« The poster | Main | Timing »
September 23, 2005
Comments
Starting with this entry we will try allowing comments. You will need to post an email address to leave a comment, and we need to approve it before it will appear. Please act responsibly. No personal or inappropriate comments should be made. Don't say anything here you wouldn't say in front of a teacher, or that you wouldn't want said to you; it is a very public place. Criticisms or concerns should be emailed directly to us here, particularly if you think we should not have posted a particular comment. (Though glowing praise will nearly always be posted for all to see.) Hints about the challenges may be blocked or held back at our discretion so don't try to publish hints about the keyword to a cipher ten minutes after we post the Challenge, and definitely don't try to tell the world what the solution is. They will find out how clever you were to crack it in 28 seconds flat when we publish the leader board. We reserve the right to refuse to publish comments without entering into a discussion about why, and to censor comments as we see fit. Finally, though we would like to, we can't be awake 24 hours reading the comments, so please be patient if it takes a while for your comments to be approved.
Posted by Harry at September 23, 2005 05:35 PM
Comments
If you wish to leave a comment please act responsibly. No personal or innapropriate comments should be made. Concerns about a comment should be emailed directly to us here. Hints about the challenges may be blocked or held back at our discretion. We reserve the right to refuse to publish comments without entering into a discussion about why, and to censor comments as we see fit. Please be patient if it takes a while for your comments to be approved.
Is don;t know if it is my browser playing up; but I don't like having to highlight text to read what it says as it blends in nicely with the background.
The cream text goes well on the grey and black backgrounds; the black on the grey background is also fine; but I appear to be having a little bit of a problem with the black text on the near enough black backround
Perhaps I should get my eyes tested, but I really cannot read the text without highlighting it.
I know it has very little if no meaning at all on the challenge; but it could make life easier for people! :)
Posted by: Andy at September 29, 2005 05:09 PM
Sorry if this is off topic, but is there anywhere I can get the challenges from last year? I would preferably like to have a go at the Solitaire Cipher again.
Thanks
Posted by: Eden at September 29, 2005 04:55 PM
Both easy, but what would you expect with the first ones?
Only trouble was submission. :p
XXXXXXXXXXXXXX
Posted by: phoenixashs at September 29, 2005 04:29 PM
Claire, you have until midnight next Wednesday to submit a solution to the first Challenge, but even if you can;t start by then you can still enter at any time during the competition. It would almost certainly be too late to win a main prize after next wednesday, but there are still minor prizes to be won, and the thrill of decoding the messages. We will also publish leader board for each separate stage of the Challenge so you can see how you do on each part you enter. Hope you can get a team together. Did you not get a poster? All the best, Harry.
Posted by: Harry at September 29, 2005 09:14 AM
Hi. I am a teacher at a school and have only come across this competition and wondered if it was too late for a group of students to join in? I would have to advertise it to them so it would be a week before we could perhaps be organised to get going.
Claire
Posted by: Claire at September 29, 2005 09:01 AM
Yes, Vicky, th eprizes for Challenge A do not depend on fast you are.
Posted by: Harry at September 29, 2005 08:21 AM
Maybe they could do a sound file with the dot and dashes as well, so that it can be accessed normally as well as with the sound. That way nobody loses.
As for the timing, I like it how it is. Our merry band of codebreakers leaves school at 3.45, but we can stay after school until 5.00. If it was made later we would all miss the start and not have as much time to work on it.
Posted by: Josh B at September 28, 2005 01:52 PM
Jason, I did mean for just monoalphabetic, hopefully no big confusion caused. I suppose that in the later stages, with no image recognition programs, to /.-.././.-/.-./-./--/---/.-./..././-.-./---/-.././ is the only plausible option.
If you were to make all of the dots and dashes the same pixelwise, rather than smoothing the image, then image recognition would be a lot easier!
Sound files are too risky in my opinion, our school has no speaker facilities, and some people working at home may have that problem as well. Also you would need to ensure that people would be able to play it using whatever software they have (I'm not sure of Mac sound players). If you could get it to work, it'd certainly be more interesting though!
Posted by: Ian at September 27, 2005 08:30 PM
Will the ciphers that are used reflect the time period the challenge is set? In other words, will there be ciphers that were invented after the challenge is set? (Less realistic, but tougher challenges, which I prefer)
Posted by: Eden at September 27, 2005 07:57 PM
How do you go about writing image recognition programs? I've never written anything like this before.
