66-70: A 2025 Archive: Hidden Clues

[alpine condor]

This is a thread to discuss hidden clues in December 2025 challenges. 👀 👀 👀

(Image: generated by DeepAi)

[frail galleon]

So we're just missing the last one right? Assuming that there really is a hidden code of course

[alpine condor]

well, it glitches

[frail galleon]

I meant if there's an easter egg in general and we're not just seeing things 😂

[alpine condor]

If someone found the clue in the first image, and is too lazy cleaning it up, here is a spoiler.

[alpine condor]

no, can't imagine we are seeing that string from spectral analysis and it is not actually there 🙂

[frail galleon]

Found the second one with || a steganography tool and the third one with a spectrogram the fourth one by using forensically|| I have no idea how to find the last one or if the ones I found are correct though

[alpine condor]

I think you need spoilers

[frail galleon]

I did use the tags. They appear normally for me

[trim blade]

can you send the string from the fourth challenge?

[alpine condor]

|| 706365626e2e637061763531 ||

[trim blade]

thx

[frail galleon]

Do the second and third one||start with the same strings? || I think I might have found the last one but it doesn't work so I need to be sure

[alpine condor]

the ones i found were the same

[frail galleon]

Then I have the last one wrong

[trim blade]

yes the third one was similar

[alpine condor]

or we have the second one wrong

[trim blade]

||742f6c2e69742f752d69396e||

[frail galleon]

The solution of second one seems too clean to be wrong 🤔

[trim blade]

if you convert it from hex to ascii

||t/l.it/u-i9n||

[frail galleon]

Yeah that's how all of them work

We're just missing || the last hex code||

[trim blade]

i didn't even found the IMO code of the vessel XD

[alpine condor]

me too 😭

these might not be ascii

[frail galleon]

it kinda makes sense though if we have to follow a link. Did you find the last one? I'm trying to play with the || colors of the image, contrast, brightness etc ||

[trim blade]

maybe something hidden in the metadata?

[alpine condor]

am I having a wrong first clue? i don't see a link

[frail galleon]

I mean it seems to me that if we unite all of the strings together they will form a link to a file or something

Or maybe it's not a link at all. In this case it might not be ASCII

[frail galleon]

I wonder if the blinking words are part of the solution or just a hint that there's more to this months challenge

[alpine condor]

🙂

[fiery viper]

Hopefully not prematurely killing a train of thought, ||It doesn't appear to be a Base64 encoded binary file. To start, it would be a very small file, fewer than 100 bytes. Also if you assume the first clue as the start of the file, the decoded start of that clue doesn't match any common "magic numbers" for a binary file.||

[frail galleon]

I was thinking about links. Can't they be practically anything? I mean the links to the photos of the challenges are just names so they could be any strings of characters no?

So far we found ||four hex codes. The first one is a bit longer than the rest. Turning hex to ASCII looks like a link probably because of the \ and . || It could be anything though

[frail galleon]

The only thing I found||was a "checksum" I'm not sure what it is exactly. It doesn't look much like the previous codes though||

[trim blade]

||checksum is a value that verifies the integrity of a file||

[frail galleon]

Anyways found a cool website though. CyberChef

[frail galleon]

Has anyone found the last code?

[empty grail]

I see some of you don't even need my hints - you're close. But be ready, there's more to come…

[frail galleon]

Does the code of the final image require more sophisticated methods/tools then the previous ones? Like a combination of various methods for example? Or any previous knowledge in programming?

[frail galleon]

I think I might have found something playing around with the || RGBA bit planes || but I'm not sure if it's relevant or how to decode it exactly. It does look a bit like || binary code which could be translated to hex I guess || but idk

[alpine condor]

@frail galleon ire you mean this?

[frail galleon]

Yes

It kinda looks like|| 0s and 1s doesn't it? Like where there's a blank, 0 and a line for the 1s || but I'm too lazy to convert it 😞

[alpine condor]

i think it is || morse code ||

well, if it is || morse code || it has the right length and digits with e and f.

the string is || 73686e6567636f7472654f34 ||

unless I messed it up, so here is an image with decoded string to verify

and link to || code: https://www.codebug.org.uk/learn/step/540/morse-code-alphabet/ ||

[fiery viper]

Like the other codes, ||that really looks like a base64 encoded string. Really starting to doubt my earlier doubts.||

[alpine condor]

well, unhex return ascii printable characters but makes absolutely no sence || h:aglcorec-M4t/lelamrdhnwt/l.it/u-i9npcebn.cpav51shnegcotreO4 ||

[alpine condor]

i don't think so, base64 uses whole alphabet, we only have hex (0-9, a-f)

now we need to wait ~10(?) days 🙁

[frail galleon]

It was ||Morse code|| indeed. Did we do it?

