HUGE number of incorrect Record Hints on Scandinavians in the tree
Comments
-
Dale Hein said: Another one. This time I'm helping someone else. There is no mention of Pennsylvania for anyone on this page. Nor that they had a daughter named, Victoria. Nothing matches -- except the name itself, Martin Tobin.
https://www.familysearch.org/tree/per...0 -
Dale Hein said: This is discouraging. Did anybody see this quote on this blog article, "Have You Seen This? Cool Features on the Family Tree App"?
Talking about Record Hints:
"These possible matches show up as record hints that can help you find records without filling out search forms or looking through record collections."
They don't even suggest the possibility that a Record Hint might be wrong. It's no wonder they are messing up so many families in the tree.
https://www.familysearch.org/blog/en/...0 -
Dale Hein said: Brian Jensen,
Can't you at least put a warning that pops up right before any Record Hint is attached? Like in red "WARNING! Remember! Hints are just hints! They may not actually be for this person!" What could possibly be Family Search's objection to doing that if their top priority is accuracy? Too many people think FamilySearch and computers can't make mistakes. I would think a warning like that would be a lot easier than trying to perfect the algorithyms, which, by the way, will never be perfect. It's 100% impossible. Because to be perfect all historical records would have to be perfect and the data in the tree would have to be perfect. Neither of those will ever be perfect. Newbies and technicians might think it's possible, but serious family history researchers know better. The ideal solution would be not put them there at all. Those of us who know what we are doing can easily find them without any help from hints. But if you won't do that, can you please at least put a pop-up warning. Do you want Family Tree to be accurate or just cool technology?0 -
Paul said: Dale
When you write, "What could possibly be Family Search's objection to doing that if their top priority is accuracy?" I assume you are speaking with tongue in cheek!
Sadly, the action of FamilySearch management has illustrated exactly the opposite in my experience. Whether it is with the careless way they suggest record hints or their complete refusal to make corrections to known errors, working towards greater accuracy always seems to be set aside. "There have been greater priorities" has been the response (these have become rarer and rarer from managers lately) to posts that show this particular issue is not being addressed. "Greater priorities" has meant making things look prettier, with the main aim to attract greater involvement, never mind whether this will be by those with little or no experience of genealogy.
True, there needs to be basic enhancements to the progam - I have sometimes suggested some myself, which I feel would make life easier for us. However, I feel now is the time for someone at the top to acknowledge things are going rather downhill at the moment and do everything possible - from improved algorithms and making basic training more easily available, to at last making changes to those incorrectly headed indexed records that can point users in the completely wrong direction.
Two weeks ago I politely requested Brian Jensen to confirm he is still following this thread. He has responded to points raised elsewhere, so I would kindly ask him again - are you taking note of the specific examples Dale been providing, with a view to improving the current, flawed algorithm?0 -
Dale Hein said: Paul,
Totally agree with all that. And, yes, I do know that accuracy is not a priority. I started another thread several years ago asking them not to allow adding GEDCOM files to the tree anymore, and I have put lots of additional comments in it, too. My two big issues have been GEDCOM's and Record Hints. Now I have three -- Ordinances Ready. No, four -- campaign emails. Ordinances Ready could be great if, again, they would tell people CHECK THE NAME AND HIS OR HER FAMILY MEMBERS FIRST. But they don't. They just say grab and go period and sing the feature's praises. "It's so easy!"0 -
iLoveMyLife02 said: I've been getting a large number of "Hints" where (1) the source is already attached to a person, with the CORRECT parents, siblings, dates, etc., and (2) the "Hinting" parents, siblings, or dates do NOT match at all. A less careful person could really make a mess of the attached sources and relationships by detaching good matches and attaching bad matches.
I'd say 70% of the hints for my ancestors over the past few days have been bad hints.
I have an attached an example, which unfortunately does not show the correct couple that this source was attached to. Take my word for it, it was attached to a perfect match of names and dates for both members of the couple, and then I got this rotten "hint".
