Historical versus current jurisdictions
Separate fields should be provided in Vitals for Birthplace, Place of Christening, Place of Death, and Place of Burial, for both historical and current designation of place names. With the current system, there is a hodgepodge of designations, and worse, the researcher who attempts to enter a historical jurisdiction, such as a location within the former Holy Roman Empire, will not be able to standardize their entry. This is a complete failure of the system to accommodate historical place names, which is supposedly what we are supposed to enter.
Comments
-
Jeffrey
It must be an extremely difficult task for the Places team to add just about every place in the world in a format appropriate to the relevant time period in history.
If you feel such places need to be added to the database, visit the Places group at https://community.familysearch.org/en/group/68-familysearch-places and put your suggestions. I believe you will have to sign-up to be a member of the group, but responses from the team appear to be quite positive, so I believe it will be well worth you doing this.
0 -
You can always standardize a place. The Places database is constantly growing so the number of exact matches between data and standard is getting larger all the time.
For those places that are not in the Places database you may have to standardize the correct historical data with the next best standard or with a larger geographical region. You do not need to have a map pin. The map pin only means an exact match between data and where it shows on the timeline map. The lack of a map pin means the location on the map is approximate.
For example in the 1700's the farm Vikanes was spelled Vigenæs. It is located in the parish of Stord which in the 1700s was spelled Stordøen. It is currently in Hordaland county but prior to 1919 it was called Søndre Bergenhus. These historically correct items are not all in or needed in the Places database but I can still enter the historical forms and standardize just fine like this:
This is correctly standardized.
If you have a place in the Holy Roman Empire that you cannot figure out how to standardize properly, post it here with what you want to display, some history of the place through time, and what the place is currently known as and I can show you how to standardize it.
0 -
Gordon
With respect, you appear to have standardised the place name to the form in which it is known as today.
For places in England, Family Tree offers me an option of formats - up to 1801, ending in "England" and from 1801, ending "England, United Kingdom".
Surely if Jeffrey does what you suggest (and I agree, he has little choice at this moment) he is acting against the Family Tree convention of standardising the place to the form it was known in any specific period of time?
Hopefully, he can persuade the Places team to add the relevant place names in the form deemed appropriate to the period in question.
0 -
I really, really wish that FamilySearch could find a way to explain what is really going on.
When they say something like "enter the place name then pick the correct standard," what they really mean is "enter the historically correct place name then link the correct latitude and longitude to it. If it so happens that the name you want to use is the same as the text representing the correct latitude and longitude, then use that text. If that text is different is some way, use your own entry but still tack on the correct latitude and longitude."
The Places database can only have one displayed name per spot on the globe, that is per latitude and longitude, for each historical time period for that spot. The displayed name can be different in each time period, however.
So you can have the time period up to 1801 parent jurisdiction be "England" and from 1801 onward be "England, United Kingdom." However you can't attach both of them to the child jurisdiction for the same time period unless you have a duplicate entry in the database with the same latitude and longitude.
I oversimplified my example. The farm Vikanes has always been called Vikanes. But go into any one old parish record book and you will find it spelled differently on potentially every page over the span of the book, variously alternating between Wiganæs, Vikanæs, Viganæs, Vikanes, and more. A different spelling is not a different name. A choice had to be made for the display name and that decision was to use the current spelling to represent the latitude and longitude for that farm.
0 -
Here is an example that illustrates the current position quite well. Go to https://www.familysearch.org/records/images/ and enter "Tendring" in the place name field. True, some of the options in the drop-down menu are for the wider area of Tendring, Essex (civil registration district, hundred, etc. - which includes other parishes). But, for the village / parish location alone there are options for pre and post 1801time periods, and for the status of Parish, Village or Populated place.
When it comes to the "Images" section of familysearch.org, it does make a difference which option is picked as the available images relating to Tendring are only found under what has been considered (applying the relevant algorithm) to be the most appropriate. But, I have to select, in turn, each of the Tendring options - as some of the image sets cut across the separate time periods (e.g. where do you place a set of parish register images covering both the pre and post 1801 periods?) and other times FamilySearch (or the algorithm!) just gets it plain wrong.
Now, when it comes to entering the place, from the same list of options, in Family Tree (under an individual's Vitals) there has been a discussion in the past (at GetSat) as to whether or not it matters which option is chosen - especially if the place name format is identical. That is, does it matter if I select any "Tendring, Essex, England, United Kingdom" format or does the category (e.g. civil registration, parish, hundred, poor law union) have to be appropriate, too? In other words, is the algorithm designed to search for records based on the place name itself or on category, too?!
All that, and there is still (as you mention) the other issue of display name as opposed to how that name has been standardised. I have been quite shocked to find that many of my perfectly correct-looking parishes in England have been standardised as a place of similar name in the United States. (Hence the algorithm leading to the offering of record hints for an event relating to someone of the same name, but who lived 3,000 miles away, I suppose!)
In summary, I was originally just trying to emphasise the difficulty the Places team must have in offering multiple options for places with (or which had) the same co-ordinates. Quite how the different algorithms are applied from that point (i.e. having added "every" time period / place related option to the database) is probably a matter that we will never get to the bottom of!
Altogether, there are seven options to choose from for events relating to Tendring, in Essex:
0 -
I share you concerns that by increasing the specificity of places, which I do think is a good thing, some strange things are happening. I think there are two problems. First that users of Family Tree who agree in principle that places need to be entered as historically accurate as possible are going to be surprised to learn, as the Places database grows, that they are going to have to really figure out what all these historical places really are. Secondly, reading what you are finding with the image browser and some of the parish records you see, I'm concerned that some of the algorithms that handle place names are getting outdated and hope that FamilySearch is working hard on updating them. Not only are some of the standards in Family Tree getting way off, in the historical records some of the original place names are being updated incorrectly. Two that I have reported are that all the records for Åkre parish in Hordaland, Norway, which were just fine, were updated to have the standard of Åkre Kirkegård which is a cemetery in a completely different parish; and that Nedstrand parish in the 1891 Norwegian census was updated to show a residence of the city of Nerstrand in Minnesota. I hope these growing pains get sorted out soon.
0 -
Thank you for adding these further comments, Gordon.
I also raised a post at https://community.familysearch.org/en/discussion/83990/how-best-to-record-correct-standard-burial-place#latest regarding the best way to record a burial place. It should be a straightforward case of using the closest, standard match to how the place was known at the time of the event, but for users wishing to locate the cemetery (perhaps with a view to paying a visit to their ancestor's burial place) it would probably be far more useful to record the current location name.
0