Auto-Standardization Error
@N Tychonievich, are you still the one to report place name auto-standarization errors to?
Here is another one I heard about today: https://www.familysearch.org/search/record/results?f.collectionId=4237104&q.marriageLikePlace=Gausdal%20H%C3%B8ifjellssanatorium%2C%20Gausdal%2C%20Oppland%2C%20Norway
All these marriages show as having taken place at Gausdal Høifjellssanatorium, Gausdal, Oppland, Norge. This is a hospital.
The place name should be simply Gausdal, Oppland, Norge.
How is it going getting these fixed? One of the first ones I reported on September 21, 2021 is still not corrected.
Answers
-
We seem to have entered yet another phase where the originally-indexed text is not shown anywhere on the index detail page, making it look like all of those marriages were indexed as taking place at the sanatorium (e.g. https://www.familysearch.org/ark:/61903/1:1:683J-X43L).
This is incorrect, but there is no way to tell, currently, what the event place field originally contained.
Likewise for one I reported two years ago as being autostandardized to Uganda (https://www.familysearch.org/ark:/61903/1:1:66D3-1RWM). It has since been corrected, but in the process, the originally indexed value of "B Gyarmath" appears to have been thrown out: there is no sign of it on the index detail page or in the index-mangler.
Or compare Gordon's 2022 screenshot for this Norwegian burial (https://community.familysearch.org/en/discussion/127835/error-report-auto-standardization-reversal-of-place-name-followed-by-truncation) with what it shows now.
Still the incorrect, truncated place, but no indication anywhere that it was chosen by the bot, not a human, nor any sign of the actually-indexed text.
4 -
@Gordon Collett Yes, I'll get that one reported too. I do see some of these marked as resolved from time to time. But I can't figure out what order they are working in--certainly not addressing the ones first reported first. A fair number of them were corrected in a system update a few months ago. Right now I have 154 open tickets for auto-standardizations--down from 300-some a few months ago. So, there is hope.
0 -
@N Tychonievich Any chance you can get a list of places that have been corrected for Norway and send it to me? It would be nice to be able to confirm that they were corrected correctly.
0 -
@Gordon Collett Sorry, but I don't have that info. When I get a notice that a ticket was resolved, I just remove it from my spreadsheet.
0 -
Ok. Guess it should have occurred to me to keep a spreadsheet of my own so I could easily go back and check once in a while for progress. I'll do that going forward with new ones I run across or am told about.
0