[alpine condor]

yes

emphasis on "we" !!!!

[frail galleon]

That was so much fun. I enjoyed it a lot more than the rest of the challenges

[frail galleon]

How did you || access the file though? || I keep getting an error

[trim blade]

||you have to write the link using the first letters of each string first, then the second ones and so on, the order is the challenge number so the first string is h:aglcorec-M4||

I hope you understand what I mean, I'm not very good at explaining things XD

[frail galleon]

Yeah, that's what I'm doing. The ||https|| gave it away lol. But I must be doing something wrong

[trim blade]

did you want the || link|| ?

[frail galleon]

No nvm I got the second and third codes mixed

[trim blade]

I now have 10 days to complete the 70 challenge

See you on December 15th

[alpine condor]

We are about to know more:

https://bsky.app/profile/bellingcat.com/post/3m7dashspis2i

Bellingcat (@bellingcat.com)

The next chapter of our December Challenge is about to unfold. Sign up for our newsletter to keep up with all the surprises we have in store for this month’s Bellingcat Open Source Challenge. bellingcat-challenges.beehiiv.com

-# Bellingcat Challenge Newsletter
Join the Bellingcat Challenge newsletter and get quick access to the latest open source challenges released by Bellingcat.

Bluesky · Dec 6, 2025, 2:51 PM

[frail galleon]

A new ARG? Maybe the page will change on December 15 and kickstart a new game!

[frail galleon]

https://youtu.be/74IB4FkuWWI?si=aeP2nPF42I2Cli_3

[alpine condor]

guess it is too late to update the post cover now 🫣

[frail galleon]

The one we have is fine. But I had a wild idea. Probably/hopefully not true. What if we have 164 hours to DECRYPT the archive and after that it will self destruct? What if we shouldn't just wait?

[alpine condor]

i don't see anything that hints at downloadable archive

[frail galleon]

Maybe nothing to do with downloading. Maybe we need to find a new link. By "Self destruct" I meant maybe the link will disappear but probably we just need to wait.

There seems to be something in the source code though

[frail galleon]

There are ||two
"Archivist Notes". Look Closer and Data Is Always Passing Us By||

[royal ridge]

I wonder if there's any significance to the number of flags listed, e.g. ||some base12 type of encoding||. Regardless, it appears there's not much more material to go by at this point in time..

[alpine condor]

technically one could do enumeration. however each link ends with a random string (8 chars case sensitive), so dunno if it is feasible to brute-force anything

[frail galleon]

Nah definitely not

[frozen yarrow]

i remember somebody mentioning ||a link provided by a spectogram of one of the chalenges, maybe they refer to looking closer to the challenges themselves?? still new to this stuff sorry if alr considered||

[frail galleon]

We've already solved that || Each challenge had a hidden string and all together formed a link ||

[alpine condor]

well, basically there are different ways to hide information in an image/audio; each challenge uses a different technique

[frail galleon]

Ah. The next chapter unlocks next week so we just wait.

But why are these notes in the source code?

[frozen yarrow]

ah mb mb

[alpine condor]

@Claire these ?
||

// Archivist note: LOOK CLOSER
// Perhaps these will help with the reconstruction.

||

[frail galleon]

Yes. And one more || data is always passing us by ||

[alpine condor]

maybe we need it next week? 🤔

some flags seem to refer to challenges' locations

[frail galleon]

If you're referring to the country codes I think they are just some of the elements that scroll down the screen along with the latin and Greek alphabet. It's been a while since I took programs lessons but I think they are selected randomly for now. Maybe next week they won't be that random. I don't know why they selected these countries though. Maybe that's another clue.