Janet
0 -
Dale Hein said: Janet, thank you so much for posting this. I wish everybody who finds them knew about this thread and would post them. My experience is that 70% sounds just about right.0
-
Dale Hein said: Wow! Here is a perfect example I just found in the pioneer lineage tree of a friend of mine whom I am helping. I saw a totally empty line in the mid-1800's in her fan chart and right away I thought there has to be a mistake here. This member has a pioneer tree that makes it hard to find any ancestors to do temple work for. So I went to the page where the ancestor now shows no husband in the tree and check this out. Somebody attached THREE sources for a totally different person who lived in a totally different place had totally different children and she added all those children and now she has a bunch of cards printed for totally bogus ordinances. Obviously, this person hasn't a clue what she is doing, so you know they were Record Hints! This is terrible.
https://www.familysearch.org/tree/per...
Does anybody with authority at FamilySearch International care that this is happening????? A lot!!! I don't understand why people in the trenches who are actually working in Family Tree are assumed to be dummies whose feedback is of no value.0 -
joe martel said: That URL to L2SQ-P2R has a husband and I don't see the 3 Sources from one User, or that a husband relationship was recently deleted. Did you mean a different PID?0
-
joe martel said: I'm not that familiar with this family. But from the screen shot you provided could you provide more info about why you think this isn't right? Thanks0
-
Gordon Collett said: It looks to me like the hint is already correctly attached to a sister born five years later and the hinting routine is picking up on the E middle initial to suggest the two sisters might be the same person.0
-
Robert Wren said: Joe, I think Dale is referring to the second of the three couple relationship, which appears to have different children and the third couple with a 'similar' spouse & one child.
Looking at the 12 sources for Anna L2SQ-P2R, it appears one person added 5, two others 3 each. FWIW0 -
Dale Hein said: I must have seen the three sources from one person on a child's page. But my friend was just here. L2SQ-P2R is her gr-grandmother. So I'm not familiar with her family, but she is. All you need to do is click on each of the children in the first marriage to see that there is a problem. Some had their work done in 1934 and some recently. The whole page is a mess.
It's taken me awhile, but I think I finally found the problem. Somebody attached death certificate for her on Feb. 24, 2018 which says her mother is Anna H. Swenson, and at the same time she merged her with an Anna Sophia Gustafson, so I am going to restore the correct mother before I do anything else. But I'm waiting to hear back from my friend first who is going to check with someone who knows her ancestry better than she dies if they know for sure the mother is Anna H. Svenson. But I also found another crazy one Hint in my friend's tree. I'll post it.0 -
Dale Hein said: Here is another really crazy one. There are 7 birth Record Hints on this father's page for his children.
https://www.familysearch.org/tree/per...
Not one of them is for Ostergotland where all 8 of his real children were born. They are all for Kalmar, Jonkoping or Kronoberg. Go figure. It's not like the father has an unusual Swedish name --Samuel Johansson.
He had two real wives -- Lena Jacobsdotter and Stina Larsdotter. Only one of the 8 mothers in the 8 sources matches one of his real wives. The others are one of these three:
Stina Persdotter
Stina Eliaesdotter
Lena Larsdotter
I guess they all had either the first or last name of one of his wives, but Stina and Lena are VERY common Swedish names and so are Persdotter and Larsdotter.
Go figure.
And Samuel Johansson only had ONE child with any of the names of the 7 children and that child has an exact birth date in the tree and the birth date of the other child with a different mother and birthplace was born two years earlier and 6 months before his parents marriage date which is there. It happens, but just one more reason to question it.
I'll leave these here for a little while, but I promised my friend I would save her tree from the destruction of FamilySearch Record Hints, so you might have to see them as rejected Record Hints in the Changes. 0 -
Dale Hein said: I don't like the way you have to click on Previous at the top of the thread or you don't see the huge list of bad Record Hints that have been posted. So I'm posting this again.
Here is another really crazy one. There are 7 birth Record Hints on this father's page for his children.
https://www.familysearch.org/tree/per...
Not one of them is for Ostergotland where all 8 of his real children were born. They are all for Kalmar, Jonkoping or Kronoberg. Go figure. It's not like the father has an unusual Swedish name --Samuel Johansson.
He had two real wives -- Lena Jacobsdotter and Stina Larsdotter. Only one of the 8 mothers in the 8 sources matches one of his real wives. The others are one of these three:
Stina Persdotter
Stina Eliaesdotter
Lena Larsdotter
I guess they all had either the first or last name of one of his wives, but Stina and Lena are VERY common Swedish names and so are Persdotter and Larsdotter.