Posted by: bob at September 27, 2005 06:42 PM
I'd agree with Dave - coding a morse code recogniser seemed easier going than dealing with a load of glyphs. Especially as you know what form the symbols will take beforehand :)
Posted by: Robert at September 27, 2005 06:15 PM
I don't like what Ian said in the last comment: "submit the possibly flawed result, and then remove errors." This does of course work with monoalphabetic ciphers (where each letter is replaced by the same different letter, position doesn't matter) but with more complex substitution ciphers and even the most simple transposition ciphers, the end result will be partially or even completely destroyed after decryption with the key. But then, of course there will be no end result for us since we don't have the key, and so deciphering the text would be impossible.
Now, image recognition. I didn't think about that. How do you go about writing an image recognition program anyone? My solution was to .-.././.-/.-./-.//--/---/.-./..././/-.-./---/-../. , print out the cipher text image and write the letters straight into the computer. Then you can decipher and submit. But if we get the text anyway...
Now, a sound file. Wouldn't that be a good idea! This wouldn't be too time consuming if you know morse code (and the message is short). You should be able to write down every letter in one continuous run then. Or of course some kind of morse code recognition! If sound doesn't feature in the challenge then why not put it somewhere on this silent website!!!
About the time problem. It is too early for me. My chances of winning hang on the later challenges where inability to decipher the text quickly shades out the speed of getting the cipher text from the internet and out of images. And the speed of decrypting ciphertext hangs on knowledge/understanding of the ciphers and computer skills i.e. automating tasks and instant decryption.
Well, this comment is probably too long but it is valuable info!
Posted by: Jason at September 27, 2005 05:44 PM
David, I found it quite a bit easier to write an image recognition program for the morse code than for the glyphs of a couple of years ago, since there are only a handful of different symbols to recognise. But maybe I did it in a different way to you?
Posted by: Dave at September 27, 2005 03:37 PM
Good point Ed. Beginning with Challenge 1 you will find that the small text in the gif on the left of the page acts as a link to either a text version of the page or a more scalable graphic. Just place your mouse over the image and the text will trun red to show you are on a link, then click to go to the text page or download the graphic. Hope this helps.
Posted by: Harry at September 27, 2005 02:22 PM
i think that the morse code is to small and it hurts your eye theey should make it bigger so you can copy and paste
Posted by: ED BALLINGER at September 27, 2005 02:15 PM
I couldn't possibly comment about what might or might not be in the Challenge. Just wait and see.
Posted by: Harry at September 27, 2005 11:26 AM
Harry, you're not going to do anything like only giving us a sound file for a message in morse code are you? That might make it realistic but would be time consuming to deal with.
Posted by: David at September 27, 2005 11:10 AM
Hi David. Those of you who have been taking part in the competition for a while might guess that this David is twice winner of the competition, but is now too old to take part. Good to hear from you.
Posted by: Harry at September 27, 2005 10:32 AM
Although image recognition programs are a pain to write and test it is so cool when you get the cipher text in a couple of minutes compared to typing out 2320 characters by hand. Morse code recognition looks slightly harder than glyph recognition since a dot (if it is small) could easily be mistaken for a mark on the paper based on the way I was doing it for glyphs.
Posted by: David at September 27, 2005 10:25 AM
Feng, we can deal with this here. If there is a Playfair cipher (and there might or might not be one) you should leave the padding characters in, if only to make it easier for you. We treat them like punctuation and remove them before processing your submission.
Posted by: Harry at September 26, 2005 10:16 PM
In regards with the correct solution thing again. There was a problem last year with the Playfair with the additional "x"s one puts for a doubled letter. What would be the thing to do this year if the Playfair does come up again? Does the letters need to be removed from the solution? Or kept in?
Posted by: Feng at September 26, 2005 10:06 PM
Actually, I've quite enjoyed the picture file. It wouldn't be a challenge if it wasn't difficult. It will certainly be a test of accuracy and patience. However, there is always the problem of making the challenge biased towards those who can afford the best image recognition software. On the other hand it may even the playing field as most ocr make some mistakes so there will be an increased element of manual work?
Posted by: AC at September 26, 2005 09:48 PM
Ian, maybe you should try using image recognition software software then! Don't worry, you certainly won't need it at the beginning. Harry
Posted by: Harry at September 26, 2005 09:29 PM
Ah, but I expect GCHQ would have image recognition :P If they do use images at all!