[alpine condor]

ye, it looks to me like the symbols are falling down in the background;
but, it also says decrypting archive, and when you decrypt it might be relevant what symbols are used in a plain text

I don't see anything to compare it to now, so probably a clue for next week.

actually, it is an archive, it is probably relevant for decompressing too

[frail galleon]

Well all I could gather from looking at the code is that || when the bar reaches 100% the message will change to Finalizing integrity checks and will reload constantly? ||

[alpine condor]

i think it will change in 10 secs after that

[frail galleon]

I think it means it will || reload every 10 seconds no?||

[alpine condor]

yep, but the page can be replaced on the server

[frail galleon]

Yeah probably it will be replaced or a new link will be added somewhere

[empty grail]

Countdown complete… 👀

[alpine condor]

neeeeew challenges 👀

[royal ridge]

I wonder whether the character set is in any way restricted, or if any Unicode character is allowed in principle (especially relevant because of all these (old) Greek characters)

[frail galleon]

I hope so. The Linear B picture is going to be impossible otherwise

I don't have much time this week but I also wonder whether the differences are all || photoshopped changes or they can also be found in the metadata, like coordinates or time taken for example||

[trim blade]

||a challenge requires checking the metadata||

[frail galleon]

Always! But I wonder if the ||differences I found are part of the challenge or not||

[trim blade]

So far, I've only found ||Photoshop edits, except in one challenge.||

[frail galleon]

In || 2 || for me I think

[alpine condor]

how do you find photoshop edits? I don't get that image forensic stuff 🙁

[frail galleon]

I just compare the photo with others I find online of the same area 😂

[alpine condor]

character set should be utf8

there is also some processing:

.toLowerCase()
.normalize("NFD")
.replace(/\p{Diacritic}/gu, "")
.replace(/ /g, ""),

[royal ridge]

Am at 610 'solved' now, of which I'm only absolutely 100% sure of 25 to be completely honest 🙂

[frail galleon]

At first I thought the one with the church would be the easiest but it turns out it's probably the hardest

[flint schooner]

has anyone gotten it yet?

[royal ridge]

For the photo booth, ||strictly speaking, aren't we looking at a 3 char difference rather than 4 char?||

[frail galleon]

I'm missing 2 but I'm not sure if I got the rest right. Especially the one with the sign

I have found the correct location of the ones I'm missing but I can't find recent photos from the correct angle.

I can't find the church and the park (photos 9, 5 I think?)

[royal ridge]

yeah those are very tricky indeed, am also still looking for those

[frail galleon]

For the booth the characters I found were 4

[frail galleon]

Does anyone who has found most want to cross check our answers?

[royal ridge]

Interesting. ||In the GPS coordinates? And all 4 chars are different from the corrputed version?||

For the church, I'm fairly certain I know what part is corrupted, but am still on the hunt for the correct 4chars.

[frail galleon]

To be fair I don't remember if all of them were different. I'll have to check when I get home

Yeah I think I know which part is different but I can't find the original

[royal ridge]

Anyone here happen to live in ||Athens||? 🙂

[alpine condor]

they are not all literal, right?

[frail galleon]

That would be super helpful. I was thinking about it too 😂

It would be easy to just go to the spots and see for oneself

[frail galleon]

What do you mean? Some changes are ||photoshopped and some can be found in the metadata||

[alpine condor]

ok, then no 🙂

[alpine condor]

we could probably do a spreadsheet, who solved what

[frail galleon]

I think maybe we should search independently at first and then compare since we don't really have a way to check our answers. If we find the same answers then probably we're right.

[alpine condor]

was not going to put answers there, just "solved, maybe, no"

[frail galleon]

Oh ok. I'm just missing 5 and 9. I'm not entirely sure about 12 and 18. Also about 11's format

[flint schooner]

https://docs.google.com/spreadsheets/d/13dCcgEsydGAF6Xc54MRPq9M5Rmnhi5TjVDKGT88uG10/edit?usp=sharing

[frail galleon]

@flint schooner found what happens when we get them all right so I guess we can now star collaborating more actively since we do have a way to check our answers.

Bellow is what I have so far. We should cross check our answers and then focus on the ones we don't have yet.

1. ||haus||
2. ||50||
3. ||ε||
4. ||sra||
6. ||s||
7. ||6||
8. ||K||
10. ||tarta|| (?)
11. ||23.7274|| ( format?)
12. ||Β ΣΥΓΚΕΝΤΡΩΣΗΣ|| (?)
13. ||Ямбол||
14. ||ΣΙΛ||
15. ||ΚΙΑ||
16. ||9893||
17. ||Ω||
18. ||2025 (or 4 spaces? Not sure )|| (?)

