Please fix the wrong place names in the "Russia, Lutheran Church Book Duplicates" record set!
So today I've been browsing through the searchable collection "Birth, Marriage & Death, Russia, Lutheran Church Book Duplicates, 1833-1885)" and I'm utterly appalled by the quality of the indexing there. The most egregious problems are with the place name indexing, which I'm assuming is a combination of a) system coding only allowing for standardized place names, and b) the "Help fix place names" volunteer task and people who know nothing about given locales being able to do this and thus making horrible mistakes.
The birthplaces in most of these records will be small hamlets or even individual farmsteads that do not appear on any sort of modern map, yet they get mapped to "standardized" locations all over the world - such as in Haiti, Malawi, Belize, Suriname, etc., all unlikely places for a 19th century Russian Empire peasant to be born, especially if the record was still being created in the Russian Empire.
Even in records for Saint Petersburg, if the birth place is just listed as "Petersburg" it gets mapped to "Petersburg, Friesland, Netherlands" rather than the more logical answer of where the record was created.
On a similar note, "St P-Burg" - a clear abbreviation for "Saint Petersburg" is instead being transcribed as "Suriname".
"Reval" (the old name for Tallinn, Estonia) gets mapped to "Reval, Nord-Ouest, Haiti" when Tallinn is the far more logical answer here. And for some reason Riga, Latvia, is instead mapped to "Riga, Irkutsk Oblast, Russia". And Libau (old name for Liepaja, Latvia) maps to "Libau, St. Clements, Manitoba, Canada" instead of the more logical nearby answer.
"Gnadenthel", a Bessarabian German village, instead maps to "Gnadenthal, Orange Walk, Belize". Other German-named villages or German cities are mapped to namesakes in Canada or the United States instead of their proper German or Russian Empire locales (Leipzig, as one example).
These are just a few examples.
And there are still others with places that don't even appear anywhere in the record, old names or no! I have no idea who has been indexing these records, but please stop. As a professional, I know that these are obvious mistakes, but there will be plenty of amateurs going in there and going "oh cool!" and just taking these horrendous transcriptions as facts.
Answers
-
@Antra Celmins When you looked at some of these entries, did you notice if some of the locations were tagged with the word "original"? If that is the case the the issue likely stems from the place name corrector program. See this thread for more info. https://community.familysearch.org/en/discussion/116253/existing-historical-records-issues#latest
0 -
@Áine Ní Donnghaile No, I'm not seeing any "original" there. As an example:
I don't know in which part of the process it is done, but it is happening a lot.
0 -
Hmm - I just searched for the 2 records at the top of your screenshot, and each displays just "Reval" - with no country.
URLs if someone else wants to check:
Pauline https://www.familysearch.org/ark:/61903/1:1:QVRK-YRY9
Christine https://www.familysearch.org/ark:/61903/1:1:VT9Y-VHV
0 -
@Antra Celmins - Are either of the following two threads helpful? They were both started by the same member.
0 -
@Antra Celmins, let me try to address the "Reval" issue. First, I don't see anything that suggests an indexing issue with regard to the Birthplace. The index record lists, "Reval," which is exactly what is found in the original record. What is interesting is the Search Results, which shows the Birthplace as "Reval, Nord-Ouest, Haiti," which is your concern. The question then becomes, where did the Search Results find that place name.
I think we find the answer in the place names that are available to FamilySearch; we see these names at: https://www.familysearch.org/research/places/?pagenum=1&pagesize=20
Notice that when you enter "Reval" into Search Places, there is only one match: Reval, Nord-Ouest, Haiti. It appears that the Search Result uses the standardized places names that the algorithm determines best matches what it sees as an event place. This is the reason that we always advise against using only a single place name when doing a search - for example, using only "Petersburg" as a Birth Place may well return results that you might not expect.
I believe that what is needed, to resolve the "Reval" issue, is the addition of the place name: Reval, Estnia, 1219 - 1918 (or something similar that may be more correct, etc.) to the set of standardized place names in FamilySearch. You can do that at the above URL, by using the option to Suggest a New Place.
Please note that I used the Wikipedia site, Tallinn, which is now linked to the article, to come with the dates for Reval.
