British Succession Project

Welcome to the home page for the British Succession project!

Aim and Scope

The aim of this project is to track the timeline of the line of succession to the British Throne through automated (programmatic) means. The current scope is to track all descendants of the Electress Sophia of Hanover. These descendants, according to the Act of Settlement 1701, are the rightful heirs to the throne. The existing monarch and line of succession is therefore applicable from the reign of George I (1 August 1714).

Licensing

The code for this project is (and always will be) open source and licensed under GPLv3.

Implementation Notes

Please do not raise issues with this project without acknowledging the following limitations/caveats.

• This is not a genealogy project! We rely on well-established external sources for ancestry/descendant data. There are no plans to begin curating our own data of this kind. Therefore the accuracy of this project is entirely dependent on the accuracy of such public sources.
  • Inaccuracies in these data sources should be reported/contributed to the curators of those data sources, not here.
  • We do however curate a list of illegitimate persons (and dates); contributions to this list are welcome.
• The change from male to absolute primogeniture preference due to the Perth Agreement (for persons born after 28 October 2011 and effective 26 March 2015) is taken into account.
• No attempt has yet been made to tag Roman Catholic persons as illegitimate to the line of succession. This could be made to fit within the existing framework with minor amendments. Special consideration would have to be taken with respect to descendants of such persons, as well as the Perth Agreement (which ended disqualification of spouses of Roman Catholics).
• The abdication of King Edward VIII on 11 Dec 1939 is treated as becoming illegitimate on this date.

Data Sources

The requirements of a genealogy source are as follows.

• Well curated (especially for errors and duplication).
• Updated for new birth/death events in a timely manner.
• Able to be programmatically accessed via an API (or at least downloaded and parsed).
• Provide a sufficient amount of information publicly (it is common for platforms to have private profiles).
• Provide a license to gather and display data with appropriate acknowledgement.
• Ideally handle duplicates internally in an attempt to form a single "source of truth" tree.
• Persons/profiles should minimally contain the following information.
  • Name;
  • Gender;
  • Birth date (and accuracy);
  • Death date (and accuracy);
  • Parent links;
  • Children links.

An incomplete list of sources which meet some of the above criteria are as follows, in order of preference.

• Geni: It is the most complete World family tree available online. The Geni API is based on primary data types of Profiles and Unions. The public API is rate limited to about 1 request per second, which does bottleneck full updates. A large proportion of living profiles of interest are private (e.g. the children of Zara Tindall). The database has been found to be full of spurious and/or unverifiable relations (see the illegitimates file).
• WikiTree: Although not as extensive as Geni, WikiTree appears better curated and the WikiTree API is extremely easy to use fast and at scale. Unfortunately the profiles of living British Royals are set to private (see e.g. descendants of George VI). This makes it essentially unusable for this project.
• MyHeritage Family Graph API: Due to the extra complication introduced by having separate family trees, this API has not yet been explored as an option, although it does appear very extensive. The MyHeritage website is a paid service.
• The Peerage: A personal project of Darryl Lundy tracking British peerage and European nobility, curated since 1988 and still regularly updated. The information section also contains links to several other large databases.
• Geneanet: Some detailed trees appear to be available with no official API. For example:
  • Tim Dowling Family Tree;
  • Simon de Solomy de Palmas Family Tree.
• Static GEDCOM files: These exist scattered around the internet. For example:
  • Royal92.ged;
  • Queen_Eliz_II.ged.
• Persons eligible to succeed to the British Throne as of 1 Jan 2011: A monumental snapshot of the line of succession (up to 5753 people) as of 2011. Unfortunately this does not allow for tracking throughout history.

The problem of combining trees from multiple data sources is not simple and has not been attempted. The project currently uses Geni.com as the sole data source. This may change in the future.

Python Development

The scripts are currently written and tested using Python v3.8.5 (however should work down to v3.6.x) which must first be installed. To set up the virtual environment with all required packages installed, run the following in Linux (similar commands exist for Windows).

python -m virtualenv venv
source ./venv/bin/activate
pip install -r requirements.txt

Data Gathering

The data gathering step will gather all relatives and descendants of the Electress Sophia (profile-56847813). This may take an hour or so due to the rate limit of the Geni API. Some parallelisation is implemented via threaded workers. The rate limit must be respected or your IP will get blocked by Geni's DDOS protection (watch for 429 response).

python geni.py --seed "profile-56847813" --db db.json --workers 6

This script will generate two files. The db.json file contains the raw profile and union responses from Geni (expect it to be up to 30 MB in size). This file is then converted to output a simplified row format (geni.yml) which is compatible with the main script to calculate the line of succession. The minimal format of entries in this file is as follows (additional fields may be stored for convenience, e.g. for identifying illegitimates).

