French software engineer may have cracked the Zodiac killer's code

boing
1

Originally published at: French software engineer may have cracked the Zodiac killer's code | Boing Boing

…

5 Likes
2
11 Likes
3

French software engineer may have cracked the Zodiac killer’s co

Now onto that whole French 257 different types of cheeses, we must have an answer to this!

4 Likes
4

To go with the wines.

5 Likes
5

tenor-9

7 Likes
6

I guess in six or seven months, someone else will claim to have cracked the codes, yet again. (Or are there any left undeciphered now, assuming we accept this latest claim?)

5 Likes
7

I’m not sure why the Times is covering this. I’m no Zodiac expert, but this guy does seem like run-of-the-mill Zodiac crank.

My favorite part was when he posted the solution for cipher Z13 as “KAYE” even though the solution he found was “KAYR” because it “resembles” the name of suspect Kaye. So he based his solution on prior knowledge of a lead in the case, was still 25% wrong, and he lied about it. Some nice code-solving there, buddy.

I also think the whole “magnetic coordinates” stuff might be ridiculous on its face; I doubt that in 1969 the Zodiac killer would have had access the precise-to-three-decimal-places coordinates (in a system different than “the more familiar geographic coordinates”) for a high school.

9 Likes
8

The code broken last year was the last cipher that experts considered long enough to be accessible to cryptography. This article is reporting on someone who has claimed to have deciphered the final two ciphers, which are generally considered too short to decode–there’s just not enough information there.

8 Likes
9

I’m not an expert either, but the solution sound convincing to me given other known facts. What I’d love to know are other solutions to know if there are a large set of similarly plausible solutions.

3 Likes
10

I think the main issue with his solution is that it was based on those other known facts. He was working from prior knowledge. There have been hundreds, maybe thousands, of solutions proposed to these two ciphers. There’s just not enough ciphertext to do any meaningful work.

3 Likes
11

Hey, if it keeps the Zodiac Killer from getting re-elected to the Senate, I’m not complaining.

21 Likes
12

But it’s “My name is Cruz.”

9 Likes
14

TIL that this is a thing :confused: (He even admits to it!)

15

I like that the Zodiac Code Breaking Forum has a rule against posting solutions.

9 Likes
16

There’s probably close to that if not more in terms of italian cheeses, and I think probably in most european countries, even the UK used to have loads of regional and local specialties, most were lost to the industrial revolution/WW1…

2 Likes
17

It is a rule against posting solutions to ciphers that are impossible to decipher (or, I guess, solutions that cannot be proven to be accurate). It’s like the rule against trying to patent perpetual motion machine ideas - best not to encourage those people.

4 Likes
18

“Drink your Ovaltine.” Yeah, I know it’s obvious.

4 Likes
19

One of the cryptographers interviewed in the story says there are hundreds of theories about those shorter notes, of roughly equal rigor to this one, floating about. Indeed, this guy seems to be a garden-variety Zodiac crank, and not an unprecedented cryptographic genius.

1 Like
20

As a computer scientist myself, one of the problems with the field is that we learn a whole lot about a subject, get requirements from end users, design a system that fits their needs, then support them while they use it. And that gives us a familiarity with the subject with very little work- we don’t actually become experts in the field, we just borrow expertise from the people whom we support.

But it gets into your skin. You feel like an expert. You begin to think of yourself as an expert. You start seeing these different fields of study, apply the computer science methodology to it, and think you’ve found a solution that has been eluding people for years. And occasionally, we do. But for the most part… we don’t. We just find a good way of approximating a good enough result or we don’t know what we don’t know that makes it invalid.

So, always beware when a computer expert finds a solution in an unrelated field.

(Technically crypto isn’t a completely unrelated field; but no one described him as a crypto expert, so I’m guessing it’s not his specialty…)

3 Likes
21

Anyone ever given thought to it just being gibberish from a madman?

Or deliberate nonsense meant to tie people up for decades as it has done?

1 Like