Historical Places - Cities that Change Name
Historical Places - Cities that Change Name
by Geoffrey Slinker, Ancestry Employee
The postings on this site are my own and do not reflect the views or opinions of Ancestry.
The postings on this site are my own and do not reflect the views or opinions of Ancestry.
Renamed Places
For example:
North Carolina Hamburgh changed to Glenville.
1827 Hamburgh (later Hamburg)
1891 Glenville
Hamburgh -vs- Glenville
Let's try to find this record.
If you were to type in the place of death, "Hamburgh, North Carolina", and you do not choose from the drop down "place picker", that is you type in the value and click outside of the typing area you would probably expect the search to find the record.
However, there are no results.
How can this be?
If you type that place into "Keywords" it finds the result.
Let's try the search again, and let's use the place picker. Type in North Carolina and pick from the list that appears "North Carolina, USA".
The results contain the record we are looking for.
Why did that work? Because of a concept known as data normalization.
Data Normalization
Data normalization is converting a piece of data into a format that makes it findable.
For example, how many ways can a date be represented?
- Jan 1 2020
- January 1 2020
- 1 Jan 2020
- 1/1/2020
If you were to take a book and make the index for the back of the book for every date in the book, and the book has dates in various formats, would you make an index entry for every representation of the date or would you pick one date format and use that in your index to reference every page for any version of that date?
Now imagine your book is 1,000,000 pages and has at least 100 dates per page, all in different date formats!
Now imagine your book is 1,000,000 pages and has at least 100 dates per page, all in different date formats!
That is the problem computer indices have to address.
However, normalization of different formats is not as easy as you might think.
To tie this to the purpose of this post, and that is historical places, place descriptions can be in many formats and change over time. The process of normalizing places can not possibly know of every place that has ever existed.
However, normalization of different formats is not as easy as you might think.
To tie this to the purpose of this post, and that is historical places, place descriptions can be in many formats and change over time. The process of normalizing places can not possibly know of every place that has ever existed.
Place Searching Tricks
So, why didn't the search for "Hamburgh, North Carolina" find anything? Because it is not a known place. One day it might be a known place and then the search would have found something. So, if you don't find anything what do you do?
Broaden The Search
Broaden the search and choose from the place picker. As in the second search example shown above, pick "North Carolina, USA" from the picker. If there had been other data it will show up, and you will have to page through the results, but at least you have results to work with.
For example, if you didn't know the person's name was Hoty, the results would look like this if you used the place picker for "North Carolina, USA".
There are only 108 results you have to look through.
Use Keyword Search
Keyword searches do not usually use normalization. This gives you a way to search for text as it appears in a record. It has a drawback in that it isn't limited to just the field you are searching for. What does that mean? If you put in the keyword "Louis" it could find the text in a person's name, or in the name of a place, or any field that might have the word Louis in it.
Use the Place Picker to Normalize
You can also type in the place and then pick from the place picker.
The place normalization actually converted it to the only part of the place that it recognized and that is "Jackson, North Carolina, USA".
You can tell that it didn't recognize Hamburgh by hitting "Edit Search" on the result page and examine the query.
You can see that the place picker normalized the data to the nearest known place. This automatic broadening is done so that the search will at least find something. It can be confusing because you type in a place and it changes it to something else. At times this is very helpful and at times it is not what is expected. But understanding what is happening makes it easier to come up with alternative searches that can lead to finding the record you are looking for.
Conclusion
Place names change over time. This can cause difficulty and confusion when searching for places.
If you don't find a record when you search for a place try the following:
- Broaden the place search. Search by State instead of by County.
- Put the place name in the Keyword field.
- Type in the place and then pick the best match from the place picker, which will then search on what places that are known. If the search is for "Chicken Bristle, Kentucky, USA" it will search for "Kentucky, USA".
Comments
Post a Comment