US1910Project
The BYU US1910Project is creating duplicate records in Family Search. There are not enough data points between those US1910Project records being created in Family Search and existing records! PLEASE STOP this project!
This example shows that the father was originally created on January 15, 2017, then he, wife, 2nd daughter and son were created June 24, 2021 AFTER the US1910Project duplicate family. Even though the family was mostly created AFTER the US1910Project, the database didn't find a duplicate for the wife, husband or 2nd daughter on June 24, 2021.
US1910Project duplicate family created April 15, 2021:
Willy Wilson GCRT-FBV father
Lula Wilson GCRT-WKP mother
Grace Wilson GCRT-L47 daughter
Mable Wilson GCRT-GZL daughter
Existing family in family search created January 15, 2017:
William H. Wilson 27 July 1883 – 24 March 1962 • LR8Q-FP5 created January 15, 2017 (merged with GZLT-N2D created June 24, 2021)
Lula Ryan 10 June 1890 – 11 February 1950 • GZLT-F94 created June 24, 2021
Mable Clara Wilson 21 September 1908 – 5 April 1980 • GZLT-N3L created June 24, 2021
Elmer James Wilson Sr 30 January 1911 – 4 November 1968 • GZLT-4LJ created June 24, 2021
Grace L. Wilson LBG3-V37 originally created May 6, 2018 (with no parents) as a wife to George J. Johnson LBG3-V37
Best Answers
-
Great example of why this Project should be abandoned.
2 -
okay having reread it, I think I understand this to mean ... for any given 1910 census record, if that record does not hint at a person to attach it to, they create a wholly new person without doing any research to see if that person already exists?
I would say that if they're going to continue using such a shoddy procedure, perhaps a different US 1910 Census Project should be started essentially to get there first... except, as we've seen, they would instead move on to census records from a different year.
Such a waste of everyone's time. What are the students even learning from this exercise?
2
Answers
-
Your issue is being sent to an internal department for review. If more information is needed, you will be contacted by private message.
0 -
Agree, this project is a nightmare and should be stopped. Duplicates are being created for people whose ordinances are already done or already reserved. When you go to your reservation list you are NOT going to know that a duplicate was created AFTER you reserved the name and someone will eventually be completing all of the ordinances again. Examples:
Charles Lee Beadle GM5L-RSG (all ordinances complete) and Charles Beedle GZ3J-2SZ
Hugh Anselm Beadle GQTN-TZK (reserved) and Hugh Beedle GZ3J-ZD5
Why are we talking about reducing duplication of ordinances and then having a project like this?
BYU figure out something better than this.
8 -
Regardless of the problems relating to ordinances, whatever is FamilySearch allowing here? This project appears to run completely contrary to the declared aim of having just one ID for "every" person who has lived on Earth.
I don't know the background of the project (is it really a BYU one?) but believe it will cause the same sort of damage as all those GEDCOM imports, if it is working as is being suggested (new ID for a huge number of individuals indexed from the 1910 US census).
If this project really is working as described, the whole thing is against FamilySearch's "policy", is completely outrageous and must be escalated to the highest level, so the potential production of a huge amount of duplicate IDs can be halted immediately.
8 -
Looking more closely at the other comments, it appears duplicates should be being picked up, but many are slipping through regardless. Perhaps I am slightly over-reacting in my comments above, but there surely still needs to be at least a suspension of the project, until the flaws (that are wrongly allowing duplicates to be created) can be addressed.
4 -
This is a project of the BYU Record Linking Lab. Please email them directly and share your concerns with them about duplication. They welcome the feedback. Here is the email address: rll@byu.edu
4 -
Comments still need to be made here. It's not the BYURLL who needs feedback (other than STOP!), it's familysearch, Familysearch needs to be aware how much extra work this is causing in addition to the confusion. I've grown weary of unwinding the duplicates created by the US1910Project and the seemingly endless merges they force. I'd prefer familysearch delete every record with only the US1910Project as the contributor and blocking further contributions.
One issue is the 1910 census records the number of marriages. US1910 simply defines all children in the household as the child of the current wife.
A second issue is the creation of records under the married name of the wife.
That they'd do these two things is a HUGE flag that they don't have a fundamental understanding of what they are doing in a genealogical sense OR Family Tree sense.
