Home Development Blog A Look at the Competition
A Look at the Competition PDF Print E-mail
User Rating: / 0
Tuesday, 28 September 2010 11:49

After putting my initial concept down on paper, the next sensible thing to do is to have a look at what else is already there.   In this day and age it is rather naive to suppose that any idea is genuinely new and/or that the competition is remote enough to ignore.

Sure enough, I found a few sites doing more or less exactly what I had proposed.   Including one with simple monoalphabetic substitutions, an interactive client, a large database of puzzles and tools to anayse the results.   Another offered daily puzzles on Android phones.   I'm sure there is an iPhone equivalent.   Hmmm!   What now?


Well, it occurred to me that the average time in which these puzzles were being solved on this other website were rather short (50 seconds, 85 seconds, etc.) and I wondered whether there really were people that fast or tools that allowed them to get the puzzle solved in that time.   I found out that I barely had time to type the puzzle text into a third party english langauge puzzle solver (WinDecrypto) and type the result back into the website within the times that the website was quoting for its' other users!   Possible though...   Also, the website allowed one to reveal all the vowels in the cryptogram and when one did that, I too usually managed to get the rest of the solution within the quoted timeframes.   So, that was obviously what most of the users of this website were doing.   (Here I arrogantly assume that they are not all immensely cleverer than I am...)   The site also gave sources for all their quotes, so a search via a dictionary of quotations was also a valid option for solving the puzzles.


Also, talking to my Dad - a committed Sudoku fan - revealed to me that many people actually prefer a bit more substance to their puzzles.   He quoted the half hour time span of an average daily commute as typical.   So maybe the puzzles should be a bit more complex?

I decided that I would seek my competitive edge by:

  • Offering a range of difficulties in the puzzles.   At the simplest end, monoalphabetic substitutions.   Moving up the scale to as complex a set of hand-ciphers that I could implement in the tools.   So, firstly I would move to non-linear mapping of letters (to mess up simple statistical analysis) and then on to polyaphabetic ciphers, and then ...   I did not want to get into machine ciphers - that would lead to a copmletely different league which might be fun to implement some day, but not yet.
  • There is a lot of background information that makes a puzzle easier to solve.   I planned to make this available, but to also make it clear what the user had accessed when submitting a solution to the league tables.   The simplest help would be to show the white space in the puzzle, then perhaps the punctuation and then perhaps ding the frequency analysis.   Access to this information might also be time-locked, so that e..g white space is only available half an hour after the puzzle is first released to the web?
  • Bot protection, or preventing cheating access to the plain text means that the client program should not have access to the solution directly, but should submit its' solution using a one way hash, e.g. SHA.
  • Since the puzzles were to be more complex, I should allow users to save their partially solved puzzles so that they can come back to them later.

I decided to go ahead with the project and started on some brainstorming which I will talk about in my next blog installment.

Copyright © 2019 Crypto League. All Rights Reserved.
Joomla! is Free Software released under the GNU/GPL License.