I am curious about your comment that "Tallinn" would be a more logical place name than Reval. That would certainly be true if we were talking about a current event; however, the events that we are looking at are in the 1800's (taking the case of Pauline Therese Krüger Schneider as an example). Generally, and if possible, it is a good idea to use place names that match the the timeframe of genealogical events.
I hope this helps in answering some of your concerns.
@Áine Ní Donnghaile, I tried to duplicate your search, but did not get the same results as you; I saw only "Reval, Nord-Ouest, Haiti" as the Birthplace. I tried several browsers, but always the same result. I thought that was very strange, and certainly have no idea why you would only see "Reval" as the Birthplace.
0 -
@Mike357 I'm using Windows10 and Chrome 98.0.4758.102.
I started my search from the Collection https://www.familysearch.org/search/collection/1469151
0 -
@PiperTWilson No, if it was the same issue, I wouldn't have made this post. My post is not about the renaming of locales into different languages based on the current administration of the area, my post is about how places in the former Russian Empire are being mapped to locales all over the world that are not remotely anywhere in the boundaries of the former Russian Empire. Haiti, Malawi, Papua New Guinea, Belize, etc. have never ever been a part of the Russian Empire.
@Mike357 It looks like you're misinterpreting my words. Never did I say that "Reval" was not in the original record. Never did I say that it is inappropriate to use Reval in a historical context. I'm saying that that problem is that the search maps it to a Reval, Haiti, when the record would mean Reval, the former name for Tallinn, Estonia. I will try the "suggest a new place", but the problem is that there are hundreds if not thousands of records in this dataset that are mapping places in the former Russian Empire (and some in the former German Empire) to places all over the world. This is a problem in the system that needs to be addressed. I can't possibly fix all of them. Nor am I the right person to fix them all, because my specialty is the Baltic countries, and I am not as familiar with old spellings of places outside of there, but even I can tell that where "Huwa" is written in a record as a birthplace for hundreds of Finns living on the shores of Lake Ladoga in the mid-19th century, that it isn't meant to be a place in Papua New Guinea.
0 -
Can you please have this database checked for the placename standardization program error?
I've now tested on 5 browsers - Chrome, Firefox, Vivaldi, Edge, and Opera. All except Chrome display Reval in Haiti, which is incorrect.
Collection is https://www.familysearch.org/search/collection/1469151
Thanks, as always, for your help!
0 -
FamilySearch has been aware of this type of problem since at least September 2021. See the previous discussion
"Standardized Locations" in Puerto Rico are shown as being in "Australia / New Zealand"
0 -
I have another one for you, NOT a German From Russia example. I have an ancestor couple who lived in Knox County Indiana Territory early 1800s, but when they got married in 1810, they were able to pop over to Knox, Marshall Islands in the Pacific. Must have been a great beach wedding and the perfect place for a honeymoon because Knox Atall Marshall Islands is completely uninhabited to this day. This index cannot be corrected, by the way. Edit is grayed out.
https://www.familysearch.org/ark:/61903/1:1:41X6-D42M?from=lynx1UIV8&treeref=29HC-9Z8
0 -
@Gail S Watson N Tychonievich has asked that each report of this issue be made as a separate post as that makes it easier for her and her team to be sure each problem is handled.
AFAIK the option to edit the index on FamilySearch is not available when the image is not viewable.
1 -
Áine Ní Donnghaile I see a lot of moderators creating new posts from an existing. Do I need to create the new post or will it be done by someone else?
0 -
@Áine Ní Donnghaile Is your suggestion "each report of this issue be made as a separate post as that makes it easier for her and her team to be sure each problem is handled" only directed to Gail as a method of separating out those records from the ones I'm talking about, or are you saying overall that each individual incorrectly mapped record have its own post? Because if the latter that's going to be thousands of posts (and I'm not sure where to make these posts).
0 -
Mine was put in the queue to be fixed.
0 -
@Antra Celmins I have replicated the problem with the Russia, Lutheran Church book Duplicates, 1833-1885 and will get a report off to the engineers. It looks like they fixed an auto-standardization problem--but not completely. When viewing the pages showing the indexed information, we are not seeing the errors. For example: https://www.familysearch.org/ark:/61903/1:1:VT9Y-FKC. But we are seeing the errors in the search results.
3