About 5. I think we're either looking ||for a graffiti on the second bench or the orientation of the image in the metadata. It was taken at the national archeological museum's garden ||

And 9, || my guess is the date is wrong. The church is ιερόσ ναόσ αγίου νικολάου πευκακίων but I can't find any pictures from the right angle and close enough to see the inscription||

[frail galleon]

A typo. 6 is ||s|| I hope we don't need the ||glyph||

[frail galleon]

Oh. 16 also has a typo. It should be || 9893|| but I'm not sure about the format again

[frail galleon]

I think we only need the ||numbers after the . even though technically the 7 is also wrong. Otherwise the characters are 5 || but I don't know. There's a similar issue with the statue too

[fiery viper]

Not sure if this helps get to an answer, ||but Ionian Greek used letters to represent numbers:||
||https://en.wikipedia.org/wiki/Greek_numerals||

[frail galleon]

You mean for the church? That's interesting

[fiery viper]

More that ||if all of the answers are numerals except for the Greek letters, it might be a sign that it's Ionian numerals||.

[frail galleon]

So far I found a mix of || Greek, Latin, Cyrillic, Linear B and numerals || But of course I don't know if I have the correct values

[royal ridge]

You can edit your message, too

[royal ridge]

Thanks for sharing!
I agree with your solutions for 1, 7 & 17.
I share your hesitation with 18 🙂
I disagree strongly with your solution for 8, and disagree somewhat with solutions for 2, 3, 12 & 16
Happy to compare notes!

[frail galleon]

For 8, I think one of the|| cards is missing a K.|| What did you find?

What about the rest?

2 || the address I think is 50 not 49||
3 || the logo seemed different||
12 || I found some similar signs online|| but I really don't know
16 yeah I don't know how to format this one

[royal ridge]

For 8, I'm pretty sure it's about ||a priest smuggling a certain amount of cocaine|| (or smth like that, I don't remember the specifics). I went through ||newspaper archives and confirmed the front page was indeed different on that one crucial character.||

[frail galleon]

You're right! I guess the card really does miss the K 😂

[royal ridge]

2 ||I believe it's the DOB that's wrong||
3 || I found the edition number of one of the books to be incorrect||
12 ||meeting point vs assembly point. Got there after a few rounds of translation back/forth||
16 || There are different ways of formatting these coordinates, I've looked at different ways and can get to a 4 char solution, but it would not be such that ALL 4 chars are strictly different...||

[frail galleon]

2. Oh yeah right. I forgot. That's why I have || 50 ||. So why do you disagree? Is the format wrong?

3. I found the bookstore and their || logo definitely has an ε instead of φ. It's this one https://www.stamoulis.gr/ ||

4. What did you get? I found these images || https://www.horosimansi.com/products/πινακίδα-διάσωσης-με-τίτλο-σημείο-συγκέντρωσης-a-asp4||

5. yeah for this one I think we need to find the correct firmat

For 2. || https://www.instagram.com/athenslikealocal/reel/DRKti3GDJqV/||

[royal ridge]

2 ||DOB is 10 not 50, but your instagram link is convincing!||

[frail galleon]

Oh, probably it means she || lived there from 1950 not that she was born then||

[royal ridge]

yeah

12 || ΣΥΝΑΡΜΟΛΟΓΗΣΗΣ -> ASSEMBLY (note that your link shows assemPly, and 'my' solution does not require the awkward 'B' addition before the word).||

[frail galleon]

But I think || συναρμολόγηση is about putting stuff together, not meeting people https://www.wordreference.com/engr/assembly||

[royal ridge]

3 ||good catch! I found this 6th edition instead of 7th https://www.protoporia.gr/mpatsinilas-epameinondas-sugxroni-logistiki-tomos-a-6i-ekdosi-557927.html but I think your solution is the correct one||

[royal ridge]

can you remove spoilers?

[royal ridge]

Well given the 'ASSEMPLY' example, I don't think these translations are often done very carefully 😉
But yeah the fact that in this case it may be incorrect in the Greek case is def worth considering

[frail galleon]

Sure but, won't this spoil the answer for everyone else though?

[royal ridge]

No I mean ||mask|| them 🙂

[frail galleon]

mask them? I'm not sure I know what you mean

[royal ridge]

Put || around them

[frail galleon]

But I already have it tagged as a spoiler

[royal ridge]

Apologies, you have! I didn't see them anymore since I had 'spoiled' them for myself! 🙂

[frail galleon]

Ah ok. 😂 That's why I couldn't understand what you were asking me to do

So, for which ones are we 100% certain so far?

1,2,3,7,8,17?

I'm also pretty sure about 4,13,14,15 but if someone has these too it would be helpful to check them

[royal ridge]

Yes, nice find on 13! I agree with that.