Go figure.
And Samuel Johansson only had ONE child with any of the names of the 7 children and that child has an exact birth date in the tree and the birth date of the other child with a different mother and birthplace was born two years earlier and 6 months before his parents marriage date which is there. It happens, but just one more reason to question it.
I'll leave these here for a little while, but I promised my friend I would save her tree from the destruction of FamilySearch Record Hints, so you might have to see them as rejected Record Hints in the Changes. 0 -
Dale Hein said: Which is also a problem. Somebody will attach it and merge the sisters.0
-
Dale Hein said: The other thing many of the bad Record Hints I have posted, and I can only assume those others have posted, need to be really looked at along with the data on the page to see the problems. I personally don't have time to explain it to people. I have to spend my time fixing the problems. I really don't think anybody who takes the time to post bad Record Hints here doesn't know what they are doing. Well, other than that we are wasting our time posting (and sometimes explaining) bad hints in a desperate hopeless attempt to make those in control see the problem. I guess in that respect we don't know what we are doing.
0 -
Amanda Spackman said: Please stop attaching record hints to Scandinavians. Two weeks ago I spent at least 4 hours detaching over 20 records from one of my ancestors that had been attached by 3 different people. I haven't even made it to her husband yet who the exact same records were attached too. Last week I had to detach some of those same records again because one of the same people who attached them before attached them again even though I was very specific in why the records did not belong with the ancestor. I messaged to woman and she said that none of my explanations showed up on the records, which leads me to believe that she did not bother checking the recent changes page. She just saw the records there again and attached them. The other thing is that I have tried dismissing the hints and they just stay on the page. They don't go away. NONE of the records even matched with my ancestors as far as dates, ages and places. The only thing that was partly a match were names and even then some were stretching things a bit. When the last name of the person in the hint is completely different from the person that it is being suggested to attach it to, you would think that that would be a clue that something is not right.
I understand that the purpose of the record hints is to "help" people and give them clues to where records for their ancestors are. But it is not working like that. People see a record hint on their ancestor and just attach it without checking the actual source first. This makes a real mess for those of us who actually know what we're doing and know how to research properly. If I have to spend hours every week fixing another persons mistakes on our common ancestor, that takes away time from me being able to do research on my other ancestors.
In my opinion, record hints do not work for a community tree. If this was Ancestry, where everyone has their own tree and they choose to attach wrong records and mess up the tree, at least it doesn't greatly effect others.0 -
Robert Wren said: Dale,
I'm sure you know all of this, but there are multiple inherent problems with Sweden and hinting, which is likely due to the patronymic naming system prior to 1900 and the location change between Län (county) and Länskap (Province) causing confusion. Of course, the very common use of names, with various iterations and similar problems with parish names found in many different counties adds to the problem.
However, their Lutheran Church has done such a great system of keeping track of everyone, for centuries almost precludes the need for hints. ArkivDigital is easy to use and very complete.
I, too, see the same hinting problems, but thankfully the lines I work on don't have as many people 'working' on them.
Scandinavia does create unique hinting problems and I salute you for your diligence in trying to verify, properly source and create 'worthy records'
Tack så mycket!0 -
Dale Hein said: Yes, Robert, I sure do know all of that. I searched microfilmed Swedish parish records page by page for 30+ years for my ancestors. Had to pay to borrow each film, wait two weeks or more to get it, drive to my FHC, spend hundreds of hours searching hundreds of rolls of film, and then typed hundreds of family group sheets (on a typewriter) for many generations of many lines of my non-LDS grandmother's ancestors, sent my temple submissions through the mail, and waited weeks to get back the paperwork I needed to do their ordinances, and then drove 5-1/5 hours to our closest temple where the office printed my cards and with the help of other family and ward members got their work done. NOBODY else was doing any of those lines but me. I went back to the very early 1600's an most lines. And when nFS first started, I could look at their ordinances and know right away which family member did them for me by the year and temple. And until Record Hints, nobody had worked on ANY of those lines but me. The lines on the other side of my family from about 4 generations back were constantly being played around with carelessly, so that kept me busy fixing those. And I would check my grandmother's Swedish lines and think "thank goodness nobody is working on those!" But then Record Hints happened, and all my years of meticulous researching and recording of data is being destroyed by church members who have never done research (obviously!) are dutifully doing just what they've been told -- look for Record Hints and attach them and then do the temple work that hasn't been done. "See how easy we have made this for you? Anybody can do it!"
