Changes I made to my 1. e4 story. (Patreon)

 A couple days ago I posted a webpage where I'll be indexing all of the stories and audio recording I post. You can find that page here. You'll notice that it has a few stories without audio. (without audio YET!) One of these stories, "1. e4"  is from my collection It's Too Late to Say I'm Sorry, but you may notice that there are some differences from the original version. I thought it would be fun to use this space to talk about the edits I made, and why.  (Also, it would be a fun excuse to link to an article in Phrack magazine  haha.)

To read the story:

New version: 1. e4

Original version: 1. e4  on Eyeshot.net

1. e4

This is a story about finding meaning and finding a connection with a stranger through a shared obsession. It's a story that I still like quite a lot, even though I wrote it well over a decade ago. I normally do not update stories or books after they've been published (with some very obvious exceptions) because the temptation is too great to go back and fix the flaws in how I used to understand the world. To censor mistakes I made in thinking or in empathy. Changes like that can undermine a story, though, and can undermine a reader's connection to the story. Once a story goes out there, its imperfections are part of it. 

But when I sat down to reread this story, there were some glaring problems with the cryptography involved. Problems which distracted from the heart of the story, and I believed I could solve the problems with a few short changes - and without compromising how it reads or feels otherwise.

1.e4 is about chess, as well as cryptanalysis , and it was written when my interest in both was much stronger than my understanding of either.  Fortunately, the chess is not discussed in any detail at all, and so I didn't give myself any opportunity to make mistakes. 

Unfortunately the cryptanalysis as discussed in the original version of the story is impossible. 

     Original: 

For a month I spent my mornings playing chess against an unknown opponent through encrypted messages in the personal ads. It started with a simple mono-alphabetic cipher, an opening chess move, repeated day after day until I stumbled upon it. I have no idea how long that message had been repeating, but I found it and I answered in kind, with a move of my own. 

The problem with this, as it reads, is that an opening chess move is just 4 or 5 characters long. 

1. e4, or say 1. Nf3

Solving a mono-alphabetic cipher (like crypto-quotes in the newspaper) relies on frequency analysis. How frequently does a certain character appear, etc. Then there can be a lot of trial and error and contextual guessing involved, depending on how big the encrypted text you are working with is.

An encrypted text of 4 characters long is not solveable. So, I changed it so there was a rambly message of indeterminate length, and then the chess move. 

   Edited version:

For a month I spent my mornings playing chess against an unknown  opponent through encrypted messages in the personal ads. It started with  a simple mono-alphabetic cipher, which decoded into a rambly greeting  and an opening chess move, repeated day after day until I stumbled upon  it. I have no idea how long that message had been repeating, but I found  it and I answered in kind, with a short response and a move of my own.

The rambly message would give the cryptanalyst enough data to solve for every letter, though probably not the digits 0-9 which are essential in chessnotation. But when starting a chess game in algebraic notation , I thought maybe context should be enough to solve for numbers - with some careful play and a few assumptions? My first attempt to solve this problem used that assumption. But it doesn't quite work.

X. eX  - the first unknown here has to be 1. So then we know what 1 is. The 2nd unknown has to be either 3 or 4, the only two possible squares for the e pawn on the first move. 1.e3 is obviously a garbage move for jerks. So, we ASSUME it is 4. The second move will come as X. whatever, giving us a en encoded character 2.  But the problem is that the 2nd player needs to know the numbers on THEIR side of the board before given enough information to solve for them. 1 ... c5 doesn't work unless you know 5. Nor does Nf6. 

Still - the problem of how digits are handled is not a problem the story needs an actual explicit solution for. A solution is possible - the rambly message could be carefully constructed to include digits in a context that makes clear which is which, for instance. And if you are writing secret codes with the INTENTION of someone cracking them, some solution like this isn't just possible, but likely.  So - I don't need to walk through, step by step in the story what the solution was, - that's not the kind of story this is - I just needed to change it from impossible to possible.  Leaving the reader with questions is fine, as long as there are possible answers.

After that, aside from removing a specific reference to DES encryption, the only other change I made was to remove a reference to the main character's loneliness being "since my wife passed away," because despite what I said about not wanting to fix my old way of thinking, this line bothered me. People can be lonely without dead wifes or husbands.  Many of us are.

-Joey