"In het verleden behaalde resultaten bieden geen garanties voor de toekomst"
About this blog

These are the ramblings of Matthijs Kooijman, concerning the software he hacks on, hobbies he has and occasionally his personal life.

Most content on this site is licensed under the WTFPL, version 2 (details).

Questions? Praise? Blame? Feel free to contact me.

My old blog (pre-2006) is also still available.

Sun Mon Tue Wed Thu Fri Sat
Powered by Blosxom &Perl onion
(With plugins: config, extensionless, hide, tagging, Markdown, macros, breadcrumbs, calendar, directorybrowse, entries_index, feedback, flavourdir, include, interpolate_fancy, listplugins, menu, pagetype, preview, seemore, storynum, storytitle, writeback_recent, moreentries)
Valid XHTML 1.0 Strict & CSS
Day 2: Broken technicalities

(Disclaimer: This post is about a game. It is all fiction.)


The monday night meeting was an hour later than Sunday's. Fortunate, since it gave me an extra hour of time to sleep between dinner and the meeting. So arriving there fresh and (a little more) rested, we would receive the next puzzle... and the consequences for the lemon juice afair.

So, after an explanation of Sunday's puzzles, we were called forward. Frank explained the mistake we made and declared our punishment: From now on, we would no longer be team E.V.I.L., but team L.I.E.V. ("cute"). So much for our image. Fun :-).

For another report on the past two days, see Brenda's weblog.



Our assignment today was simple: The mainframe we found yesterday has an admin interface. We could use our own logins here, but not without a valid pincode. To help us find a pincode, we received an envelope with a clue: A paper with 6 Kakuro puzzles. After making a number of photocopies we went home to solve them. After first starting out with one puzzle each, we quickly realized that this would take too long. As soon as we found out that these damn puzzles were called "Kakuro" puzzles (thanks Diederik!), two people started search for kakuro solvers on the internet. In the end Brenda found a number of solvers, from which the first worked (Which proves you need a woman to actually find stuff...).

By the time we had entered a few puzzles into the program, Frank solved one by hand. The others were solved by my notebook. On each puzzle, a number of squares were marked in color. Using the numbers in these squares, we got six 6 or 7 digit numbers. Also, we got the digit "3", which was in a special square that had double borders. This turned out to be no clue, the double border was accidental...

What do do with numbers?

Our first idea was of course to use these numbers as a pincode, though since we assumed to be looking for one single pincode we didn't expect this to work. It didn't. Later this appeared to be caused by some kind of technical problem, but for the moment we started trying other stuff.

Summing the numbers, calculating differences, using modulo 26 letter conversion, offsetting with the magic number "3" we found, nothing appeared to give anything useful. One of the numbers, "4891886" IIRC appeared to be a telephone number (all on-campus telephone numbers start with 489). After I quick google without results I decided to just pick up the phone: Worst case I would piss someone off for waking them at 0200, but I could always just hang up in that case. No such luck: The number didn't exist...

Problems with the cool stuff

After a while it became apparent that the organisation was having some trouble with the technical side of the puzzle. We retried entering one of the numbers and it accepted it this time, showing me an empty map with an empty legend. I remembered that this same application caused problems on my laptop the day before (There is no macromedia flash player for AMD64 and the open source player I use doesn't work fully with this app). So, moving to another laptop (we had 4 up and running), we tried again.

This time (using a second pincode, since the first stopped working), we got the legend, but there was nothing on the map. We tried all 6 pincodes to see if one might provide access to the "Scanpoint" dot on the map, but it didn't.

Time to complain. After some "hmm"'s and "I dont' understand"'s from the technical dude, fizzgig, they suspected someone was bruteforcing pincodes and was causing trouble. But, after that problem was fixed, the dots still wouldn't show up. Apparently, the problem was one of scaling: One client worked, more somehow killed the connection between the (flash app in) the webpage and the server.

So, instead of a position marker for the "scanstation" and a moving marker for the guard's location (fed by WLAN positioning, very cool) we got an empty map.

Syrinx, the mainframe with humour

To prevent everyone getting stuck with the puzzle, the organisation quickly deployed their backup mainframe: Syrinx. Through a state-of-the-art IRC interface, we were able to enter pincodes and retrieve the needed information.

This lead us to a clue at the pond behind the Spiegel building. After a big shootout there (we weren't the only ones solving this) we found another puzzle: a large sequence of (29) increasing numbers. Without any indication about what to do with them, we resorted to the default solution in these cases: Try to find a pattern in the numbers given.

Most obvious was of course taking the differences, which gave an (again increasing) sequence. Taking the second order difference showed that these were mostly "1", with a few exception. There seemed to be some imperfect symmetry in second order derivative. Also, every run of "1"'s in the second order derivative of course corresponded with a run of adjacent numbers in the first order derivative. Not a coincidence, but what would it mean?

Do the math

After a number of attempts at establishing patterns in this data (using spreadsheats and Matlab for plotting) we decided to buy our first hint. This hint read "Try plotting a graph of the difference". Right, we had done that way before, but it didn't show anything useful. Now knowing this was indeed the direction to focus on, we did some more thinking. Looking at the difference sequence showed that all were 2 digit numbers. Somebody suggested using the first digit as an x coord and the second as a y coord in a small graph.

Trying this gave us a number of points which, when connected properly, formed the number "381", which is a flat on the Witbreukseweg. Hoping to find the final clue there, four of my teammembers went to look there. Initially not finding anything, I reminded them to check the "381-Laag" part of the flat, which is a little way off the main part.

There, we found the final clue. Quickly entering it, we got the satisfying message "De authentication-codes zijn geaccepteerd. Je kunt nu gaan slapen" (The authentication codes were accepted. You can go to sleep now). We did.

0 comments -:- permalink -:- 19:50
Copyright by Matthijs Kooijman