What's up with the duplicate WW 1 draft cards?
What's up with the duplicate WW 1 draft cards? I am getting hints for people for whom I posted this
Answers
-
Many records sets have been indexed more than once. I have many folks with 4 or 5 copies of the index of their birth records.
-1 -
My question had to do with draft cards. I am used to linking the same marriage over and over (although it is a waste of time), but in the past couple weeks something huge has changed with WW I draft cards. I am getting hundreds of hints for records I already linked, in many cases years ago. In some cases when referencing the "old" source, I get a message that I should refer to the "new" source. I have wasted some time linking the new DUPLICATE source, but have chosen to ignore the hint as much as using it. I do not see that the new source is in any way better than the previous one. If the image quality is different, I haven't seen I just don't get it. Seems like SOMEONE could explain this.
0 -
And something done about the excessive hints currently being generated. Either replace the old source system-wide and/or stop generating hints where the source has been linked previously.
0 -
This is pure speculation, but perhaps they're testing out new computer-assisted indexing methods by applying them to already-indexed images?
I don't deal with U.S. records much, and hence haven't seen any such duplication, so I truly don't know: are these really duplicates? As in, is it the same scans or photographs of the same pieces of paper?
1 -
The images appear to be the same digital images, but one I viewed a few days ago had 2 different indexes. The record was for my great-uncle. His middle name was indexed incorrectly, 2 different ways. So, I take that as evidence of 2 separate indexing projects. Whether one or both was handwriting OCR, I can't say. Certainly, I didn't understand why the name was misspelled in the index, but then I know what the name should be.
I edited/corrected each one independently.
Edit to add: the last update of the WWI draft registration record set -United States World War I Draft Registration Cards, 1917-1918
32,580,666
10 July 2024
1 -
@AdrianClift Thank you for bringing this to our attention. We are going to escalate this issue and see what we can find out. Thank you for your patience. We'll let you know what we find out.
1 -
@AdrianClift
To help us resolve this issue, could you provide specific examples (Names & PIDs) for the individuals you are seeing multiple hints for from the WWI draft cards?1 -
@Caleb L Since I corrected 2 for a great-uncle just a few days ago, I have the info close at hand. PID LHJ6-H87 for Edgar Lumpkin Burgess.
2 -
I am not getting duplicate HINTS. I am getting loads of new hints for folks for whom the source (former source?) has been long since linked to individuals. If after linking a duplicate record to the same individual, if you bring up the previously linked card, you get a message to the effect that IT is a duplicate and ahas been "retired." I don't see what is being accomplished.
0 -
From personal experience, I'd suggest that you check carefully whether the retired source record and the new one, really are duplicates, as is suggested.
We had a series of "retirements" some time ago in English and Welsh parish records. We were told by the system that the retired entry was a duplicate of the other - our problem was that sometimes the retired entry was not a duplicate and indeed it contained more data than the (presumably) newer replacement. Both entries pointed to the same event in the same parish register, but they had been indexed differently.
We asked for the cessation of the retirement process and reinstatement of the non duplicates, but I don't recollect any specific outcome nor explanation of why the tidying was thought necessary.
My only reason for bringing this up now, is to suggest a close check on the equivalence of the data content of retired and new entries would do no harm. The detection of duplicates in our case was not infallible.
1 -
On the WWI draft registrations, I have seen no message of duplicate/retired. As I mentioned earlier, the images appear to be exactly the same, but the indexed information is not an exact match between the 2.
In my experience, the appearance of 2 sets of the same draft registration is less recent than the original poster. I've been seeing them for quite some time. For my maternal grandfather, I attached 1 version in 2016, and the 2nd version was attached in 2020. For his younger brother, I attached both versions on 19 Nov 2020.
-1 -
@AdrianClift If you would like us to help you, we need examples of what you are seeing. Will you please provide PIDs for some of the people that have had this issue? More examples can help us figure out if there is a reason for what you are seeing or if there is a bug that needs to be fixed.
@Áine Ní Donnghaile Thank you for the example you provided. 🙂 In looking at the two entries, I can see that they are two different record numbers so they are separate entries in the database. I'm not sure why though, so that will be my next step. I will ask around and see if I can find out why.
