Census occupations missing

djb2

As of some "update" a couple of weeks ago, occupations are no longer showing in census transcriptions. This affects at least the England & Wales census transcriptions for 1851, 1861, 1881, and 1891. Those for 1871, 1901 and 1911 still retain them (or at least some). I've checked a random sample of people from my tree in several different areas, people for whom I've previously recorded an occupation, so I know it was in the transcription! No notice, no explanation, just vanished. Anyone know anything about this? Is it permanent? or just some over-eager web interface designer shoving out some new (probably unnecessary) code without properly testing it?

MandyShaw1

I've just looked at the underlying JSON data for this 1891 census record: /ark:/61903/1:1:4264-WPZ.

While the occupation (Police Constable) is absent from the display it still shows in the data:

… "facts" : [ { "type" : "http://gedcomx.org/Occupation", "value" : "Police Constable", …

The corresponding 1911 entry shows the occupation both in the data and on the screen.

So this sounds to me like a user interface problem only.

djb2

Thanks Mandy! So yes, it looks like some display update has caused the field to become invisible in the interface… I do wish web developers would stop mucking around like that, online auction sites are another one where they can't just leave alone!

Paul W

There have been many issues with the E&W census collections over the years. FamilySearch has responded by suggesting that Find My Past formats / updates the records and they just publish them in the way they are sent across to them. Find My Past has taken a different view, saying certain matters (such as the loss for a number of years of millions of records from the 1871 collection) is due to no action on their part.

From viewing https://www.familysearch.org/search/collection/list?ec=region%3AEngland%2CplaceId%3A1986340%2Cregion%3AUnited%20Kingdom%20and%20Ireland&fcs=placeId%3A1986340&offset=0 it appears all collections - except for 1901 & 1911 - were updated in June (2024). Unfortunately, these updates appear to have led to a "downgrade "in the data contained in the past, and the latest exercise appears to follow that pattern. You will have difficulty in getting a definitive explanation on this and can only hope that this particular issue can be put right, either before or at the time of the next update.

djb2

Thanks for the answers!

How do I view the underlying JSON data (not that I can do it for every record until they fix the bug) but it would be useful to be able to check for some key individuals.

Thanks,

Dave

MandyShaw1

https://www.familysearch.org/platform/records/personas/<id> will give you the underlying data for  /ark:/61903/1:1:<id>.

Hit the JSON button to make it easier to read.

Doesn't work for all 1:1: records though (not sure why).

djb2

Wow, thanks, I never knew that! Amazing what's hidden away isn't it?

Incidentally, it doesn't work for this person in 1861:

https://www.familysearch.org/ark:/61903/1:1:M7R4-TRX
https://www.familysearch.org/platform/records/personas/M7R4-TRX

…gives me "401 Unauthorized - Failure in upstream call: FailureToGetPersona (upstream 401)"

However, the same person in 1851:

https://www.familysearch.org/ark:/61903/1:1:SGFS-ZZL
https://www.familysearch.org/platform/records/personas/SGFS-ZZL

…does indeed give me the raw data (albeit with no occupation, as he's only 10 at the time).

I wonder whether the 1861 failure is to do with the fact the record is unattached to a tree, but the 1851 record is? Anyway, I've learnt something very useful, many thanks!

MandyShaw1

It appears to fail on the ones showing 'Image Unavailable' and to succeed on all others I have tested (whether or not the record is attached to the Family Tree).

Re Searching

@MandyShaw1 Thanks for this enlightenment. I used it to examine the place name data for a record in the 1851 E&W Census. It revealed that the data presented in the 'Event Place' was sourced from EVENT_PLACE_ORIG and that was wrong, it had been replaced with the County and Country. If the three items EVENT_DISTRICT_ORIG, EVENT_COUNTY_ORIG and EVENT_COUNTRY_ORIG are concatenated, it would be better, because it would at least reveal the District. Now I want to try to discover what update caused the place to get destroyed in the original. It is also strange that the entity EVENT_TOWN_ORIG is not present at all.