Fiction from the New Source Linker
Scenario:
Widowed mother living with her children and grandchildren. In Source Linker NOW there is an "Unknown Spouse" to be added to the profile of the long-deceased spouse.
Specific example: PID L2HJ-DSN in the 1900 census.
Surely not.
Thank you.
項留言
-
Fiction? But, as they say, based on a true story…
For what it's worth, I suspect that this "Unknown" spouse for Mira Rushford is inspired by the fact that the original image shows her as being married 43 years ago - and widowed now. In the desire not to lose anything from the index, an imaginary spouse has been created for Mira who she married in 43y ago in 1857.
There's also a further "Unknown" on that Household Census's Unfinished Attachments, who might, on that basis, represent the person who Mary Shawcross married 11y ago.
Now, I can imagine the response of certain people to my suggestion being along the lines of "But the index is only a Finding Aid, it's not supposed to include everything…" And I wholly agree with that. In my view, a deceased person should not appear in the index to the Census.
Further, there is a FamilySearch precedent for deceased people not appearing in indexes. I keep reading the same basic question in relation to the indexing of directories - basically, the OP asks "If the entry says 'Mary Smith, widow of John', am I supposed to index John Smith?" And, so far as I remember, the answer is always "No, you do not index John Smith, he's deceased and isn't in the directory". Surely, this is the same thing? There is a deceased Mr Rushford (presumably) - should he be indexed? Between the 1900 Index and Source Linker, he apparently has been indexed. Based on the precedent of the directories, he should not have been indexed.
In other words, I agree with the OP.
Just to be clear (maybe), if I were starting with that 1900 census form and no data had been entered for that family, I might create a Mr Rushford who married Mira 43y ago - but that is my decision as a researcher. There are risks involved with that approach, such as names being misleading - I should not be prompted down that path by an index.
1 -
I've been hoping I would encounter a census to attach where I had NOT already entered the deceased spouse to see what the source linker would do with that scenario.
And I spend time merging "unknown" profiles into known persons. I surely don't want the source linker creating more of them.
0 -
As I read the indexed record, the census knows that Elmira is the mother of Joseph. But there is no father listed in the census. Thus, the only thing that the source linker can present is that in the INDEXED RECORD the spouse is unknown. You do not need to attach, but there is no other expected solution, since source linker will not look beyond the indexed record for data on the left. Deceased or not, it still will show unknown as a placeholder in Source Linker.
0 -
@ScottSeegmiller No, please no. The old source linker didn't have this fiction.
There is no husband for Elmire mentioned in the 1900 census. The solution should not be to create an "unknown" placeholder.
Joseph's father, Mary's father, and Elmire's husband are all the same person. Alexander died in 1891, and his profile is heavily documented. He doesn't need to be recalled in a record that does not pertain to him. Joseph is an adult and head of the household.
With the Tree Builder Project and the US Census Project, all those unknowns will create clutter for the rest of us to clean up. Again.
2 -
@ScottSeegmiller said
"… source linker will not look beyond the indexed record for data on the left … "
And I'm in total agreement with that.
"… the only thing that the source linker can present is that in the INDEXED RECORD the spouse is unknown. You do not need to attach, but there is no other expected solution … "
But there I do have a problem. I do wonder if we're slightly at cross purposes at this point, so forgive me if I seem like I'm repeating anything. I would say that there is another solution - namely that the unknown spouse should not appear in the index record at all.
This was the way that I understood census index records to work - there are 6 lines in this household - therefore there should be exactly the same number of personas in the census index record. Instead, there are 8 personas - made up of the 6 from the census page plus 2 unknown spouses. So either by some miracle, I've never come across an incomplete census record before, or things have changed for at least this census record.
… You do not need to attach [the unknowns] …
Unfortunately, this isn't a particularly good solution. If we don't attach the Unknown Spouse, then what we have is the "Unfinished Attachments" message, which clearly worries a lot of users. I'd be perfectly happy not to attach the Unknown Spouse and then deal with the "Unfinished Attachments" message by Dismissing it. But I do worry about how many users are happy with that method.
Conversely, if we do attach the Unknown Spouse, then that probably leads us into extra merging downstream when the Unknown is merged into the real spouse. Extra work…
2 -
And, if "we" don't get to the record first, the Census Project or the TreeBuilding Project surely will and create more "unknowns" for us to fix later.
2 -
So waitasec: did Source Linker up and take it upon itself to add (superfluous and erroneous) index entries to the database?!??
Something certainly did, because the index detail page has two entries for "UNKNOWN" in this family: https://www.familysearch.org/ark:/61903/1:1:MS3X-C9B.
Were those (inexplicably) already there before the source was attached, or are they really the (equally inexplicable) result of the new Source Linker?
1 -
I truly believe that this **** is a New Source Linker invention. I had not seen it before this frightening occasion.
We who have indexed for years know that we don't create fields for someone not in the record.
They were there before I attached the census to the enumerated family, and I did NOT attach the unknowns for obvious reasons.
1 -
Lets send it over to the engineers and ask them to give us an opinion.
1 -
The engineers replied:
The "UNKNOWN" persons are on the record data; Source Linker doesn't ever create record persons.
0 -
I'd like to hear from the Indexing team because we know we don't create a field when indexing. Somewhere, somehow (post-processing?) that field has been created.
3