Thank you to the OP and to RJGKennedy1 for referring it to an internal department at familysearch. Paul W: you *weren't* overreacting.
8 -
Today after merging THREE duplicates for Willitt Hammond 29 April 1877 – Deceased • 9X74-VFT, of course there is a FOURTH one to merge from the US1910Project.
Willitt Hammond 29 April 1877 – Deceased • 9X74-VFT (created before April 26, 2012, all ordinances done except sealing to spouse)
Leona M Johnson March 1884 – 5 September 1943 • LRGQ-QV4 his wife (3 duplicates merged, first created 13 March 2017 with her parents and siblings, ordinances reserved)
Samuel C. Hammond 6 September 1907 – 17 January 1987 • LBR3-JTZ (son, added May 18, 2018 ordinances reserved)
Still working on the research for this family....
Then, US1910Project created:
Willett Hamlin 1877 – Deceased • GCZ3-TYX (husband/head, created March 26, 2021, duplicate #4 because the last name is spelled wrong on the 1910 census record)
Leona Hamlin 1884 – Deceased • GCZQ-WLC (wife, created March 26, 2021, duplicate #4 because the married name is spelled wrong on the 1910 census record)
Samuel Hamlin 1908 – Deceased • GCZQ-WLX (son, created March 26, 2021, only duplicate #1because the last name is spelled wrong on the census record)
Leopal May Hammond 28 April 1910 – 18 June 1988 • 9J7Z-K88 (daughter, created April 29, 2012 only with husband, one duplicate because the last name is spelled wrong on the census record, only needs sealing to newly attached parents)
Can we just get a definitive answer to find out if this project is no longer adding records to Family Search. so that we only have to deal with the residual duplicates?
PLEASE?
No more "referrals", just an answer, posted for everyone to know?
2 -
Another example:
The US1910Project created ID's for all nine members of one household, then cloned those nine members and assigned different ID's to the clones. All on the same day.
Here's the head of that household, John Luchansky, GZMN-8F2 and GZMN-VJL See: https://www.familysearch.org/ark:/61903/1:1:MG4F-5CM
This might sound flippant, but it illustrates the need to address the US1910Project now, I neither wish to see a US1900Project nor a US1880Project.
3 -
Yes, STOP this travesty, US1910Project, for all the reasons outlined above.
2 -
Since that is a project that is not sponsored by FamilySearch, you might want to check with BYU Record Linking Lab projects, to see if it is continuing now. Here is the URL: https://rll.byu.edu/projects
0 -
A point made earlier is an important one. Basically, who controls the running of FamilySearch - its own management team or someone at BYU? If the former (as surely should be the case), then the team at BYU responsible for this project should have been told to suspend it immediately.
5 -
I also agree.
I have done dozens of merges for duplicate people based all over the country that were made by this US1910project group.
Not only have they created entire families based on 1910 census records, once those families are made with different spellings and places and dates of birth that the records don't show up as hints for the correct family, nor hints for Duplicate people.
Every time I find a family made by this stupid project I end up spending time fixing something that never should have happened.
3 -
I agree that these census projects should be stopped immediately. My sister consistently finds whole families duplicated as she works on our lines in California. I work mostly on English lines and I'm told they are doing the same thing with the 1911 English Census, which makes me heartsick.
It's never a good idea to add people to Family Tree based on a single source, and census records are especially problematic:
- Married women are listed by their married surnames, but women are supposed to be added to Family Tree based on their maiden surnames.
- Ages may be reported incorrectly (in addition to the variance caused by the person's birthdate relative to the census date). For example, I've noticed that women may not age 10 years between censuses :)
- Relationships may be misstated for blended families (e.g., a child is listed as a son when he's really a step-son). In fact, step-children may wrongly appear with their step-father's surname instead of their actual name, contributing further to confusion and duplication.
- US censuses only give states as birthplaces, which makes finding duplicates problematic.
- A census is only a snapshot of everyone in a certain residence on a certain day. Rarely does a census list all family members.
I also agree that the responsibility falls on FamilySearch to stop these projects. Several of us have tried to express our concerns to the RLL but they seem determined to continue despite frequent duplicates, low quality information, and other problems. They have erroneously concluded that it's their job to "grow the tree," as they say. But that isn't their responsibility. That responsibility rests mainly with family, not computers or strangers.
