When I was a young warthog, a favorite book was The Secret Of The Old Post Box. In it, some kids find some codes from the Revolutionary War. They’re Book Cyphers, often also called Ottendorf Cyphers. I did some googling, and MANY sites had this same sentence:
The Ottendorf cipher is presumably named after Major Nicholas Dietrich, Baron de Ottendorf who worked for the British, organizing spies in the French and American camps.
I can work in a couple different ways, but all of them involved both ends of the message user having the same piece of text, usually a book. In the movie National Treasure they were in both The Declaration of Independence and the Silence Dogood letters.
The person sending the message writes it out in plain language, then takes the text and finds the words. So for example, in the message “The British are coming”, they look for the word “the”. They would take note of the page number, the line number, and the numbered position of that word in the line.
In a simply text like a poem or letter you might simply do line number and word number. There are lots of options, which is one of the reasons it can be hard to crack.
I’m going to go too much more into them, you are read LOTS online, and I recommend you do.
My own cypher
Sometime in the 2000’s I was using a blog platform I wrote myself, and I made an ottendorf cypher system that used post id and word position. it worked ok, but not great. When I moved to WordPress I ported it, and again it worked ok, but not great. When Gutenberg came along and started putting a bunch of extra code in the post it was almost useless.
Recently I put a bunch of work into it and it seems to work quite well. If you want to try it you can do so here: https://topher1kenobe.com/blog-cypher/
Just put in some plain text and it generates a key pair string. Put it back in and decrypt and you should get your message back.
I’ve released the plugin at https://wordpress.org/plugins/wp-blog-cypher/.
In order for it to work well, all of your words need to exist in the blog. My blog has 23 years of content, so it has a pretty good library.