Let me check my notes right now for the others

The ones I'm 99.9% certain of (can never be 100% with these things ;-)): 2,3,7,8,13,17

[frail galleon]

15 is also || most likely correct. The car is KIA not ford I think ||

[royal ridge]

I'm not sure what to do with 4. ||The original name was sans rival. Yes, there are 3 (legible) letters left, but in the corrupted version we see 5.|| Do we just ignore that? It's a bit 'meta'

[royal ridge]

I agree, it's pretty certain. The only reason it gives me pause is ||that e.g. toyota also made a car that looks just like this, and I'm not familair with all car types to 100% rule out Ford :-)||

[royal ridge]

(for the puzzle makers: I do appreciate the 'NEWEL' joke ;-))

[royal ridge]

Plus, ||the 'I' is also still somewhat legible, making it hard to put the solution in 3 chars||

For 11, it might just be something like ||23.727417, or one less in case we need to include the period||

[frail galleon]

For 1, I think the cafe || changed it's name|| but if someone could check it too would be nice. That's why I have || haus ||

[frail galleon]

Probably yeah. One of the puzzle makers should probably confirm if the number of characters is correct for 11 and 16

[royal ridge]

I'm looking into the decrypt function now, and think we may be running into some utf8 encoding issues with the current set of solutions (with placeholder characters for the missing ones for the moment)

lol the animations are cool

[frail galleon]

On Google maps only the || SRA || are visible. The || I's box is still there but the letter itself is faded|| so I don't know if it counts

[royal ridge]

Thoughts on 10, the Hondos Center?

Also, I believe it's OK to use the glyph in #6 ❤️‍🔥

[frail galleon]

This is the address || Pl. Omonias 3, Athina 104 31 ||. I think the word ||TARTA|| directly below 381 has been erased. There's another||word to the left but the building hides it on the photo of the challenge

||

[royal ridge]

wow nice find, yeah that could be it

[royal ridge]

OK I have a theory about #18, which again touches upon my pet peeve of ||although there's a difference, strictly speaking not all 4 are different. It's taken in Nov 2025, so more likely the calendar is for 2026. And also, I've identified the calendar next to it, which is of 2026||

[frail galleon]

That was my guess too but then I found this||https://www.toumpis.gr/wp-content/uploads/2025/05/CALGC1_CatsNew_MONO_site.jpg||

So I really don't know what the answer should be

If they're making the ||same designs just for different years it would make sense why they don't have dates on their website and 2026 is indeed the answer we need but none of their calendars has the year on the cover ||

[flint schooner]

i am almost certain 6 is ||𐀯||

[frail galleon]

If ||glyphs|| are accepted then definitely

[flint schooner]

it did change its name to ||blue haus|| but its menu online also changed the prices. my gut is its the name though given how recent the photo was taken

[flint schooner]

so its not ||K||?

[royal ridge]

Highly doubt it; where do you see this?

For practical purposes, it would really help if we know whether the final concatenated string should indeed be exactly 66 chars in length and/or how we should treat things like periods and 'changes' in dates (e.g. is "2007" --> "2013" a valid 4 char transformation?) @empty grail

and I guess related whether the answer should consist of consecutive characters found in the originals

[royal ridge]

@frail galleon Regarding 12, one argument for || ΣΥΝΑΡΜΟΛΟΓΗΣΗΣ is that this is what Google translate yields on the corrupted image: ||

And also, Googling for || ΣΗΜΕΙΟ ΣΥΝΑΡΜΟΛΟΓΗΣΗΣ || yields lots of such signs, whereas Googling for || ΣΗΜΕΙΟ ΣΥΝΑΝΤΗΣΗΣ || yields none 🙂

[trim blade]

||i think the right number is 7 not 2||

[royal ridge]

other way around, you mean? ||As in, '7' is corrupted||

[trim blade]

yeah, my bad

||for the number 5, https://maps.app.goo.gl/g8i97kjwiMD77uW89 this is the best view I managed to find, I can see a small graffito on the bench, but it is not readable||

[royal ridge]

Here's a different angle, ||of the row of benches closer to the museum: https://maps.app.goo.gl/FPtYw5DyoMBcvgov7 Graffiti often repeats, || so this may be a clue (although I don't immediately see the solution for 2chars)

[trim blade]

i don't think is the same row of benches of the photo

[royal ridge]

right, that's what I said

[frail galleon]

It is definitely||2|| but I thought it was a K too because there are two cards and only one of them has the K. The other only the figure

[royal ridge]

While we mull over #5 and #9, what about #14? Curious about what people think

[frail galleon]

All of the photos I found had the word ||"ΣΥΓΚΕΝΤΡΩΣΗΣ"|| I can't find a sign with the other word