In addition, these census projects teach false principles about accuracy, using historical records, and working in a shared tree--particularly to the inexperienced college students who participate and don't yet know any better.
If the RLL wants to create trees based on census records, they should do so in a separate database, not in a shared tree where low quality records impact all users. Novice users may not recognize the errors, so they build on them and (as has been pointed out) are likely to do duplicate temple work. Experienced users are more likely to find the errors, but then they spend hours cleaning up damage instead of making progress on their own lines.
This post is getting long, so I'll stop for now :) But if anyone is interested, I can share other reasons why these census projects are damaging and should be discontinued.
5 -
I heartily agree with all the concerns posted so far. I constantly run in to duplicates (a family of 8 the other day). When I see US1910Project as the user, I cringe knowing that I have a lot of merging that I will have to do before I can continue with my real work. FamilySearch - PLEASE revoke the API keys for the US1910Project (and any similar projects) and then they can work like the rest of us - manually researching names instead of just dumping in duplicates.
I understand that the perceived reason for dumping in all these names is to "grow the tree" and help new folks who are just getting started; however, while dealing with dupes is an annoyance for those who are experienced, it can be intimidating for those new folks who have to deal with the dupes they run into.
As mentioned above as well, allowing GEDCOM dumps at this point is just as problematic. Everyone should be required to manually enter names and have them vetted on entry.
The Lord wants a correct tree, not a huge tree that is bulging with duplicates resulting in temple attendance time wasted on those whose ordinances have already been done under a correct record, when there are so many waiting for their work to be done for the first time.
3 -
I agree with all of the comments that are critical of this project. It is creating more problems than it is helping. Most of the comments deal with people. Tonight I found a person where the entry for "Immigration" was added and the date of 1854 was inserted. The problem is that I have no idea if this is correct as I have not had the opportunity to research this matter. However, my notes have these possibilities:
IMMIGRATION: To U.S.:
per 1900 census = 1850
per 1910 census = 1854
per obituary = 1846 (1913 - 67)
Why should the date of 1854 be put in the immigration category when there is no evidence of this? This is crazy!
I am 77 years old and have been working on genealogy since I was 18. This project is unacceptable and goes against all the "rules" for proper sourcing of data.
I strongly suggest it be discontinued.
3 -
Perhaps we could be provided with an update on this issue. Has this been escalated to the appropriate level, in order that BYU (which I assume is not represented on the FamilySearch management team) are / have been ordered to stop this damaging work?
1 -
I'll see what I can find out.
3 -
Yesterday, an error made by the 1910 census project cost me 90 minutes to straighten out. If this project would just do 10 minutes of research, say, by looking at the 1900 census for the same family, they might be able to avoid creating so much damage. No, this was not my family; yes, I am retired, so I had the time to take to fix the problem. But why do such shoddy work and create so many problems in the first place?
3 -
Here's the scoop folks. I emailed them beginning last fall. I just followed up because of the current interest
Sent: Tuesday, September 14, 2021 10:41 AM To: rll@byu.edu Subject: US1910Project
I don't see this currently listed as a project on your website, at least by this name.
For regular FamilyTree users this is a nightmare. Duplicates are being created for people whose ordinances are already done or already reserved. When you go to your reservation list you are NOT going to know that a duplicate was created AFTER you reserved the name and someone will eventually be completing all of the ordinances again. Examples: [etc]
The first paragraph in the response below absolutely floored and quite frankly infuriated me. Totally disingenuous, they are NOT "working really hard to reduce the duplication of temple ordinances'. That's why it took me this long to respond.
From: Joe Sent: Tue, Sep 14, 2021 at 3:19 PM Subject: RE: US1910Project
Would you have an interest in helping with our projects? I agree that we want to reduce duplication on the Family Tree, especially when those ordinances are shared with the temple system. We are working really hard to reduce the duplication of temple ordinances and would love your help if you are willing.
We are also open to your ideas on how to best allow people to find their family on the Family Tree when they come for the first time. Our projects stemmed from experiences helping African Americans who were new converts to the church. Their experience looking for their family on the Family Tree was very different than that for White individuals. This is what prompted much of our tree growth projects. We would love to test out any ideas you have for how to help people better find their family. There are lots of countries where the tree is empty and we could have a huge impact. We always love help.
