"We detected a possible relationship issue in the Family Tree" message - is this new?
https://www.familysearch.org/search/linker?ark=%2Fark%3A%2F61903%2F1%3A1%3AQPCJ-46GY&id=L6PK-4JM&hinting=%2Ftree%2Fperson%2Fdetails%2FL6PK-4JM illustrates what I am seeing every time I try to attach a source to a Robert Wrightson, who is shown as the father of Frances Jane Wrightson on the Tree. I have just worked out the problem here: she is also shown as the daughter of a THOMAS Wrightson.
I can carry on attaching the source(s) after clicking on the "proceed to attach" link, but am not happy that a bot appears to be dictating my priorities - i.e., I need to give attention to this "relationship issue" before proceeding with any other work. (The sources I am attaching do not relate to Frances Jane at all).
Surely, such matters should be highlighted against the profile(s) in question, not when trying to attach a source that has no direct bearing on my attaching sources to one of Frances’ "suggested fathers"?
In summary, it appears a user has attached Frances to two sets of parents, without reason. I will remove the incorrect relationship (to Thomas Wrightson) later, but am retaining my post in a hope of responses about the “warning” message appearing at this stage, and whether this is a new feature, etc.
UPDATE - Just noticed that Thomas appears to be Frances Jane's grandfather and that she has been accidentally added as both his daughter and (correctly) as that of his son, Robert.
Answers
-
I haven't seen that one before, @Paul W.
Based on the wording, that could be an AI "analysis." I hope AI is not presuming to guide experienced (or novice) researchers.
2 -
Thanks for the prompt response. Once again, I was probably editing my post at the same time, but I don't think my original wording has effectively changed my question - or your response!
I'm not denying it was a positive warning to raise - but that it pops up when carrying out an action not directly related to the individual mentioned, yet there is no warning against her profile (LWB4-9RR) - where it is specifically stated there are "No Data Problems" - all seems rather odd.
2 -
I think it is new. I ran into it once a few days ago and it was concerning a situation that would have resulted in a looping pedigree that I would not have noticed otherwise. Why this only appears when entering the source linker is a good question. I would wonder what is going in as the source linker screen is being generated that this kind of relationship problem presents itself only then. I also wonder why the new feature can't point out what the problematic relationship is. In the case I ran across, like Paul's, it was not a relationship that was involved in the source or that showed in the source linker. It took a bit of searching to find the problem. If I remember right, a man had been entered as both the son and the son-in-law of the same person.
I think it is a very useful new feature but agree that it is a very weird place to have it show up.
4 -
a very weird place to have it show up
A bit like some of the messages in the PQS that don't quite match up with the place or profile where we see them.
0 -
Of course, it may not be all that new. It may just be that the type of error it identifies is rather rare and it is just a coincidence that Paul and I both ran into one the same week.
The source linker does pull a lot of relationship information while creating its page which shows the focus person's parents, siblings, children, and others. That may be why only the source linker can recognized such errors.
3 -
The idea that Frances Jane Wrightson is the daughter of both her father and her grandfather surely cannot have caused an issue by itself??? Our family histories are full of children who have been "granny-reared" and who will therefore have that pair of relationships. Admitted, one should tweak the Relationship type to say that the relationship to the grandparents is adoptive (probably) but I'll bet plenty of people can't get their head round that.
I just set up exactly this "granny-reared" relationship on Beta-Site to test, and the set-up of the double sets of parents works quite happily. That doesn't mean that I will be able to attach all sorts of sources without giving the Source Linker a thumping headache and….
Bingo!!!! I just tried to attach a simple birth registration index to the granny-reared child and got the same error message!!!
Even when I set the relationships to the grandparents to be adoptive, I get the same error message.
This is simply unacceptable. Firstly, the scenario that gives rise to this error message is perfectly normal.
Secondly, the error message is unintelligible - "There are two persons in the record with the same id: Emily Cadman (KGL8-MNX) and Emily Cadman (KGL8-MNX)" - which record for goodness sake?
Then it says, "If our analysis is correct, the historical record was incorrectly transcribed…" Well, no. My set-up was nothing special - in fact, it's actually what happened in real life to the person whose profile I chose to use for my test. All the historical records have been correctly transcribed.
Even weirder (perhaps?), the error message in my case refers not to the child with 2 "mothers" (Edith Cooper) who was granny-reared (by Emily Beech) but to her mother (Emily Cadman - who is the daughter of the Emily Beech, who reared the grandchild). Though whether I should have expected the child to be the subject of the error message, I don't know.
(Sorry if there are too many Emilys there but as I say, I chose a real-life example).
4 -
I wonder if we should ask to move this thread to the New Source Linker Feedback group.
0 -
Or should I raise the same matter in the NSLF group (with same heading as here) and add the URL for this "question" as a link? No, I'll tag Ashlee for now, as she (and Scott) appear to deal with most of the topics raised there.
@Ashlee C. for your call, please!
2 -
@Paul W If I'm understanding correctly, this issue only happens when you are using the source linker. So it is a perfect discussion for the New Source Linker Feedback Group. That is where the engineers are monitoring posts about the source linker.
0 -
Yes, that certainly appears to be the case, so will leave it in your hands as to whether you can move the whole discussion across, or advise the alternative (e.g., whether you want to escalate directly from here, or if I should raise a separate discussion at the NSLF group, as suggested earlier). Thank you.
2 -
@Ashlee C. and @Paul W - For what it's worth, I'd agree with directing it to New Source Linker. I just retried my example and the New Source Linker screen flashes up immediately before the error message appears.
When I retried it, I also dug a bit further and clicked "Proceed to Attach". I had to mess around a bit with the Source Linker, opening and closing Parents and Siblings but I can now see Emily Cadman (biological mother) in two places in the Source Linker screen - with the same "Role". At the top, she is "Emily Cadman (Mother)", i.e. Edith Cooper's Mother, and at the bottom, she is "Emily Cadman (Mother)" listed as a sibling of Edith Cooper (because the siblings are the siblings from both biological and adoptive mothers).
1