As for the timing, it's quite good for me.
Typos are a pain with the image files, but as long as you're fairly accurate, you can use whatever program you have anyway, submit the possibly flawed result, and then remove errors.
Posted by: Ian at September 26, 2005 09:24 PM
So you guys think that GCHQ phones up their targets and says "I say old chap, its a terrible bore having to enter all this data ourselves, do you think you could send us a textfile with your secret message in it. We'll be able to crack it so much faster that way." ?
Posted by: Harry at September 26, 2005 07:39 PM
I agree with Ian, Image files are horrible. And they're just a test of typing ability, not coding. One mistake and you'd kill yourself later.
Posted by: Michael Bryant at September 26, 2005 06:52 PM
Simon, I'm afraid not. Since we only use the order in which we receive solutions, and not the absolute speed it doesn't effect the score and we don't publish this info. Harry
Posted by: Harry at September 26, 2005 04:18 PM
Harry
I was just wondering, would it be possible to publish how long it took the 10 fastest teams to submit a 100% correct entry? I'm just interested to see how fast people can crack the codes.
Posted by: Simon Chu at September 26, 2005 04:14 PM
See, this is exactly my point. If we move it earlier it gives an even greater advantage to those who do get access straight after school. If we move it later it increases the advantage to those with home access. I won't post any more comments on this topic, but promise to take all the feedback into account when designing next year's challenge.
Posted by: Harry at September 26, 2005 02:30 PM
To add to the time debate! My school finishes at 3:15, so I feel that 4pm is too late! I am trying to run an after school club, and it would be great if teams could get together on thursdays and work through the ciphers. Some students will have to wait until friday lunchtime before they see the ciphers as they do not have e-access at home.
Posted by: Chris Smy at September 26, 2005 02:26 PM
Chris, thanks for your comment. This is an issue that comes up every year and the answer is that there is no good time. If we post the Challenges an hour later only those who have internet access at home can take part, so we time it so that participants can use school facilities. We know not everyone gets access after school so this is not a perfect solution.
Posted by: Harry at September 25, 2005 09:43 PM
Why can't the ciphers be posted a bit later than 4 o clock, i get home at about quarter to five after school so i lose 3 quarters of an hours time on most other people throguh no fault of my own and i know lots of other people in the same position.
Wouldn;t it be a lot easier for everybody to have the ciphers released at say 5 o clock instead so more people have an equal footing to start the cipher
Posted by: Chris at September 25, 2005 09:03 PM
Thanks for all the nice comments, Harry dV, but we can't make any promises about image files. Why don't you wait and see, ...
Harry
Posted by: Harry at September 25, 2005 04:57 PM
Tom, we always use the earliest entry when you submit two or more entries with the same "correctness" score. All the best,
Posted by: Harry at September 25, 2005 04:55 PM
Yes, I think this year is going to better than all the others. The website's way better too, well done!
I also would like to echo Ian's hope that there won't be any more image files, 'cos they take ages to work with...and time is of the essence!
Finally, good luck!
Posted by: Harry de Voil at September 25, 2005 03:26 PM
I think someone asked this last year.
But, if you submit a 100% correct solution, but then someone else in your team or something, 10 minutes later, submits the same solution again, will your position be worked out for the first correct submission you made or the second?
Posted by: tom at September 25, 2005 02:36 PM
Tom, excellent question. Our cipher checker deletes all the other characters in your solution (including spaces, numerals, punctuation), does the same to our model solution, and then compares the results. So yes you can ignore numerals if they appear, however you are more likely to make mistakes if you try to pre-empt this by removing the characters yourself, so I would leave them in. This is not to say we will include numerals by the way, just that we reserve the right to do so.
Posted by: Harry at September 23, 2005 11:34 PM
In the rules it says:
"A solution will only be deemed to be correct if, disregarding the punctuation [and spacing], the deciphered plaintext (only involving the Roman characters A to Z [UPPER or lower case is fine]) is letter perfect as compared to our master solution."
Does this mean we can ignore any numbers that appear in a message?
Like last year the transposition cipher had numbers in it, but can they be ignored in this years?
Posted by: tom at September 23, 2005 11:28 PM
Wow! This cipher challenge is looking to be easily the best so far...
*sigh* Image files :P Can we expect more of these? Or will the Morse (if any) be in text (. and - and /) format?
Posted by: Ian at September 23, 2005 07:30 PM








