Correcting a location error
I have noticed on numerous occasions of a location which is shown as the following (without quotes):
"Stokes, Pitt, North Carolina, United States"
The correct naming of the location should be
Stokes County, North Carolina, United States or possibly Stokes, North Carolina, USA.
The reason for this request is to correct a huge number of errors for births, marriages, and deaths showing as a Stokes community in Pitt County, North Carolina (e.g., Stokes, Pitt, North Carolina). However, there is no "Pitt" in Stokes County, North Carolina and the counties that adjoin Stokes Coiunty are:
Patrick County, Virginia - to the north
Henry County, Virginia - to the north
Rockingham County - to the east
Forsyth County - to the south
Surry County - to the west
This error in naming location applies to a huge number of births, marriages, and deaths as well.
Best Answer
-
Just a general note from a user: Blind data search+replace is rarely straightforward. The reality is that supremely-crafted search terms invariably scoop up stuff that wasn't intended and the replacement can do a dozen unexpected things. Getting replacement right without damage can require setting up dummy data to do test runs on - where you might learn that your dummy data is supplying it's own quirks.
I'm just saying that the reality with IT is that few things 'just work'. Almost everything blows up and you have to take time to examine the debris and then you get to watch it blow up again in a new way.
1
Answers
-
The marriages in Stokes County, NC are often labeled as "Stokes, Pitt, North Carolina" when in fact there is no "Pitt" as part of Stokes County, which borders Forsyth and Guilford and Surry counties.
There may be a "Stokes" as a community in Pitt County, NC but Pitt County is on the eastern side of North Carolina.
I've corrected some that I run across and corrected the "Stokes, Pitt, North Carolina" location, but please change all of them to Stokes, North Carolina, United States.
0 -
No one has contacted me, or corrected the error on the location code "Stokes, Pitt, North Carolina, USA."
Once again, there is no "Pitt" as part of Stokes County, NC. However, there may be a "Stokes" in Pitt County, North Carolina but am not sure. The number of errors with the "Pitt" attachment for Stokes County, NC is "HUGE" and it makes little sense to have FamilySearch let the users correct all of them, when FamilySearch could probably correct all of them at one time using your technology.
I have ancestors in both Stokes and Surry counties, but can tell you there is NO PITT as part of Stokes County, so please improve the site by correcting this egregious error.
0 -
This sounds like one of the many auto-standardization errors that were introduced last year. There are some long posts that go into this extensively but I'll just say here that the FamilySearch ran a routine to improve place names in the historical records that worked fine maybe 99% of the time. But when you are talking about billions of records, even a 1% error rate is huge. You can recognize these errors because the records involved will have two event places. The second of the two used to be labeled Original and is the base place name originally set for the indexed batch and is usually just fine. The first listed is the auto-standardized version that has the error. Because the event place is the same for the entire indexed batch, the error will be the same on all of them also.
If you post a specific example of one of the records you are seeing this problem with which you have not corrected in the Search category here in Community under FamilySearch Help, the right people will see it and send it off to the engineers to add to the list of corrections. We've been told it may take a long time to fix them.
Here is an example of one I posted a couple of days ago: https://community.familysearch.org/en/discussion/138713/auto-standardization-error-fjelberg-parish#latest
Do report this in an original post with a clear title.
The trouble usually arrises when the place name was incomplete and the auto-standardization misinterpreted the level of the place name. There is a Stokes in Pitt county:
So the error was that the routine took Stokes as the city name rather than the county name in the records you are seeing.
0 -
Hi @HVJ64 You are correct. There does not appear to be a Pitt, Stokes, North Carolina. (https://www.familysearch.org/en/wiki/Stokes_County,_North_Carolina_Genealogy) However, according to https://www.familysearch.org/en/wiki/Pitt_County,_North_Carolina_Genealogy, there is a location called Stokes, Pitt, North Carolina.
What records specifically are you seeing where the town Pitt is being named in Stokes County? Please provide a link to so the issue can be reviewed.
Thanks so much. Maile 🙂
1 -
For Stokes County, NC (which was created out of Surry County, NC in about 1787), there are literally hundreds of thousands documents for births, marriages, and deaths thart were erroneously labeled as "Stokes, Pitt, North Carolina". However, this pertains to a community or town in Pitt County, NC which is on the east coast of NC, and not in the piedmont.
I would hope FamilySearch has the computer background sufficient to go through all births, marriages, and deaths in Stokes County, and correct the revise the "Stokes, Pitt, North Carolina" to just Stokes, North Carolina, United States.
The problem is in the sort criteria in which shows "Stokes, Pitt, North Carolina, United States" appears just after "Stokes, North Carolina, United States". This leads to a huge number of incorrect data.
0 -
I understand your frustration, but I need a url link to a record that shows the error so that I can submit it for investigation
0 -
Here is an example:
The problem is many people who are posting births, marriages, and deaths in Stokes County, NC are doing it incorrectly since the option of choosing "Stokes, North Carolina" come right before "Stokes, Pitt, North Carolina". There must be thousands of incorrect records in FamilySearch in which the records are in error as it pertains to Stokes County, North Carolina. I've run across well over 100-200 records so far, but am well below what I believe is a huge problem.
