Adding a Source by Example

Thursday, July 23, 2020 by Fouke Boss

Earlier this month, user [SH] asked me several questions on how to improve the source references in Centurial. After composing a detailed reply, I realised other users might benefit from these practical guidelines too. In this blog, we take a look at improving several everyday source citations.

Introduction

In Centurial, all information and claims are entered in the context of a source. This way, Centurial will always be able to tell which evidence originated from which source. Before entering the data from a source, Centurial requires you to enter the exact details of the source you are studying.

As described in several earlier blogs, entering these source details serves 3 goals:

  1. We need to know the exact provenance of a source, so we can classify the quality of the data in the source.
  2. We need to capture the source, repository and layout details of a source, so we are able to retrace the source later on.
  3. We need to create a source citation, which is a well versed summary of all the source details. The source citation identifies a source in a unique way.

In this blog I'll discuss 3 every day examples of sources, and I'll have an in-depth look at the best ways to improve provenance and details.

Example 1: VanGoghMuseum.nl

For this first example, let's have a look at a biography of Vincent van Gogh - the famous painter that always serves as an example in these blogs - on the website of the Van Gogh Museum which is located in Amsterdam, the Netherlands:

https://www.vangoghmuseum.nl/en/art-and-stories/vincents-life/from-dark-to-light

I could enter this source manually, but in Centurial there is a better way of adding source details from the internet: Let's use the Centurial browser extension. This browser extension tries its best at importing all source details (and genealogical information and claims if possible) from the web site you are currently viewing in your browser:

While importing the data, the browser extension scrolls through the web page to capture a screenshot of the web page. Sometimes, like in this example, the result is less-than-perfect (some parts of the screen shown twice), but such a screenshot is still useful as it captures the state of the web site at the time of import.

As shown in the top left corner of the Source View that pops up, you can see how Centurial has created a source reference based on the available data in the web site:

Although this source citation is looking good already, it's always advisable to double-check the source details manually after an automated import. By clicking the button (marked with the red arrow), we can have a closer look at what exact details Centurial imported:

By default, the browser extension sets the provenance of any web page to web page, as shown in the provenance panel (1). For sources that come from the world wide web, Centurial supports several other source types, as you can see in the source type panel (2) on the left of the source dialog. In this particular case however, the source type Web page is the suitable provenance: An unnamed person (an employee of the Van Gogh Museum) has authored an online biography of a part of the life of Vincent van Gogh, based on an array of sources. As this biography is not one of the more specific source types (it's neither a Blog, a Social Media Post, a Wiki Entry, or any of the other types), the generic Web Page is the prefered source type.

The browser extension was also able to import the various fields almost perfectly. The web page title was copied to the Title field, it was able to extract the correct web site title and url from the contents of the page (together creating the Repository), and it correctly copied the URL of the web page into the Url field. Also, it set the Accessed date field to the current date.

After closely studying the structure of the website of the Van Gogh Museum more closely, one improvement can be made to these source details. Starting at the home page of the website, this web page could be accessed by selecting Art & Stories > The Life of Vincent van Gogh > Biography, 1886-1888 respectively; this is the Path of the web page. Setting this field accordingly yields the following, improved first reference note for this source:

Van Gogh Museum (https://www.vangoghmuseum.nl/ : accessed 19 July 2020), Art & Stories > The Life of Vincent van Gogh > Biography, 1886-1888, “From Dark to Light”.

In the source list, the shorter bibliography item is presented like this:

Van Gogh Museum. https://www.vangoghmuseum.nl/ : 2020.

In this first example, the source details were imported almost perfectly by Centurial. In the next example, this is certainly not the case.

Example 2: BillionGraves.com

For the second example, let's take a look at the website BillionGraves.com, which is an online database of gravesite information, which includes transcriptions and images of many cemeteries all over the world. In particular, let's have a look at the grave of Vincent van Gogh:

https://billiongraves.com/grave/Vincent-Van-Gogh/27891546

Again I use the browser extension for automated import. This time the screenshot created by Centurial is much better. And it even was able to import some person information for Vincent van Gogh and his brother Theo, as can be seen in the Information Panel:

Again, Centurial has imported the source as a Web page, and it comes up with a first reference note similar to the one in the first example:

BillionGraves (https://billiongraves.com/grave/Vincent-Van-Gogh/27891546 : accessed 19 July 2020), “Vincent Van Gogh 1853 - 1890 BillionGraves Record”.

