for Katherine).  None of them list any children for Dudley with Katherine, but several sources say they had three children together.

It gets worse:  Over on the right side of the screen above, Thomas Dudley is listed as a child in 12 different families headed by Roger Dudley (all the same FSFT ID), with Susaana Thorne identified as his mother in 11 of them (there are 11 different FSFT IDs for Susanna).

This Profile for Thomas Dudley had one "Sources entry - a  citation for everything on the page (none were tagged to an Event):

 The source given above is from a website page for Thomas Dudley at  That is not an original source, but it appears to be a fairly reliable derivative source.

I was curious about the "Discussions" section:

Two years ago, two persons added a comment about the daughter Sarah and the second wife's name.

3)  That's the problem!  There's nothing that says the other 13 profiles for Thomas Dudley aren't attached to more duplicate or wrong spouses, or duplicate or wrong parents and siblings, or duplicate or wrong children.

Do you see all of the extraneous information, the multiple entries for persons (in this case, Thomas Dudley, and spouses for both Thomas Dudley and his father, Roger Dudley)?

How can that be worked to make Thomas Dudley (1576-1653) into one person Profile, with one person Profile for his two spouses, and one person Profile for his 8 children?  And then we have to deal with the children, and their spouses, and the parents of Thomas Dudley and his spouses, ad infinitum it seems.

It will be a really BIG challenge to do this one Person Profile and all of his family connections and get them right.  Then there are millions of other person Profiles to add, correct or delete.

Do you understand why I'm leery about doing Thomas Dudley's FSFT profile using RootsMagic? I'm quite sure that the challenge ism ore than I can handle.

FamilySearch thinks that if everybody works together and does their part that the task will get done and there will be this one really big MOAHPFT (that's Mother Of All Historical Persons Family Tree).  That's a great thought in principle - but making it work is the challenge.

What are your thoughts?  How can this be made to work for historical persons like Thomas Dudley?  Are you willing to contribute?

The URL for this post is:

Copyright (c) 2013, Randall J. Seaver


goneresearching said...

After the RootsTech 2012 presentation on FamilyTree I went into FamilySearch thinking I'd correct the incorrect and source it properly but what I found made me want to cry.

The Dunham entries I wanted to work on I could not because they were IOUS (Individuals of Unusual Size) -- all of them. Guys with different names were smashed together as one person (when they were clearly not the same guy) and at the same time married to the same woman multiple times (when they should have been merged). Add to that the same kids multiple times and then multiple parents.

And different people are editing them each thinking the guy is someone else. It is a mess.

Yes, it is a nice thought/plan for "one tree" but there is no good way to un-knot the spaghetti even with scissors. I'm in the process of submitting unmerge information to FamilySearch for them to do in newFamilySearch but it takes a couple days just color-coding person identifiers for just one Dunham guy and then writing it out who is really who. And I still don't know if they'll do the correction.

All I wanted to do was source some information on my guys so the correct info was out there, but it is taking longer to get it cleaned up than it would to do the sourcing.

I thought I was alone in this problem. I sounds like you got a case of it too.

Donna said...

I've been working on my tree as well and it is a nightmare to try to get the duplicates merged. Additionally, If the first name wasn't fully gender specific, many people input them twice - one of each gender. You can't merge people of opposite genders and you can't delete the profile either. So you have to contact Family Search with the information and let them take care of it. There are also many people of the same or similar name with no parents, no spouse, little to no information on when or where they might have been born, so those are going to just remain cluttering up space with nobody ever able to guess who they have been intended to be. It is my opinion FamilySearch should clean up their own mess.

SearchShack said...

I've been working to clean up the duplicate SHACKFORDs in FSFT and am finding the same issues that you raise -- moreso on the original settlers - 9 duplicates, then duplicates of their differing parents, etc. Also see the issue with anyone with lots of children as the parents were entered separately for each birth. I've made a small dent in cleaning up the tangles. It is very time consuming and frustrating but in my case it actually helps me think back through the genealogy relationships as I work through the corrections. Also but I'm now seeing a few others posting on these same trees via the WATCH feature and am getting replies and some new collaborations when I write to those others interested in the same family lines.