- _id: 684074c6-e3bc-5bb2-9948-476a780db75b
  name: Jacob Pleydell-Bouverie, 8th Earl of Radnor
  short_name: Jacob Pleydell-Bouverie, 8th Earl of Radnor
  gender: male
  birth_date: '1927-11-10'
  birth_accuracy: day
  is_alive: false
  death_date: '2008-08-11'
  death_accuracy: day
  parent_ids:
  - aeccd4c9-5377-5ead-90e7-79b5739b889d
  - 07c8a85e-2afc-50f9-b897-24467eab56c7
  children_ids:
  - beb53e66-0a8b-5179-8254-2d710d190ac0
  - 29de1077-f901-55b4-8156-3e08c622e3f5
  - cbd02cdd-a6a5-5dc0-8e86-01cfb925e4e7
  - ead42fd9-cdfa-599e-8bfb-806cc071f793
  - 9129b7b5-b590-5b4f-b6de-1139c696b869
  - c153a3b7-8c24-5997-b425-ca0dd88ebbe1
  external_url: https://www.geni.com/people/Jacob-Pleydell-Bouverie-8th-Earl-of-Radnor/6000000009607153132

To run the script again for a complete data gathering step, you will need to delete/rename/change the database file location (otherwise only the conversion step will run). This is done to allow the conversion to run independently of raw data gathering.

Illegitimate Persons

A list of illegitimate persons is maintained in the illegitimates.yml file as an input to the succession calculation. The format is as below. Persons with a key/value pair in the match list will be tagged with the date given (or their birth_date if this is null). Citations should be provided as comments where possible.

# Abdication of Edward VIII
- date: '1936-12-11'
  match:
    _wt_id: 4928583
    _geni_id: profile-3413739
# Children of William IV and Dorothea Jordan
# George FitzClarence
- date: null
  match:
    _wt_id: 7903372
    _geni_id: profile-20732210

Succession Calculation

The line of succession calculation will use the files in the format above as inputs to calculate and output descendents in order of the line of succession. The script can be run as follows.

python main.py --descendants geni.yml --seed "0557aac6-264c-5a83-8f1e-a3f6cfac8b9a"

The output is successors.csv and successors.json with the following columns/fields.

• _id: An ID unique to the person. Must be non-null.
• name: The name of the person. Should be non-null.
• birth_date: The best known birth date. Must be non-null.
• death_date: The best known death date. May be null; if so then the person is assumed to be alive.
• illegitimate_date: The date from which the person became illegitimate. May be null.
  • Equal to the birth date for persons illegitimate from birth.
  • Equal to the birth date of the earliest illegitimate ancestor if a descendant of an illegitimate ancestor. If the person has a legitimate bloodline then this will take precedence.
  • Equal to the abdication date for Edward VIII.
  • Equal to the effective Perth Agreement date (2015-03-26) for persons born between 2011-10-28 and 2015-03-26 whose succession order is effected. This results in two entries for the same ID; the second entry has a legitimate_date of 2015-03-26.
  • In principle can be used for Roman Catholic persons and spouses including changes due to the Perth Agreement, however this behaviour has not yet been implemented.
• legitimate_date: The date from which the person becomes legitimate. May be null.
  • This is currently used for handling the Perth Agreement as above.
• external_url: Link to person on external data source website. Should be non-null.

The order of the output is the line of succession order if everyone were still alive today. The line of succession at any given date can then be determined from this list by a simple algorithm.

1. Filter out not yet born;
2. Filter out dead;
3. Filter out illegitimate;
4. Filter out not yet legitimate.

The ordered list of remaining persons is the line of succession.