Stokes, Pitt, North Carolina is on the east coast of North Carolina and I presume the preface of "Stokes" in this option refers to a town or township in Pitt County, NC.
Regardless, the entire database of names in FamilySearch is messed up with this error as it pertains to Stokes County, NC. I live in Guilford County, NC which adjoins Stokes County and have many family connections to those who lived there, so seeing this type of egregious error is troubling.
0 -
Non-mangled link: https://www.familysearch.org/ark:/61903/1:1:QLMM-Q4KK
Huh: that example actually doesn't look like auto-standardization -- there's only one Event Place field. I think the error happened in index pre- or post-processing, because of the uniformity: out of 1636 index entries on that film, there are a grand total of 68 that say Stokes county instead of Pitt, and all of the ones I looked at out of those were ones that someone had corrected using index correction. The film is correctly identified in the catalog as Stokes county, so I don't know what happened: does index processing now also use the flawed autostandardization routine?
0 -
My concern is that members are not being made aware of their errors, and FamilySearch is not trying to rectify all the existing errors.
If you look up anyone with the "Stokes, Pitt, North Carolina" designation, also look at where they are supposed to have been born and also buried. The issue is, the likelihood of someone being born in Stokes County, NC and yet shown to have married or died in Stokes, Pitt, North Carolina are very slim at best.
There are still hundreds of examples (or in the thousands) of FamilySearch showing borh Stokes County and also Stokes, Pitt, North Carolina.
0 -
Users or members have nothing to do with the origin of this kind of film-wide error. In fact, no human being was likely involved in the origin of this error. That's the whole problem: a human being would have noticed that hey, waitaminit, Stokes is the county, it can't also be Pitt. (Or more likely, a human being would've picked the correct county from the list in the first place, maybe noting with amusement the existence of the town in the other county before moving on to the next task.)
1 -
We appreciate you bringing this problem to our attention. The issue has been reported to the appropriate team, who will research the error. We will get back to you as soon as we have an answer. Thank you so much for your kindness and patience while this is being addressed.
Right now, the best option for fixing these errors is to report them one at a time.
1 -
Thanks a bunch!
As I go thru the edits, I’m correcting them as I edit those I see which are incorrectly assigning the Pitt version to Stokes County
Also, people seem to presume that a current location today is the same as it was in the late 1700’s & 1800’s, which in many cases they are not the same. Example, Forsyth County in North Carolina was created from Stokes County in 1849, lots of people just put Forsyth on records before 1849, but those records are still in Stokes County. What’s fueling the issue is Findagrave does not provide the correct option of defining a city in Forsyth County as also being correctly assigned to Stokes. Example, Winston-Salem wasn’t created until 1913 (merging of Winston & Salem) but people post the City on births, marriages, & deaths before 1913..
Guess you can tell I’m a stickler for accuracy, since lots of people making edits are just compounding the inaccuracy of records and could care less about it, which then makes your job more difficult.
Findagrave needs a lot of work on accuracies of locations. Keep up the good work and let me know if I can be of more help.
Bob
0 -
@HVJ64 Thank you for pointing out the error in place in the North Carolina, County Marriages for Stokes County, North Carolina. We can see that this is a system-generated error. I have reported it to the team that can get the error fixed. We can't predict how long it will take for a fix to be in place. Fortunately, you can edit the place when you encounter the error in records of your own family.
2 -
It is Stokes County, North Carolina and not Stokes County, Virginia which you have in the first sentence. 😊
Thanks for the follow up. I’ll continue to make corrections as I find them. Have no clue if Pitt County, NC has/had a “Stokes” community, but will let you research that one. If not, then the fix would seem to be an easy computer search for “Stokes, Pitt, North Carolina, United States” and replace them with “Stokes, North Carolina, United States”
Thanks for the follow up
0 -
I notice numerous errors in defining a marriage location for Stokes County, NC. Also, I have made several suggestions on this topic, but never get to see any changed that should be made
In a huge number of cases, members have entered Stokes, Pitt, North Carolina as a marriage, birth, or death location, but that is incorrect. There really "is" a Pitt County, NC but there is also a Stokes County, North Carolina and for some reason a "huge" number of records indicating Stokes, Pitt, North Carolina should be changed to Stokes County, NC.
I used to be able to make edits the files, but it appears that option is no longer available to the users and members of FamilySearch to update (or correct) the records. If that is the case, please bring it back, but I'd rather you go into the system and change all the incorrect locations as show in the above paragraph.
It might help if you changed the location to Stokes County, versus Stokes, North Carolina and also Pitt County versus Stokes, Pitt, North Carolina. That way it is clearly distinguished from each other.
0 -
Maybe many are errors, but you seem unaware that there is a town called Stokes and it's in Pitt County NC. It's northeast of Greenville.
1 -
@HVJ64 - not all records have the option to edit. Editing the record is a relatively recent addition to FamilySearch, and more records will have the option to edit in the future.
0 -
N Tychonievich's 20 May comment states:
I have reported it to the team that can get the error fixed. We can't predict how long it will take for a fix to be in place.
The list for fixing these is very long. We do not know when you will see a change. Until then, you are welcome to fix the ones that have the ability to to edited.
Mod note - I have merged all your discussions about this topic into one place.
0