Author
Message
Lysithea
Unfettered
Joined: 04 Feb 2004 Posts: 468 Location: Australia
What I have so far:
Spoiler (Rollover to View):
t CONGR HCDSUGRXN FLBDH HINPOMP YLALMLB
g PBATEUPQFHT EKASYO QUUV AC BZCL YNYZYO
c TFEXIYTU JLXIO EWCSU YYZE GFDGPCRCD CS
a VHGZKA VWL NZKQGYEUWAA BGIHFIRETE FEU
t HACNCQNPPHUT NFAIRFVCLDCEJQ T VYMBY
g U NPA PDA CCUHGAS NV ESI PYQ PRWDGIL ZO L
c YR TETH EG GYLKE WRZI WMT CU TV AH KM PDSP
a ATVGVJGIIAN MGYT BKYOVEWVX CJ MORF UR
t FVBYPO WFI AMA ARXWJXD NABGA RWWAXEPZ
g SIOLC BJSVNZNNEKJWKQANO TNEJJNKRCM
c WMSPGFN WZR DRRION AO UE RSX RIN NROV GQ
a YOUR IHPYBTFTT KQPC QWGT UZTKPP TQXIS
t VBANSMAR GVHUSTA GHS
g IONA FZNETI UHFGNTU F
c MSREJDRIXMYLJKRXYJ
a OUTGL FTKZOAN LMTZAL
Which spells out something like:
Congratul(a)tions you have completed your technical test.
Visit ????? html for your confirmation code to gain provisional martian status.
Sorry about the formatting. Does anyone know how to use a fixed-width font and bold at the same time?
I think that the canine gene sequence may still be needed.
EDIT Found another word in the messge
_________________"Some people grow into their dreams instead of out of them." - Miles Vorkosigan
Posted: Sat Mar 26, 2005 4:02 am
Last edited by Lysithea on Sat Mar 26, 2005 5:54 am; edited 1 time in total
jbd
Decorated
Joined: 27 Jul 2004 Posts: 295
My guess is 16176 is a starting point in a chromosome, but the message doesn't specify *which* chromosome. Grf.
Posted: Sat Mar 26, 2005 5:23 am
chippy
Entrenched
Joined: 19 Feb 2005 Posts: 789 Location: Leeds, UK
guys, gotta run now (easter family etc), but found this... 16176 gene sequence. gogogogo!!
http://www.tigr.org/tigr-scripts/tgi/tc_report.pl?tc=TC16176&species=Dog
Posted: Sat Mar 26, 2005 10:26 am
deadman
Boot
Joined: 07 Mar 2005 Posts: 57
Great job Lysithea!!!
Spoiler (Rollover to View):
For the others out there, the possible page we are supposed to go to will be
www.aresstation.com/ABCDE.html
where ABCDE
A: L Y C E
B: D Q U W
C: C P T V
D: E R V X
E: J W A C
That's 4096 combinations. I'm tempted to brute force it. The sequence that was posted did not have the pattern we have already derived.
Posted: Sat Mar 26, 2005 11:37 am
Ehsan
Entrenched
Joined: 09 May 2003 Posts: 992
I've been thinking about brute forcing it all day, considering that we have done everything we can to solve it, and actually know how it is supposed to be solved.
But brute forcing just doesn't seem right.. I mean we will find the next page, but something will just be missing.. will you really be satisfied if we don't find that damn sequence??
Posted: Sat Mar 26, 2005 12:32 pm
chippy
Entrenched
Joined: 19 Feb 2005 Posts: 789 Location: Leeds, UK
dont give up yet!
theres also this which seems more on the money?
UniGene Cfa.16176 Canis familiaris
http://www.ncbi.nlm.nih.gov/UniGene/clust.cgi?ORG=Cfa&CID=16176
so, is this puzzle a vignere?
Posted: Sat Mar 26, 2005 1:14 pm
deadman
Boot
Joined: 07 Mar 2005 Posts: 57
I couldn't find an overlap in there either. Here are the translations we should expect to see in the run up to the address.
Spoiler (Rollover to View):
This is the code used for "technical test. Visit"
CCTCTAGACTCGAGCGGC
Keep in mind that we might be seeing the other side of the DNA strand in the published version online. If that is the case, then online, the other side would look like
GGAGATCTGAGCTCGCCG
If we find a match to the second one, then just take the next five letters, and make all the C's into G's and visa versa, and swap the A's and T's.
Posted: Sat Mar 26, 2005 1:34 pm
Ehsan
Entrenched
Joined: 09 May 2003 Posts: 992
I thought this was it! But I can't find anything useful...
Code:
>gnl|UG|Cfa#S22487276 tl23g08.b7 Right Cardiac Ventricle (DOGEST6) Canis familiaris cDNA, mRNA sequence /gb=CV998263 /gi=56268241 /ug=Cfa.16176 /len=340
TATAGGGCGAATTGGGTACCGGGCCCCCCCTCGAGTTTTTTTTTTCCCTTTTTTTTTTTT
TTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTGAAGGCGCAGGGGCCTTTATTAA
AACCCGGGGAGGCAACCCATTCGCTGAAAAAAAGGACACTTTTGTTCCAGAACCACGTCG
CCCCCCCCCGGCTGGGGGGGGGGTGGGCCAATCTGGCCAGACCCCTTTTTCATTACAGAA
GGGACCCGGGGGGCAGGGATGGATCAAAACGTTCTTAATGCCGGGCCCTGGTTTAAGGCC
TGGATTTGGGAGCAAAAGGGAGGGGGAAAGGGGGGGTGAA
And no it's not vigenere, it's more like a ROT using the sequence as a key..
