Who is responsible for the Accuracy of Place Names?

Question

I have addressed this issue before, but here goes.

Ancestry.com, when they were creating their Gazetteer, erroneously assigned many settlers' location as Monroe, Amherst County, Virginia.  This location has been used extensively through many dozens, if not hundreds, of profiles.  The time frame this location was used was late 1700's to early 1800's.  Monroe is, today, a small, unincorporated  community of of less than 5,000 population.  It did not exist during this timeframe. It is not anywhere near the obvious migration routes from Augusta county to the Greenbrier River region where most of these settlers ended up.

The correct location should have been Monroe County, Virginia, after 1799, Greenbrier, between 1778 and 1799, Botetourt, between 1770 and 1778, Augusta, between 1745 and 1770 or Orange County before 1738.

Answer 1

That is whgy Managers should not rely on Ancestry without checking the data.

lots of Ancestry trees assign Resiudence to the USA for people who lived and died in the UK, just because some emigree founded a town named after their British place of birth.

Answer 2

The person entering the place name is ultimately responsible for its accuracy.

It would be nice if software backed up the accuracy by doing a data check on the place name and time period before the profile is saved.

The best method would be to have a database of correct place names and time periods and have the person select the correct location for the time period of the event.

Comments

The best method would be to have a database of correct place names and time periods and have the person select the correct location for the time period of the event.

This is what the FamilySearch location suggestions attempt to do. We may not always agree with the suggestions, but it is definitely a start.

Creating our own database that could do this (accurately) would be nearly impossible, or require resources that would be impossible from a free site.

---

Creating our own database that could do this (accurately) would be nearly impossible.

So WikiTree has a database of locations and dates. Are you claiming that this information is not accurate? Are you saying that a locations database can't be created from the existing data that WikiTree has? This doesn't sound right to me.

require resources that would be impossible from a free site.

What resources are needed and what is the cost?

---

So WikiTree has a database of locations and dates. Are you claiming that this information is not accurate? Are you saying that a locations database can't be created from the existing data that WikiTree has? This doesn't sound right to me.

You are correct that WikiTree has an existing database of locations and dates. However, the issue at hand isn't the existence of the database, but rather the accuracy and reliability of the data it contains. The current databases on are primarily based on user-provided data. This user-generated nature means that the information can only be as accurate as the data entered by individual contributors.

For example, if a user enters "XXX" into a location field, the database now includes "XXX" as a location. Clearly, this does not correspond to a recognized geographic location. So, while the data in our current databases is invaluable, it isn't inherently verified or standardized.

While it is technically feasible to create a more accurate database using the existing data as a starting point, doing so would involve complexities and resources that extend beyond simple data aggregation.

What resources are needed and what is the cost?

Let's tackle this question from another perspective:

Numerous well-funded and well-staffed projects such as Pleiades, the Getty Thesaurus of Geographic Names (TGN), China Historical GIS (CHGIS), and the European Holocaust Research Infrastructure (EHRI) have attempted to create comprehensive databases or digital archives. These projects, albeit on a much smaller scale than a global historical place name database covering all of Anno Domini, still encounter significant logistical, technological, and financial hurdles.

These initiatives often have the backing of major academic institutions, international organizations, or significant grant funding. Despite these advantages, they face ongoing challenges in data accuracy, technological adaptation, and sustainable funding.

Regarding financial considerations, the initial development and continuous updating of such a database would require a substantial financial outlay. Projects funded by entities like the National Endowment for the Humanities (NEH) often receive grants ranging from $50,000 to $300,000 per project phase. Larger or more complex projects might secure more substantial funding, especially when spread across multiple grant cycles.

Given these precedents, the resources needed for a global historical place name database would be considerable, encompassing not just financial investment but also extensive expertise, time, and sustained collaborative efforts.

---

(continued)...

If there was enough interest to tackle this on a much smaller scale, either as part of WikiTree or as a standalone project open to external funding, I would be interested in working on it. We could begin by clearly defining the scope of the project, perhaps initially focusing on a specific region or a particular historical era that resonates with our community's interests. Once we develop a viable method and solution, we could then leverage that data within WikiTree.

---

Have you been speaking to Ian about our ideas about a solution for this? 

---

Steven, I think there is plenty of interest in this, and I would suggest creating a separate channel on discord under ‘WT Apps’ so that the discussion can commence.

---

I have thought about this also, but concluded that it is too big a project for a volunteer organization. Family Search already has a database of Places which WT uses so the better choice would be to feed any changes/updates to FS.

---

The problem with that approach is that their naming standards are not the same as that of WikiTree, so you will never get them to correct things to what WikiTree wants. (I have tried.)

---

It's a big job, but we can do better than FamilySearch.