This time that default choice for Web page is not the best possible choice. First of all, notice that the BillionGraves website is actually a virtual archive (a database), which you can tell from several clues: the home page is centered around a search form, allowing you to search the archive; the URL of the search result contains an identifier (27891546), which is probably the ID in the corresponding database record; and the layout of all the result pages is exactly the same. This is how to mark a website as a virtual archive:

Looking at the database entry for Vincent van Gogh more closely also reveals that the entry contains data that was extracted from the gravestone: the name and dates were transcribed from the stone, and a picture of the gravestone was uploaded to the database. So the history of this source (the provenance) starts with the gravestone being created in the 19th century. Many years later, on 4 June 2018, a volunteer named Viluk photographed the gravestone, and uploaded this to the BillionGraves database. Today, I access this database entry on the BillionGraves website. So the correct provenance would be:

Database entry of a photograph of a cemetery

Let's use the source dialog to update the provenance of the source:

Next, add the missing details to each of the layers. The first layer is the database entry layer, for which most fields are already filled correctly by the browser extension. I only have to supply the Entry for field:

For the second layer, the photograph, I enter the photographer and the date of the photograph:

In the final layer, the cemetery, I copy the cemetery name, place and access data from the website. Note that I leave the section, row and lot fields empty, as they are not provided by the website. Also the accessed field is left empty, as I did not visit the cemetery myself. Instead, I accessed the database entry, which therefore has its accessed field set to the current date.

Now with the provenance updated and the additional fields provided, the first reference note is looking much more accurate:

BillionGraves, database (https://billiongraves.com/grave/Vincent-Van-Gogh/27891546 : accessed 19 July 2020), entry for Vincent van Gogh; citing Cimetière Communal d’Auvers-sur-Oise (Auvers-sur-Oise, France; Avenue du Cimetière), Vincent van Gogh gravestone; photographed by Viluk, 4 June 2018.

In the source list, this source is grouped like this:

BillionGraves. Database. https://billiongraves.com/ : 2020.

Please note that besides cemetery and gravestone information, websites like BillionGraves or Find A Grave also display additional information from other sources in their database entries. This Vincent van Gogh entry from BillionGraves for example also shows a Life Story (which was taken from Wikipedia as clearly indicated) and exact birth and death dates (which were not taken from the gravestone as this only shows the years). If you were to use this information too, the provenance note should be changed into database entry of an unknown source, as the information is then coming from multiple sources of which some are unknown.

Example 3: FamilySearch.org

In the final example of this blog, let's have a look at the birth certificate of Vincent van Gogh using FamilySearch:

https://www.familysearch.org/ark:/61903/3:1:939F-KXQM-B?i=368&wc=MVN2-ZNP%3A1057402601%2C1057322201&cc=2026223

The Centurial browser extension contains some specific logic for the FamilySearch website, which allows Centurial to do a much better import job:

Instead of a screenshot, Centurial was able to import the scan of the vital record. Also, it was able to tell that the provenance of this source is actually a digital image of a vital record. (Note: Centurial is not able to tell the exact type of record all of the time; in those cases, the provenance will be digital image of an unspecified source and needs to be updated manually).

In this case, the first layer is a digital image, and Centurial aced the import of the details for this source, so no manual improvements are required (in some cases, the Credit Line field starts with the word 'citing', which needs to be removed manually). Credit Line, Repository and Path fields were taken from the citation in the information tab at the bottom of the FamilySearch web page:

For the second layer Centurial was able to determine the record type (in this case a vital record) and to import some of the fields. We then manually add the missing fields based on the contents of the vital record:

Please note:

  • The Repository field is left empty, as I cannot tell with certainty where the vital record is currently being held. FamilySearch says it's being kept at the Nederlands Rijksarchiefdienst archive in 's Hertogenbosch, the Netherlands, but I have not checked that for myself; that's why this information is in the Credit Line field of the digital image layer and not in the Repository field.
  • The Page, Number, Item of Interest and Date fields are taken from the vital record itself.
  • The Label was taken from the cover of the vital record book, which is also accessible through FamilySearch:

These manual changes again improve the first reference note:

Zundert, burgerlijke stand, “Register der akten van Geboorten over den jare 1853”, p. 9r, no. 29, Vincent van Gogh (31 March 1853); image, “Netherlands, Noord-Brabant Province, Civil Registration, 1811-1942”, Family Search (https://www.familysearch.org/ : accessed 19 July 2020), Zundert > Geboorten 1850 > image 369 of 476; citing Nederlands Rijksarchiefdienst, Hertogenbosch (Netherlands National Archives, Hertogenbosch).

... which in the Source List is grouped into:

Zundert en Wernhout. Burgerlijke stand. Database with images. “Netherlands, Noord-Brabant Province, Civil Registration, 1811-1942”. Family Search. https://www.familysearch.org/ : 2020.

Summary

Creating the right source citation is a two-step process. First determine the correct provenance of the source, then for each layer of the provenance, enter all details. Enter details exactly as you find them in the source, so do not guess or extrapolate. When using the Centurial browser extension, always check and improve the import results.

Finally, if you want to further improve your source citation skills, you might want to take a look at the two example projects that come with Centurial (Vincent van Gogh and Evidence Explained, available from the QuickStart dialog), both of which contain many more everyday examples of source citations.