Source linker is doubling names
I tried to add Ellin Cooper to her family in accordance to many documents in the search results. What I find is the search from the new software can't find her so this is what I get. Note that the first and last names are duplicate in the names fields. I went back to the old version and was able to add her. Here is her sister that has the same problem in the new software.
Answers
-
It seems your problem is resolved, @Cheryl Ann Nelson. You have attached the 1930 census to the family. https://www.familysearch.org/tree/person/details/GKK6-LP5
0 -
I can see that you're using Source Linker -- which is some of the oldest software still in use on FamilySearch's website. It was introduced in 2014, and hasn't been substantially changed since.
I don't know what caused the doubled surname, but luckily, the field is fully editable, so you can fix the error.
The reason it was lining her up with Henry is that Henry's ID was in the box.
1 -
When I select a person in Research Help to add a sibling to their parents, then select REVIEW & ATTACH, I arrive at the 'Attach Historical Records to Family Tree' screen. Usually the parents line up with the parents and the siblings line up with the siblings on their respective lines.
About 3 days ago I started noticing a parent and the sibling, that I am attaching, displayed under "Other On Record" rather than on the correct line.
When I drag the parent to the correct line it displays correctly, but when I add the sibling, the name doubles.
0 -
That's a very strange bug in Source Linker, but I don't think it has anything to do with Research Help: that's just the method you used to get to Source Linker.
Unfortunately, this is not something that can be tested on the beta site, because it only has a copy of the tree, not of all the records, so while it shows record hints, they don't actually work there. I was also unable to re-create the error using the hints on my relatives; they're all lining up the parents and children correctly, and adding a child from them does not produce the doubled names that you're seeing. It must be something about the way that particular index is set up. (That would also explain why we aren't seeing more reports of this problem. I very vaguely recall someone else describing it, a while ago, but I have no memory of what became of that thread.)
1 -
It's not particularly uncommon for the Source Linker to fail to line up people on the record with the corresponding persons in Family Tree. I see this happen most often when the record doesn't contain relationship information (it may seem that it specifies relationships, but sometimes it wasn't indexed that way, or the record collection wasn't structured with relationships). When that happens, it's easy enough to drag the person into the right position, so I'm not particularly worried about this relatively rare occurrence.
The weird duplication of names shown in Screen 3 is clearly a bug. I haven't seen it. Perhaps it is a problem with that particular record collection.
1 -
This just started happening a few days ago. I have been following the same process for a few years now and they have always lined up. My concern is that someone new to this process will miss the attachment of siblings and spouses. It is happening very often now. Strange for sure.
0 -
This has just started happening to me tonight. When I have a source and move a name from 'Other on Record' up into the children area to insert it. It duplicates the first name and the last name with no space in between. Its happened 3 times in less than an hour.
And before you go there, last year my stats contribution was over 30,000 and this year I'm nearly at 15,000 and I have never had this problem before.
Granted, not every contribution I've made has been this exact same effort, but I can confidently say I've exercised this code literally thousands of times in the past 2 years.
Something has changed.
0 -
@bekasu, what index(es) or collection(s) were you attaching from?
0 -
Regarding the two different issues brought up here:
1) People showing up as "Other On Record" is how the Source Linker has functioned for years when there is no indiction in the index as to who the people are. This is similar to having parents' names line up alphabetically on the left hand side when not indexed as male or female in the record which often leads to them not matching up properly with the parents on the right hand side. Learning to check who those other people are and making sure everyone lines up properly or is added properly is just part of working in the Source Linker and the ability to shuffle things is what gives the Source Linker such nice flexibility.
How often names do not line up fully in the Source Linker is completely dependent on the historical record format.
2) Names getting doubled when creating a new person does seem to be new. I tried multiple ways to get it to happen in the indexes I usually work in but could not replicate the problem. I agree that this does sound like something in the structure of some indexes is causing the issue, not the Source Linker itself.
1 -
Use this record to duplicate the problem. GM1L-QL8
Select the research help, then review and attach. Drag the child Giuseppe Vassalotti into the Children line and insert. Check the record then click 'Add". You can see the name duplicated there.
0 -
Well, it's definitely doing the doubled names for me, too.
Whatever the cause/mechanism, it's not solely the lack of relationship and the need to drag the person into place, because I just did one where the mother needed to be dragged up from "Other on Record" and then added, and it did not double her names.
