What has gone wrong with UK Census record layout ?
New drop-down chevrons have appeared that expand to contain data from elsewhere in the family group, or alternative options that are not applicable and cannot be selected.
examples
https://www.familysearch.org/ark:/61903/1:1:M7R2-94X
https://www.familysearch.org/ark:/61903/1:1:W9WS-L2M
The Birth year drop-down is a good example of how bizzarre this is. The alternatives are the birth years of the other people in the group.
Is this a halfway step to some further impovement, or has an undemanded change been allowed through the checks and verification process ?
Answers
-
@SerraNola Could you take a look, please? The examples are indeed displaying odd alternate details.
4 -
The default position is to have the record "closed", as shown below, with the option to "Open all". The problem at present appears to be the record(s) being displayed the "other way round", with a need to "Close all" to get to the usual presentation.
0 -
The first URL that @Re Searching lists is the 1861 for John Evans, his wife and family. There are 7 people in that family but only 5 names in the persona's source records - both parents have children named after them so the duplicate names have not been added.
Secondly, I have no real idea which family member this persona in the source is supposed to refer to. This John Evans, according to the Source Record, is 10y old and married. 10y old refers to the son named John Evans and "married" refers to the father named John Evans.
if I then look in the source record at "John Evans's Extended Family", the first entry is for "Wife - Mary Ann", suggesting this John Evans is the father. But the line number is 10, which is the line number of the son! Oh - forget that - all members of the family have line number 10…!!
Shouldn't a screen output be meaningful? This isn't….
4 -
I amended my previous comments a short while ago, as they made no sense! (Said "opened" when I meant "closed".) I've just been opening already attached sources and, yes, it is rather a pain to have to keep clicking on "Close All" in order to view the record in the way to which we have been accustomed. Hopefully, this is not the new "default" viewing option and the engineers can revert to how it was, asap.
1 -
Further to the initial query, there might be evidence that some automatic manipulation is affecting the data in the records. In the following example, the nationality of all persons is presented as DUTCH when they are clearly not all Dutch.
https://www.familysearch.org/ark:/61903/1:1:XW84-T83
1 -
Again, all personas in the record have the same line number - 5 in this case, though here there are exactly 5 lines in the schedule (one schedule per page in the 1911). The nationality of Dutch has been applied to all members of the household, instead of just the wife. (The others have no nationality in the original, being British - the nationality data is correct in FindMyPast).
For anyone who doesn't have access to the original image - or can't be bothered to look - this is surely little short of disastrous?
2 -
Yet another reminder that we should always look at the record and not rely on the index/extract. This morphing of information just seems to get worse.
2 -
This is worse than morphing - weird values "hidden" behind the drop-down arrows are weird but possibly harmless unless you look. Altering the Nationality of a whole household to Dutch is corrupting information.
2 -
By the way, it isn't just UK Censuses. https://www.familysearch.org/ark:/61903/1:1:6R39-T35V?lang=en is part of the Canada, Census, 1931 and it shows the weird values behind the drop down arrows as well.
I checked a couple of people with UK censuses - all have weird values behind the drop down arrows on all of their UK censuses.
0 -
Thanks for this report. We should be able to have an update on this in a few days. If any more collections are found, please post them here.
3 -
@Adrian Bruce1 I think the line numbers have been out of kilter for a long time, but since you have raised it again, it has prompted me to think of a possible mechanism that might explain the issue.
Consider that the data in a given household group is imported line by line. Ordinarily we would expect that the line would increment as each person is added. In which case if the line number was being latched, then all records would show the first line number. What we see is that all records show the number of the last line in the group. This makes me think that the data is being imported in reverse order starting with the last record, and the line number being written is not being updated as each subsequent record is read.
The presence of all the additonal data in the drop-downs makes me think that the same buffer record is being used to populate from import and then write out to the next record in the group. This will create an edit history of all data in all records.
0 -
Oh dear @SerraNola - I should have expected you would ask a very pertinent question.
https://familysearch.org/ark:/61903/1:1:MN6T-KK8?lang=en is part of "United States, Census, 1870" and has weird values behind the drop down arrows. Eg every name, age, estimated birth year appears behind the drop downs concerned.
"Denmark, Census, 1901", has the same weird values behind the drop downs e.g. FamilySearch (https://www.familysearch.org/ark:/61903/1:1:QL6P-1RHF : Sat Jan 18 15:31:31 UTC 2025), Entry for Johan Christian Juliussen and Thora Astrid Vilhelmine, 1901.
Probably no surprise that the Scottish censuses have the same issue: "Scotland, Census, 1901", , FamilySearch (https://www.familysearch.org/ark:/61903/1:1:7Y37-TCPZ : Wed Feb 19 23:13:24 UTC 2025), Entry for James P P Bruce and Maggie Bruce, 1901
Also perhaps to be expected is the 1939 Register for England and Wales: "England and Wales, National Register, 1939", , FamilySearch (https://www.familysearch.org/ark:/61903/1:1:7G2X-KFN2 : Sat Mar 01 03:01:48 UTC 2025), Entry for George Bruce and Mary E Bruce, 29 Sep 1939.
I guess if these appeared so quickly, it might be possible to disappear them easily... Hmmm Well, one can hope...
2 -
@Adrian Bruce1 Oh boy…..this seems to have affected every collection that has multiple personas in a household. Thanks to all of you for discovering where the data has been corrupted. Many collections are still matching what was indexed, but others are quite messed up. If anyone has additional feedback on this issue be sure to post. I am meeting with engineers this afternoon so hopefully will have an update tomorrow.
3 -
@SerraNola - "… multiple personas in a household…" That was my suspicion - or my guess, to be more accurate!
I tried a very few other collections such as Electoral Registers, tax lists, and passenger lists but all seemed OK - none had the weird data behind any drop down arrows. And none seemed to have anything like a household - not even the passenger lists I looked at - I can see where the family is because I know who they are, but it doesn't look like there's the equivalent of a household in the data. I probably missed something though!
1 -
I think this is fixed! This is a great example of Community working together as we should to provide adequate information for the engineers. Thank you!
3 -
@SerraNola , @Adrian Bruce1 Yes it is indeed fixed, with the exception that the line numbers are still identical for every record in a household group as Adrian pointed out. This is as it was last week.
0 -
I agree that the examples that I gave above have all been fixed (excellent, please pass my thanks to the team@SerraNola!) with the exception of the line number oddity, which isn't always present - while 1911 census for England & Wales https://www.familysearch.org/ark:/61903/1:1:XW84-T8W?lang=en uses the same line number for all personas in the household, a 1931 census of Canada https://www.familysearch.org/ark:/61903/1:1:6R39-T35V?lang=en has different lines numbers.
Frankly I'm not too fussed about the line numbers - I tend to just look for the family in the image and take it from there - the 1939 Register quotes line numbers but they're not actually visible on the sheet of paper, you have to count down the page, which is too much like hard work for me!
0