Lauri said...

As I was connecting people in a family tree I also had one of these. Over 99 instances of the same person that needed to be merged. I got through some and then too many and couldn't merge anymore. The children have similar problems and it is a disaster. Decided at that point to just wait on that branch of the tree.

John said...

It is time for the Powers That Be at FSFT to issue some guidelines/rules. For example: 1. Governor is preferred over Gov, Gov., (Gov), (Gov.). In other words avoid abbreviations wherever possible.
2. Titles, like Governor, should be entered as a prefix, not as a suffix.
3. Christian names and surnames should be entered in standard upper and lower case format, e.g. Thomas Dudley, not Thomas DUDLEY
I am not saying the above rules are right or wrong. I just believe we need some rules.
John in Victoria, BC

Geolover said...

FamilySearch Family Tree is chock-full of such messes. While all of the duplicates (with or without duplicate and/or incorrect spouses, children, parents) may have been combined in the older new.FamilySearch tree, some crew at FS has been separating out versions of individuals, and this may be what you are seeing. in new.FamilySearch one could not truly merge persons or undo what someone else had done, for the most part.

Some of my ancestors have more than 100 duplicates in FS-FT. How could one even select the least wrong version to start with, in less than a couple of hours (if the program were running a lot faster than usual)?

Stuff that was extracted from actual records and put in IGI years ago appears now in FS-FT stripped of any identifying information as to source.

In my forays I've found that as "sources" and reasons for presenting some data, people are giving internet trees, message board posts and "from GEDCOM."

And some entity identified only as "FamilySearch" keeps adding wrong dates, spouse(s) and as many as several sets of wrong parents to individuals I look at occasionally because they are subjects of Widely Held Mistaken Beliefs in myriad publications and trees.

In addition to those problems, given the hodgepodge of database sources, place-name problems, cultural/social biases and rather odd programming bugs, it's hard to see this as a really viable vehicle for genealogical accuracy.

Kellie Reeve said...

This is one of the reasons that I prefer WikiTree over the FamilySearch tree. These issues are addressed daily by some marvelous folks over there, and duplications are easily handled. Corrections and updates are easy to make, too. I wish FamilySearch luck with their endeavor, but I think they were at a disadvantage from the beginning.

Doris Wheeler said...

Amen, Kellie Reeve. Perhaps FamilySearch is too democratic. They give equal weight to every entry, and no one profile can be "tagged" as historical and therefore protected. The same old errors keep being reintroduced, no matter how often you correct them, with sources and even notes describing why they are erroneous, or at least suspect. I've stopped trying out of frustration. Wikitree does have a better system.

T said...

familysearch has too many problems. I have decided to stay only on My reason is simple. Anyone who is researching their family will probably check the family trees. If they find someone promising they can add to their own tree without disrupting the original tree. Collaboration is possible by contacting the tree owner. There will still be wrong people added but it won't be on a master tree so that everyone who researches finds the same errors. Sorting people from one tree is much easier than sorting from 100. I think that eventually most people will subscribe if only for a short time to see what they might find there. I am not going to pay next time around but my tree will still be there and I can still work on it. Maybe 6 months down the road I will subscribe again but for now I have exhausted everything ancestry has for me.

A world family tree may be possible but not in my lifetime. My own relatives don't agree on ancestors. How can there be an accurate world family tree? Caleb Johnson discovered a major error in one of the Mayflower families. How many years worth of genealogy has that affected? And does everyone know? Not likely so the misinformation goes on. Work done by others for my own family is proven wrong but the old researchers won't change what they have because that's what grandma told them. Well, grandma was wrong. So comparing one of their trees to mine, there are conflicts. Who decides which one is right? Which tree will future researchers use? Until every baby is micro chipped at birth and followed their whole life, it will continue to be a mess. And chipping will be mandatory any day now!

Kathy Matlock Wigley said...

Mercy Dudley is the Mother of my 8th Great Grandmother, Martha Woodbridge who married Capt. Samuel Ruggles Jr. My Grandmother is Nellie May Ruggles. Here is a link to my family geneaology blog I am working on for those interested in finding connections.