hacker.org

how did you solve the challenge ?

I don't think that its solvable manually like I tried first time ...

did you wrote own scripts ?

if yes in which language ? and could you post it eventually ? i would be really interested in how to solve such a program

I use to do cryptoquotes in newspapers all the time. I just spent some time guessing out the letters starting with guessing that 3 letter words were "the" and going from there and so on.

I did write a little program to make the substitution easy so I didn't make any mistakes but mostly just guess work.

Yes, this was pretty easy to solve, I used a subsitution website to assist me, but you don't really need things like that.
As MerickOWA pointed out, scanning the text for probable "the's" - which has two of the most common letters in English and the first word with the duplicate "L" was all you had to do, before many other words where easily solvable.
Google helped me with the last missing letter ('m') - since "adumbrate" doesn't belong to my vocabulary.

I think it's a bit unfair when there's a letter that's only used in the solution word, and the solution word isn't a real word. Edit: I see, this is a real word, my bad. /usr/share/dict/words is a good source of greppage, if you're looking for real words matching a particular pattern.

Generally for substitution ciphers I use external Unix tools, executed on portions of a text file using vim - e.g. "tr", "sort", "uniq" - combined with sed-style manipulations which are built-in with vim.

Although it's crude, I like the way it's rather ad-hoc - it makes it easier to adapt to different techniques (e.g. analysing pair frequencies, analysing transposition ciphers).

For more advanced things I write Python programs, which I can again run as external filters over the text. But on one occasion I made an interactive Python program to analyse a cipher, because it was useful for the program to maintain state.

I did this manually without any programs. Searched for common words, letter patterns, etc.

Yeah, I did it manually too. There are some good online substitute decoders - I would try a letter and see what words would make sense. Kind of like playing 'hangman'. Then there were 2 letters left and only one made a word recognized by google.

i made it with cryptool, great tool

I've used this website for help (and also Simon Singh's helper, which is kind of similar but with more features; it's also quite buggy, at least under Opera).

Going to have a look at cryptool.

I counted the letters and then compared them to letter frequencies.

ape wrote:i made it with cryptool, great tool

Hehe, me too
At the time I attempted the challenge I was sitting at a windoze machine, CrypTool was installed, so I thought why not give it a try? It seems to be a nice tool where you can learn a lot about cryptography.
I hope it will be released for linux soon.

Reading the page, it looks like there's a Java version.

I did it also manually. Counted the letters, then found zvm, hv and hdy twice. hdy together with the frequency was very likely "the".

and so on...

Simply used gedit and search + replace.
Most challenges contain the words 'answer' or 'solution' in their ... well ... solution.
This one had both.

I took the time and searched for "answer" (quite easy to find it because there is only one 6 letter word) and did all the guessing in 2 text fields... all by hand^^

The substitution rule is rather wierd....

S?WHPXMTA?DFUGICB?LNVORKEY
^^^^^^^^^^^^^^^^^^^^^^^^^^
ABCDEFGHIJKLMNOPQRSTUVWXYZ

wtf on earth is this ???