0 -
The other two PIDs, in case they are useful: my GF is LHJ6-H33 and his youngest brother is LHJ6-C2Y.
I do see something a bit unusual in the Source Linker on LHJ6-C2Y. As I said earlier, I attached both versions of the draft reg on 19 Nov 2020. But, one version now shows an "unfinished attachment" in the list of sources. That would appear to have been captured from the name of his employer/father, with the same initials. I don't recall that field having been indexed before. Perhaps there has been some post-processing? That field is not indexed in the other version.
1 -
GQH5-CHC GQ1N-987 G3KB-9XG
https://drive.google.com/file/d/1Nrxiw041UF0TQq3Vi0IyQC31QEwAf4AQ/view?usp=sharingKN4X-8D3
0 -
GL3P-JJW LK4W-K2C 9JWR-V3Z LJ1G-RNG KH5W-W4K KCKF-RR3
0 -
I am also seeing the duplicate/retired error message on records for which only one draft card source is present.
0 -
I'll just throw one of mine in… (I haven't checked further)
Profile is for John Arthur Pickstock, PID G7V3-G64
Someone attached this WW1 Draft Registration card 2y ago: https://www.familysearch.org/ark:/61903/1:1:KZJ6-5HN
When I go there, it says "This record was a duplicate and has been retired. We recommend using the most current copy"
The "current record" that it suggests is on https://www.familysearch.org/ark:/61903/1:1:7DBX-RJ6Z
When I compare the 2, the images appear to be the same (both image 3922 of 5682) and both have the same URL for the image excepting that there's some further terms in the new image URL starting with "personaUrl". No idea whether that's significant.
The contents of the 2 index records are close. Here's the original, marked up with the changes to reach the new one.
Name
Arthur Pickstock
Sex
Male
Birth Date
13 Jan 1888
Birthplace
Eccles, Manchester, England
Citizenship Place
United States
(After this, the new index record has an additional item "Country of Citizenship = United States". Strictly speaking neither Citizenship Place nor Country of Citizenship appear on the form as it actually asks "Are you a Natural Born Citizen, A naturalised Citizen, An Alien, or have you declared your intention?")
Race
Caucasian
Event Type
Unspecified (is now "Draft Registration" on the new index record, which makes far more sense. Indeed, I wonder if that's why the new version has been created?)
Event Date
from 1917 to 1918
Event Place
Essex, New Jersey, United States
Affiliate Publication Number
M1509
(omitted from the new Index Record)
Affiliate Publication Title
World War I Selective Service System Draft Registration Cards
(omitted from the new Index Record)
Presumably the Affiliate Publication stuff would be found at the collection level… Somehow…
1 -
@ Adrian Bruce1
I still don't see what is being accomplished here. These are hardly meaningful improvements given the amount of additional work that is being generated.
0 -
@AdrianClift said
"I still don't see what is being accomplished here. …"
I tend to agree. In the absence of any FS explanation that I know of, I'm forced to fall back on experience, etc, (with a lot of "etc" 😉 ) but my best guess is that FS believe that the justification for the new indexing is the new Event Type value of "Draft Registration", which would, if I understand correctly, cause Source Linker to offer a Custom Event of "Draft Registration" when the new index record goes through Source Linker. Aside from that, there is no new data.
I have been dubious about some of these new Source Linker events but - perhaps because I have an interest in military stuff - that's actually one I'd use. But I could have set that up without Source Linker's assistance.
As it is, I have a source index record attached to my chap's profile that is red flagged. I feel I have to obey the instructions and replace it with the most current copy, but the system doesn't help me to do that - the most current copy doesn't appear as a hint to my chap. Sure, it's easy enough to follow the instructions to get to it (providing you haven't unattached the old one first!), but it's just offered me a raft of profiles that might be the correct one - including at least one woman!
So yes, I'd suggest most people will have acertain amount of work to do for, in this case, virtually no visible benefit.
By the way, unless anyone knows better, we still don't know the long term implications of "This record was a duplicate and has been retired" - will such records eventually be deleted? The only safe thing would be to carry out that extra work.
3 -
@ Adrian Bruce1
Exactly. So what happens to a "physical description" entry based on the prior source? I have been linking to both where available, but I have been entering detailed data form draft card records for maybe five years.
0
