Why was the same film indexed twice??? (And the two indexes released two days apart?

LegacyUser

Tom Huber said: Henry Stanley Huber
4 September 1893 – 1 December 1986 • 9V9H-143

(https://familysearch.org/ark:/61903/1... : 25 October 2019)
Document Information:
Affiliate Publication Number 76193916
Affiliate Publication Title Veterans Administration Master Index, 1917 - 1940
Affiliate Film Number 107A
Digital Folder Number 105208557
Image Number 03275

Henry Stanley Huber
United States, Veterans Administration Master Index, 1917-1940
Name:
Henry Stanley Huber
•
Event Type:
Military Service
Event Date:
6 Sep 1918
Event Place:
Beatrice, Gage, Nebraska, United States
Event Place (Original):
Beatrice, Nebr, United States
Birth Date:
4 Sep 1893

----

(https://familysearch.org/ark:/61903/1... : 23 October 2019)
Document Information:
Affiliate Publication Number 76193916
Affiliate Publication Title Veterans Administration Master Index, 1917 - 1940
Affiliate Film Number 107A
Digital Folder Number 105208557
Image Number 03275

Henry Stanley Huber
United States, Veterans Administration Master Index, 1917-1940
Name:
Henry Stanley Huber
•
Event Type:
Military Service
Event Date:
6 Sep 1918
Event Place:
Beatrice, Gage, Nebraska, United States
Event Place (Original):
City of Beatrice, Nebraska
Residence Place:
Beatrice, Nebraska
Birth Date:
4 Sep 1893

----

Yes, I see that there is a single line difference between the two entries.

However, the question remains? Why twice? And why two days apart?

Note that the image also has both indexes...



I have not attached either record as it makes no sense to have both be for the exact same image on the exact same film.

LegacyUser

Adrian Bruce said: My concern with this sort of issue is always that if one "digital film" can be indexed twice - it must be possible for another not to be indexed at all.

I am sure that lots of people think this is boring stuff, not relevent to their family history, but there are reasons why IT project management got lots of boring (yes) disciplines added to it - and trying to stop risks of double indexing (which is wasted effort) or missed indexing are 2 of those reasons.

Please note that I am not saying it's easy to control this much data at this level of detail (it's really hard!), but since we on the outside have no visibility of detailed progress, it gets ticky trying to understand what processes might be in place.

LegacyUser

Paul said: Tom

You write: "I have not attached either record as it makes no sense to have both be for the exact same image on the exact same film."

I really don't see it makes sense to attach the same image even if it has been subject to multiple filming. Robert Kehrer promised some years ago this problem would be eliminated and surely the digitisation process was / is the ideal opportunity to deal with at least part of it.

Once it has been clearly established the content of two microfilms is identical there seems absolutely no need for duplication. In a number of cases I have three or four identical looking sources attached to individuals in Family Tree. True, there is the argument that an indexed record might contain more (or less) detail than one indexed from another film. But where the indexed records are identical, or if neither has been indexed, there is absolutely no reason to digitise the two (or more) microfilms.

LegacyUser

Tom Huber said: I can understand the problems if a second filming takes place of the same original records and two indexes are subsequently produced. In most cases, one of the indexes will be superior to the other.

In some instances, the filming process (cameras used) have improved to the point where the original film(s) have clarity problems over the film(s) produced with newer equipment at a later date.

In this case, I have not bothered to attach either record simply because if FS does what I hope they do, they will discard one of the two sets, since the indexes are of the same (NARA) film (not two filmings of the same original records).

LegacyUser

Jeff Wiseman said: Tom, I found a good example of what you are talking about (i.e., multiple fillings, etc.) I posted a note on this since it may be of use to others:

https://getsatisfaction.com/familysea...

Note that this is definitely NOT the same as the situation that this thread was opened on. It does seem very odd that such a duplication would occur. A few years back FS had accidentally created many true citation duplicates that they subsequently cleaned up. However your situation could not be one of those as the slightly different index data that you have in your example proves that the citations were not true duplicates.

LegacyUser

Tom Huber said: I just ran into another NARA film where two indexes were created for the same image. One is more complete than the other and one of them has the 23 October 2019 date. The other date is from a couple of months earlier: 25 August 2019

In this case, it is the U.S. World War I draft cards that got duplicated... 23 October 2019 is https://familysearch.org/ark:/61903/1... index and 25 August 2019 is https://familysearch.org/ark:/61903/1... -- same film and same image number.

LegacyUser

Tom Huber said: In the above reference to the WWI Draft Card, the problem is that one index is more complete than the other. This explains the duplication, but does not explain why both indexes are still "live"

FamilySearch, please respond on what the plans are? Should both be attached, even though they are the same image from the same film, or what?

LegacyUser

David Newton said: They did exactly the same with the Hampshire parish registers. Same items indexed multiple times in a database released in 2019. They were informed by me that it had happened. Not a peep from them about fixing it.

LegacyUser

Tom Huber said: Incidentally, I posed this same issue earlier today in the Indexing community. Hopefully, we'll get a response and if so, I will post the response here.

LegacyUser

Adrian Bruce said: As I see it, there appear to be about 4 issues:
1. Are the 2 (or more) indexes identical in which images are covered? Or are some images indexed only in one?

2. Does each index record contain exactly the same data? Tom's examples include ones where they don't.

3. What should any patron do when discovering multiple indexes for the same image? Bearing in mind that duplicates have previously been withdrawn?

4. Why the duplication in the first place? To be honest, this is actually the least interesting issue to me. It might be that it was intentional - a film was partly indexed collecting one set of data values; it was realized that data was inadequate so the film was reindexed using a wider set of data. Maybe. Or, of course, control of the indexing process was inadequate so the same film was assigned for indexing twice. Hey - it happens - I guess. If a mistake has been made then please admit it, tell us that you'll do better next time but most importantly tell us the answer to issue 3 - what to do.

LegacyUser

Tom Huber said: You are correct in #2. There are two indexes of the identical film (usually not imaged by FamilySearch) and images, released within days of each other, and usually have only one or two differences in the index of the same image.

In this case, the duplicate indexes remain in the system, often both generating hints for the same person for the same image. This is where I first discovered this anomaly.

#4 is really my question. Why was a duplicate index released -- again within days of the other? The only thing that I can think of is that somehow, somewhere, the duplicated index was missed and/or somehow created. The duplicate index was either done by another firm or a duplicate project was set up with slightly different index fields.

LegacyUser

David Newton said: With the Hampshire examples it's the same image, the same film and exactly the same information indexed.

LegacyUser

Tom Huber said: Thanks for letting us know that not all indexes are different.

I have to wonder if this isn't some kind of glitch that happened when the index was loaded -- that the person that pushed the index into production twice -- the second time by accident, perhaps thinking that it didn't make it into production.

In the incident that I've cited, that also could have happened -- that it was pushed into production and then pushed back to indexing for completing the indexing effort, and then pushed into production. I haven't checked the actual dates and its late, so this is something that I need to do when I'm more awake.