I will never, ever, ever, ever understand why nobody appreciates the efforts of those who have worked so hard for so many years -- when it WASN'T easy. We come here and try to explain and nobody cares. Our work is treated like a plaything.
And you're right about it especially being a problem for Scandinavian records. In fact, I put that in the title of this thread when I started it. But when Brian Jensen asked for some examples, I have sent him plenty that aren't Swedish or Danish. They're all over.0 -
Robert Wren said: I feel your pain, Dale!
But now you can't say "NOBODY appreciates the efforts of those who have worked so hard for so many years -- when it wasn't easy."
Been There, doing That - 40+ years starting in the FHL basement scrlling thouhg microfim for HOURS!
Keep pluggin' on! Records Worthy of All Acceptation.0 -
Dale Hein said: Oh, my gosh! Somebody in FS went and removed those 7 wrong Record Hints for LH5T-97V! I'm so glad my friend whose tree it is saw them, too. They were there earlier today when I posted the above message. I just went back to declare them Not a Match and they're gone! Not declared Not a Match by a user. Just plain gone like they were never there! That's great, because at least I know somebody who can do that saw the horror of it and removed them. So maybe somebody finally believes us. I am soooooo excited!!!!!!!!!
Now if they would just remove all Record Hints from the whole tree. 0 -
Dale Hein said: Dang. I wish I took a screen shot because there's no way to prove they were there now.0
-
Mary said: It's not just Scandinavia. I just had the same thing happen with my English "Clarke" family. Somebody saw record hints for christening records for the children of John Clarke in West Haddon, Northamptonshire, England and attached them to their ancestor who spent his whole life in Virginia. And these hints were already attached to the correct people, but they detached them and created new profiles! With the record hints, they added seven new children, all born in England, to this man from Virginia. Luckily I caught it and was able to clean it up.
I keep a "backup" tree with my own research and sources on Ancestry.com, so if somebody changes something, I can quickly look at my reference and sources and see what I know is correct. It helps.
There definitely needs to be better user education and "warning labels" of some sort on the record hints.0 -
Mary said: It still isn't easy, especially when you get to Scandinavia or any other country across the pond, or even America pre-1850. The problem is, people assume it's easy. That's a misconception. It's easy to get started, but it isn't easy to do thorough, sound research.0
-
Dale Hein said: Virginia and England have been problem Record Hints in my tree, too.0
-
joe martel said: Dale, was that before or after your Merge?0
-
David Newton said: It's worse than that. People don't just assume it's easy. They get told it's easy by idiot marketers at the major genealogy companies and also by stupid campaigns from the marketing department of Familysearch (notice the common thread there). Even beyond that they get wrong-headed presentations made by various people "called" to positions involving genealogy within the Mormon organisation who tell them to willy-nilly grab as many names as possible without the slightest bit of research.0
-
Juli said: I think the person on the FS Community forum I encountered today must've fallen for the propaganda: apparently, she came up with a birthplace for her husband's ancestor by looking for people with the right name. The town she picked had 13 of them baptised in the right decade, it must be right, no? And then she attached a different baptism as a source for "her" guy -- except the one she chose says clearly in the remarks that the child died at the age of 2. Oh, and she attached the source for the apparently-randomly-chosen baptism to a different (also index-based auto-generated) profile from a different town a century earlier. People really do think that if it's the same name, it's the same person!0
-
Dale Hein said: The Hints were there before the merge, but, honestly, I don't remember if I noticed them still there right after the merge or not. I had no intention of rejecting the Hints yet, because I wanted to give the engineers a chance to see them first. Then I went back last night to do it, like I always do, figuring they know how to see them in the Changes.
But I don't believe that is why they disappeared. I just looked at the data on the deleted record and it had the same death date, birth year and one of the wife's names. If the children's names were there, I would have said that too in my reason statement. I always give very specific reason statements. And there are no children added in the Changes of the deleted records. And the place on the delete record was the same, Ukna, Ostergotland, Sweden, and the wife's birthplace didn't change either when I merged her. So nothing changed in those two merges that made the records any different than they were when those hints were there earlier.0
This discussion has been closed.