Joe
From: Me Sent: Saturday, January 8, 2022 5:48 AM To: Joe Subject: Re: US1910Project
Dear Joe,
3 questions:
Is the 1910 project still continuing? There were 92+ million people in this census. Will all of them be added to FamilySearch?
What projects are available right now to help reduce duplication?
Joe To: Me Sat, Jan 8 at 11:07 AM
Yes, the 1910 Census Project is still continuing. I gave a presentation about it to the Bridge Forum last week. Here is a link that video.
We are focusing first on the families that have at least one child living at home. This is 68 million people. 87.8% of them already have a profile. We are helping with the last 12.2%.
Here is a link to a project we have set up that you can use to help reduce duplication on the Family Tree. There is column labeled duplicate. Any row with a 1 in that column is a person with a possible duplicate. It would be great to have your help with that.
Joe
So as you can see, they know it's causing duplicates. Hopefully the same people that are working on this project will be the ones that merge the 92+ million duplicates they have created minus the ones that all of us have had to tangle with. We will just have to wonder if the duplicate merges will be correct. I am not holding my breath. It is absurd that FamilySearch is allowing this and whatever like it may be coming next. I didn't get anywhere trying to contact FS about this.
Regards to all
4 -
I wonder if you might offer any further suggestion as to how prevent this ongoing BYU project causing so much damage? You can see the response received when BYU was approached directly.
Does anybody in the FamilySearch organisation have control when it comes to the negative effects a BYU project is making on its Family Tree product? If they do, surely they should be made aware of this situation and take action?
If nobody in the FamilySearch management team is worried about this, I certainly won't be spending my valuable time merging hundreds of duplicates (usually only distantly related to me), while FamilySearch stands by and allows this BYU project to continue in its current form, with its known consequences.
2 -
What comes next will likely be the 1920, 1930, and 1940 censuses.
Norman Hellmers
1 -
I can't even begin to tell you how many duplicate people I have had to merge due to the us1910project.....hundreds I'm sure. I work on here almost daily. FS is ultimately responsible for what is going into the system from BYU. FS shouldn't be trying to pass the buck over to them. That would be like me telling someone that my employee's mistakes aren't my concern. This so-called "project" NEVER should have been allowed in the first place.
3 -
An excellent analogy! As you can see, there has been no further response from any moderator or employee to my requests for an update on whether FamilySearch intends to deal with BYU on this issue.
This is very down-heartening to FT users who work their hardest to cut down on duplicates, while the BYU project has been allowed to do so much damage in the opposite respect.
1 -
Beyond frustrating on so many levels. This project has got to stop. I've yet to see something correct. Can they be required to go through training before allowed on Family Tree? Someone has to teach this group...
3 -
I actually found a couple of cases the other day where the US1910Project actually created their OWN duplicates. I found 2 families where all the records for both families were created by the US1910Project on the exact same day almost a year ago and they hadn't caught and cleaned up the duplicates yet. I didn't spot check beyond the 2, but suspect there are more.
I left a critical review on their facebook page, but their pat answer (as in the response already shared in this thread) was "Why don't you come help?" I responded that I DO help every time I run into a mess they created, I help by cleaning it up. I really wish someone with power would be willing to cut off their access to automatically upload data...
3 -
When this current version of FamilySearch was being introduced through a 'WhitePaper' titled 'The Case for Our Tree' (or something like that).
It introduced three GOALS: Sources, Collaboration & avoiding Duplications.
Regrettably, I feel the completion of those 3 goals has not been accomplished & FSTree leaders might consider a review of that excellent document & look for ways those worthy goals might be better acknowledged & ACCOMPLISHED.
Link to the document:
https://broadcast.lds.org › eLearning › fhd › Community › en › FamilySearch › FamilyTree › pdf › The_Case_for_Our_Tree_FamilySearch.pdf
1 -
It appears the subject of your link can only be accessed by members of the Church.
Nevertheless, I agree with your comments and find it very sad that, as individuals, we are strongly encouraged not to create duplicates, yet the FamilySearch organisation (and BYU) are knowingly allowing this project to continue, with such damaging results.
0 -
Try this link:
I do no think that one must be a Member of the Church to access
Brett
2 -
😀
0