Posted: Sat Mar 26, 2005 1:53 pm
deadman
Boot
Joined: 07 Mar 2005 Posts: 57
I made a mistake before... there are only 1024 possibilities. I made a program the spits out bookmarks for all the possible pages, and then I went through them all, or I thought I did. Nothing but 404 errors, but there are so many I probably made a mistake. I'm hestitant to post the code or the bookmarks, since bruteforcing is not a good thing to submit other peoples servers to. There aren't too many, in computing terms, so I figured it was okay if I did it. I think I will try one more time.
Posted: Sat Mar 26, 2005 2:09 pm
deadman
Boot
Joined: 07 Mar 2005 Posts: 57
Okay, that didn't take as long as I thought... I found a better way of doing it.
Every single page I tried came up with a 404 error. I put in www.aresstation.com by itself just to make sure I wouldn't get a 404 error, and I did not, so I think everything is working properly.
Just so you can get a feel for how I did it, I'm posting 2 screens, out of the 41 I had to go through. Not one of these links worked. So, what do you think the web address would look like, based on our current findings?
Assuming we are supposed to go to a page www.aresstations.com/ABCDE.html, like I thought originally, does not seem like a good place to be checking. Any other ideas?
Description
Filesize
116.88KB
Viewed
273 Time(s)
Description
2nd screenshot
Filesize
116.94KB
Viewed
259 Time(s)
Posted: Sat Mar 26, 2005 2:29 pm
LordKinbote
Decorated
Joined: 25 Sep 2002 Posts: 261
Ehsan wrote:
And no it's not vigenere, it's more like a ROT using the sequence as a key..
Actually, it's a special kind of Vigenere known as a "one time pad". Instead of choosing a short key that repeats itself, you choose a key longer than the message, which usually makes it nearly impossible to crack unless you know the key or know the key is an English (or another language) passage, subject to the same letter and word frequencies as anything else.
So this key *would* be extremely hard to crack, but it only consists of 4 letters in basically a random sequence. Think about how secure it would be if it were 26 letters in a random sequence; we NEVER would have gotten as far as we have. The problem with one time pads is that they're very impractical. Unless the key is something recognizable (FOUR SCORE AND SEVEN YEARS AGO, etc) you need to carry the key around with you or have easy access to it.
Posted: Sat Mar 26, 2005 2:34 pm
deadman
Boot
Joined: 07 Mar 2005 Posts: 57
Yeah, the only reason we could get this far is because we basically found that from each letter, there should be only four things to choose from. If we could not even say that, than any letter could map to any other letter, making this all but impossible to solve without the key. Anyway, I'm not sure what good it will do, but here is a list of all the websites I tried. All of them at http://www.aresstation.com/<Fill>.html
Here's the fill part - I just modified my program to spit out the unknown string to a text file instead of bookmarks. Sorry about posting several times in a row. Anyway, I'm out of amunition for now, been working on this all morning. Maybe one of these 1024 entries is a real english word, I didn't find anything though.
Spoiler (Rollover to View):
Also, here's the full key, split into groups of 10, assuming what we have is correct. With all the long words like Congratulations, provisional, technical, and so on, you can bet a very large portion is spot on.
TTTTTACCAT GCCTGGGCTA GCGTTTGAAA CGGGCCCTCT AGACTCGAGC GGC?????CT CGAGAAAAAG TCTTCTTACC CACTCAGTTC CACCCAAGGG GATTTGGAAT TTGGT
Description
Download
Filename
tries.txt
Filesize
7KB
Downloaded
570 Time(s)
Posted: Sat Mar 26, 2005 2:42 pm
ItWasntMeISwear
Unfettered
Joined: 15 Feb 2005 Posts: 503 Location: Illannoys, USA
p
Posted: Sun Mar 27, 2005 3:19 am
Last edited by ItWasntMeISwear on Sat Jun 23, 2007 6:40 pm; edited 1 time in total
ellipses
Veteran
Joined: 17 Nov 2003 Posts: 138 Location: Location Location
Found it! It was indeed bovine (Bull) rather than Canine. Interesting - check it out here
http://www.ncbi.nlm.nih.gov/UniGene/clust.cgi?ORG=Bt&CID=16176
The first 115 characters are
TTTTTACCANGCCTGGGCTAGCGTTTGAAACGGGCCCTCTAGACTCGAGCGGCCGCGGCT
CGAGAAAAAGTCTTCTTACCCACTCAGTTCCACCCAAGGGGATTTGGAATTTGGT
Note the N which also explains the A in congratulations
Spoiler (Rollover to View):
So it's cqrtw.html - but this seems to 404 for me
Posted: Sun Mar 27, 2005 5:51 am
Ehsan
Entrenched
Joined: 09 May 2003 Posts: 992
Good work!
What I get is slightly different.. cqtrw.html .. can you recheck yours? Still doesn't exist though..
Posted: Sun Mar 27, 2005 6:35 am
Display posts from previous: All Posts 1 Day 1 Week 2 Weeks 1 Month 3 Months 6 Months 1 Year Sort by: Post Time Post Subject Author Ascending Descending