I haven't encountered one among my hints where a child needed to be dragged-and-added, but I'll keep an eye out.
1 -
I can confirm the same double name error when using that particular source record. I would not attach that record for a while in hopes that the right moderator (would that be you @Maile L ? or do you @Alan E. Brown have any connections?) sees this post and passes it on to the right programmer to be evaluated.
I do, however, have a theory! (Sorry. Theory is probably the wrong word. Pure speculation based on a wild guess is what it really is.)
This is one of the newer indexing projects that are at least partially indexed in the language of the collection as you can see by the date being in Italian. There is a second date in English:
I wonder if in the underlying data that is not displayed there is both an Italian first and last name and an English first and last name despite there being no difference between the two and if the Source Linker is pulling both versions of the name when creating the person.
1 -
You may have something there, Gordon, because while my example is also from a very new index, it's in Hungarian, which means that its name fields probably mess with Source Linker's structures regardless -- first is last, last is first, etc. etc. Therefore, whatever process is causing the doubling probably hits a snag somewhere and has no chance to manifest.
1 -
I'm getting the name doubling thing when I use SourceLinker for the 1940 US Census. I've now seen it happen about 3 different times over the last 24 hours.
0 -
I have reported the bug. I will let you know when I hear back. Thanks guys.
1 -
I would like to report a problem with the Indexed Records duplicating the given and surnames, see example:
This has only been a problem for a week or so. Thanks for your help. Arleen Craig Ellis
0 -
This has been reported here: https://community.familysearch.org/en/discussion/comment/519853 and as you can see in the first July 17 comment that the problem has been sent off to the engineers to get it evaluated and, we hope, fixed.
2 -
Julia -- sorry for the delay. Let me get back from hospital/caregiving for husband and see where I was -- definitely would have been something on my German/Prussian family info. I've been out of pocket for family health issues.
0 -
Julia -- sorry for the delay. Let me get back from hospital/caregiving for husband and see where I was -- definitely would have been something on my German/Prussian family info. I've been out of pocket for family health issues.
And, of course, I couldn't stop myself. I searched and found the following screenprint. Bottom of the right column shows the duplicate first and last name. Bekasu
And here is the history showing the activity completed with the first/last names doubled up.
0 -
So Bekasu is also working with the 1940 US Census.
Drat, so much for @Gordon Collett's multilingual-index theory.
The thing I noticed, though, is that the child being added did not need to be dragged anywhere: he's in the index as a child, but isn't listed among his siblings (presumably because he may still be living), so Source Linker lines him up below the existing siblings and gives the "Add" option without any dragging or other adjustment needed. So indexed family structure or lack thereof is apparently not relevant to this bug. I haven't a clue what is relevant.
1 -
Here's another hint that something funny is going on in the background data somewhere that the Source Linker is picking up on. I just noticed this from a collection that is quite old:
I don't really want to detach this source then use it to try to create a new person so I'll leave it to others to find a similar example where the tag insists there are two names but the indexed information shows no sign of that and to test out what happens when you create a new person through the source linker for the person.
1 -
OK, I went ahead and test it out. Does not point to the source of the problem either:
1 -
Hey guys. I merged both discussion into one place so that we can have all the info together. With that merge, references and links to the other discussions were removed.
0 -
When adding a new person using the source attachment tool the First name & Last name fields are being populated with the name twice:
First Name: Pally APally A
Last Name: HaskinsHaskins
I am using Google Chrome on Windows 10, both are up to date, have cleared cookies and tried a different browser, behavior unchanged.
1 -
Gorden -- what a kindred soul. I also find it difficult to suggest a possibility without just taking that next step and 'trying it out'. lol My great downfall when I had to estimate project time lines. Bekasu
0 -
Hi everyone- Sorry I have been so slow to jump into this discussion but it has taken a minute to figure out the correct team to get to the bottom of this. I just wanted to let you know that they are actively working toward a resolution and I will keep you posted as I get updates. Thanks for your patience as they work to clear this up!
2 -
Today it is working correctly for me
0 -
When I look at the example from @bekasu, I do not see any doubling. (https://www.familysearch.org/search/linker?pal=/ark:/61903/1:1:KWKL-GV3&id=GJ46-SVT)
3 -
I was working on several names within the Source Linker since yesterday and the name duplication is not apparent.
1 -
Let's give a round of applause to the engineers who tracked down the problem and fixed it so quickly (in terms of program debugging and updating time)!
6