Web Development

The source for the web application is contained within the web directory. Development and testing proceeded using node v10.19.0, which must be installed to set up the development environment. Once installed, navigate to the directory and run the following to install dependencies.

npm install

If adding dependencies please run npm install <package> --save to ensure they appear in the package.json. Currently the only dependencies are d3 for the chart, and lit-element to simplify data-binding between JavaScript and the DOM.

The latest successors.json file should be placed in the web/static directory. This can be downloaded directly from british-succession.co.uk/static/successors.json to shortcut the Python script.

Run the automated build watch server.

npm run watch

Run the development static server.

npm run serve

Navigate to http://127.0.0.1:9000 in a web browser.

The web application is made up of a chart and table component. Each path in the chart corresponds to a person. The path height ordering is the line of succession at that point in time. Filled circles represent birth/death. Unfilled circles represent legitimate/illegitimate dates (e.g. the abdication of Edward VIII). By default the chart/table is set to the last updated date. Click the chart to select a different date. Hover over the paths to see the corresponding highlighted row in the table. If the person is not in the line of succession at the selected date (e.g. not yet born or already died) they will appear without a number. Hover over the table to see the highlighted path in the chart. Zoom/pan the chart to see more detail. Add more lines to the chart using the add icon (initially limited to 500 for load time and performance reasons). Reset to the initial view using the reset icon.

Web Server

The production website can be reached at british-succession.co.uk. The host is currently an Ubuntu 18.04 VM with public IP 178.62.42.196. For traceability (and reproducibility) the set up of this VM is described here.

1. The master branch of the Git repository was cloned.

   git clone https://github.com/jdclarke5/british-succession.git

2. Python 3.8 was installed and the virtual environment was set up.

   sudo add-apt-repository ppa:deadsnakes/ppa
   sudo apt install python3.8
   sudo apt-get install python3-distutils
   curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py
   python3.8 get-pip.py
   cd ~/british-succession
   python3.8 -m pip install virtualenv
   python3.8 -m virtualenv venv
   source ./venv/bin/activate
   pip install -r requirements.txt
   deactivate

3. Node 10 was installed.

   curl -sL https://deb.nodesource.com/setup_10.x | sudo -E bash -
   sudo apt-get install -y nodejs

4. The website was then built.

   cd british-succession/web
   npm install
   npm run build

5. The initial successors output was transferred manually from a local version.

   scp successors.json [email protected]:~/british-succession/web/static

6. Nginx was then installed and set up for a static site.

   sudo apt install nginx
   sudo ufw allow 'Nginx HTTP'
   sudo ufw allow 'Nginx HTTPS'
   sudo ufw enable
   sudo mkdir -p /var/www/british-succession/html
   sudo chmod -R 755 /var/www/british-succession
   cp -a ~/british-succession/web/build/. /var/www/british-succession/html/
   # Enter VIM to create below nginx configuration
   sudo vim /etc/nginx/sites-available/british-succession
   sudo ln -s /etc/nginx/sites-available/british-succession /etc/nginx/sites-enabled/
   sudo nginx -t

   The Nginx configuration is as follows.

   server {
       listen 80;
       listen [::]:80;
   
       root /var/www/british-succession/html;
       index index.html;
   
       server_name british-succession.co.uk www.british-succession.co.uk;
   
       location / {
           try_files $uri $uri/ =404;
       }
   }

7. An SSL certificate was installed for the domain names, with the option to forward HTTP requests to HTTPS requests.

   sudo add-apt-repository ppa:certbot/certbot
   sudo apt install python-certbot-nginx
   sudo certbot --nginx -d british-succession.co.uk -d www.british-succession.co.uk

8. A cronjob was added to automate an update every Sunday at midnight UTC time by adding the following line to crontab -e.

   0 0 * * SUN bash ~/british-succession/update.sh

9. Later, A cronjob was added to ensure that the certificate was renewed and NGINX restarted, even though this should be done by #7!

   0 0 * * SUN certbot renew && service nginx restart

Contributions

If you want to contribute to this project please do so in the following ways.

• Send me an email at [my-GitHub-username]@gmail.com.
• Open a GitHub issue with a bug or feature request.
• Create a pull request to the dev branch with suggested code changes/improvements.

I am not a genealogist or even a British Royal Family enthusiast, just a physicist and data/software engineer. My motivation for this project is from the perspective of gathering, cleaning, and presenting the data in a new and interactive way. I welcome contributions from others to improve on what is built from here!