Has the derivation of Event Place from Event Place Original been deliberately removed?
Take https://www.familysearch.org/ark:/61903/1:1:QGV3-T47D as the starting point. This is the historical record for the birth of Anne Morrison to James Morrison and Margaret Anderson Morrison on 19 December 1874. It has an Event Place of Milford, County Waterford, Ireland.
Because I've been working on this family, I had a pretty shrewd idea that I knew who this person was and she wasn't born in County Waterford.
To test this out, I used irishgenealogy.ie and checked all 13 Anne Morrison births in 1874 and 1875. I looked at the Superintendent Registrar's Districts and not one of them covered any part of Co. Waterford. In other words, the screenshot above contains incorrect data.
The actual birth took place (I already knew this from irishgenealogy.ie) on 19 December 1874, to parents James Morrison and Margaret Morrison formerly Anderson, at Glencross (a townland) in the (Registrar's) District of Rathmullan, Superintendent Registrar's District (aka Union) of Milford, County of Donegal. Those 4 placenames are on the certificate.
Several points arise:
Firstly, I assume that the original indexed placename was "Milford", and that "Milford" has been auto-mapped to Milford, Co. Waterford, instead of Milford, Co. Donegal. To be fair, how on earth the auto-mapping is supposed to know which Milford it is, I don't know. I know which one it is because I can see the county on the original certificate. @N Tychonievich please note.
Secondly, I just said that I assumed that the original indexed placename was "Milford". Until a short time ago, I would not need to assume because the system would have recorded and shown both the Event Place (Original) of "Milford" (presumably) and the Event Place of Milford, Co. Waterford, Ireland.
Is this a deliberate removal of Event Place (Original)? Deliberate or accidental, it makes understanding the necessary derivation of Event Place more difficult. For instance, if the Event Place (Original) were actually "Milford, Donegal" then the resultant mapping to "Milford, Co. Waterford" would be obviously appalling. I don't think that was the Event Place (Original) but I simply can't be certain because the item has been removed.
Thirdly, if it is a deliberate removal of the Event Place (Original) - then please explain why this has happened. Are we assumed to be incapable of understanding the mapping process? Does FamilySearch assume that the mapping process is infallible now? I can assure you that we, and several long-suffering mods, know that it isn't. Did the person who removed Event Place (Original) understand its usage?
(I don't think that I'm the first person to note the loss of Event Place (Original) by the way)
Best Answer
-
Ah - thanks for that heads-up.
(PS - I didn't deliberately rate my own comment as the Best Answer - just finger trouble on my part, which I haven't found a way of derating…)
0
Answers
-
It would seem that the (original) notice has been removed even before a correction has been made.
I reported this strange location in an Error Report in July.
If I search for the same record today, this is the extract.
So, it's gotten worse instead of better.
4 -
@Áine Ní Donnghaile - excellent idea. I was looking at that birth event at the top of this thread but had no means of proving what it was like originally. You reversed the process to start with a screenshot of an old entry, compared it to today and showed the definite loss of Event Place (Original).
I've subsequently looked at some of my old bookmarked threads and demonstrated a similar loss of Event Place (Original) on records from England and Wales National Register, 1939 and England Births and Christenings, 1538-1975. So it's not just an assumption on my part of "Shouldn't it show...?"
The really sad thing about this removal is that it throws FamilySearch indexers under the proverbial bus.
I would suggest that the typical place-name auto-standardisation thread in the Community goes something like:
- The original poster says that a Historical Record shows that an event took place at XYZ and the image shows that this is XYZ in a specific county / place /state / country;
- The OP claims that the indexers have wilfully disregarded the county etc. applicable to this event and have indexed it to an XYZ in the wrong county etc;
- Members of this Community try to explain this by saying that the indexers have indexed only XYZ, which can be seen in Event Place (Original);
- The resulting XYZ, wrong-county, etc, can be seen in Event Place and the Community Members explain that this is the result of the background auto-placename standardisation - the Indexers are innocent, OK?
Without being able to see Event Place (Original), the Community members will have difficulty in convincing the Original Poster of what's going on as they will be referring to invisible items that might or might not exist in reality. The worst case scenario is that the OP still believes that the indexers are incompetent and further that Community members are making up excuses. Further, no evidence can be provided to a mod (for onward transmission to techies) about what's going on.
3 -
If I recall correctly, the (Original) designation disappeared for a short time once before - sometime in 2023, I believe. Pollyanna here hopes this is another short-term oops or unintentional revision. But, we really need the Mods/Managers to escalate this matter. We at least need to know what is going on.
I've tried to be consistent when I've reported the placename algorithm problems, starting the title of each with ERROR REPORT. That means I can find them again, by searching on that exact term and filtering on my username, resulting in a list of examples we can show.
3 -
@Adrian Bruce1 You may be interested in one I just reported. https://community.familysearch.org/en/discussion/157883/error-report-placename-standardization-on-the-fly
1 -
@Adrian Bruce1 Thanks for the report. I'm still seeing both Original Event Place and Event Place in some record collections and not in other. I have no idea why that is. But, when the event place is clearly wrong and I can confirm that it was not an indexer error (i.e., it wasn't must a mistake made by a volunteer indexer on one page of records), I am reporting them. I'll report this one.
2 -
@Adrian Bruce1 I've made a few reports of strange place name alterations over the last day or so. And, I had noticed that the word "original" was once again present. Seeing that spurred me to review this old thread that I vaguely remembered. And, lo and behold, the word "original" is once again present on the record, from your example, for Anne Morrison https://www.familysearch.org/ark:/61903/1:1:QGV3-T47D
It is also once again present on my example for Anne McNassar https://www.familysearch.org/ark:/61903/1:1:QPBR-4SSN
As I said on my other threads, small mercies appreciated.
3 -
😉
0 -
If there is a specialist group for reporting errors with the place name corrupting AI, can somebody please let us know. If not, then can we have one please, otherwise the I suspect that the community will eventually be overwhelmed with reports about it.
0 -
@Re Searching We've been told to report in Search, tagging Ashlee C. for escalation.
I know some FS contributors have just given up on getting these corrected.
0