[royal ridge]

True, but that's 12chars 🙂

[trim blade]

i found the exact location but not a good photo

I'm really thinking about buying a plane ticket to Athens

[frail galleon]

That would be helpful. Even if it's far away, if the sign is a bit visible it could help

[frail galleon]

I know 😢 But the translation doesn't seem right to me based on the dictionary

[royal ridge]

I agree it's questionable

Re 5, I'm starting to doubt it's to do with ||graffiti||

[trim blade]

||https://maps.app.goo.gl/rtEfdfBuVmEBf6it6 number 14||

[royal ridge]

Note that ||the street lights are on, but it's presumably taken at around 9am in the morning...||

[trim blade]

i found a video about the number 9 quality is orrible 🙁 ||https://youtu.be/RsV2ImH_kKw?si=EpwvGxiPxWygTPCy&t=130||

[frail galleon]

I found many videos on YouTube. I couldn't make out the inscription though. It was one that showed it briefly from afar. I think it was from a funeral

[frail galleon]

For 5: ||The orientation in the metadata seemed weird to me. Especially compared to the thumbnail ||but I don't know enough about it to tell for sure. I used this site
||https://29a.ch/photo-forensics/#forensic-magnifier||

[frail galleon]

Time to remember how to use suncalc I guess. Although it seem bright enough for 9

[frail galleon]

The || shadow|| looks correct to me if I positioned it correctly

It could be|| around 16:00|| too maybe? I'm having some trouble finding the right angle from which the photo was taken

[royal ridge]

I agree something seems off with the ||metadata, and in particular the orientation||

[frail galleon]

If the||orientation|| theory is correct then the answer should probably be ||18||. The image appears|| straight but it's orientation reads 90° so it was probably 180° originally?|| I'm kinda confused about how this works exactly but I think this is how it was changed

[royal ridge]

Yeah but the other photos taken with the same camera have the same ||orientation|| data too

[frail galleon]

Yeah you're right. So probably it's just how apple cameras work

By the way, concerning 9, I searched for the inscription and found ||these variations https://www.biblegateway.com/verse/grc/ΑΠΟΚΑΛΥΨΙΣ ΙΩΑΝΝΟΥ 22%3A17||

It doesn't feel like a clean answer but || if ΛΑΜΒΑΝΕΤΩ has been changed to ΛΑΒΕΤΩ, that's a 4 character difference||

[frail galleon]

If we lived in a CSI:New York world we could "enhance" the image 😂

[flint schooner]

||they very well could have just painted over it throughout the years, but it looks like you can see a faint O to the left of the current one. The same thing continues for the other characters until you get to the second epsilon. Maybe they photoshopped that? I've been trying to find a copy of the revelations verse in original byzantine/urical script but am coming up short.||

[teal tulip]

Hi, im newbie, I think I got the link to the corrupted archive but it doesn’t work, could someone please help, is it normal link or there’s also a puzzle whithin the link? Sorry im being slow

[frail galleon]

Yes there is, sort of. You have to figure out how to put them in the right order.

[teal tulip]

So it’s not that it doesn’t work anymore, this is not how it suppose to look?

||https://challenge.bellingcat.com/corrupted-archive-n95OMwn14||

[glass wind]

Double check your decoding, you are SO close

[royal ridge]

Re 9, ||the phrase seems correct. But what if instead of 1999 there used to simply be a reference to the bible verse, i.e. 22:17?||

I found the exact phrase in ||the Third edition of the Greek New Testament (aka Stephanus 1550)||

I've also tried to go through ||any available death records to find Χαραλάμπας, ΚΥΖΙΡΙΔΗ||, to no avail

That ||bible edition|| looks quite impressive btw 🙂

Re 5, ||regarding the streetlight, perhaps it's just AM -> PM (which would once again violate the assumption that ALL characters in the solution must be perturbed, but if that is not a condition then it could work)||

Re 9, ||the year 1999 has always looked suspicious to me, especially because at the pixel level some of the 9s look very similar. However, if indeed a condition of a valid solution is that ALL characters must be perturbed, and assuming the 9s are coming from some other position, I don't see a valid perturbation.||

(I don't think I explained myself very well there)

[frail galleon]

Yeah, me too. I couldn't find anything about his ||death||. There was a Facebook page with ||a young guy going by the same name who uploaded a black square on 2014|| if I remember correctly, but it looks more unrelated than not.

I was pretty sure at first that the ||year|| is what has been changed. It fits given the length of characters. I thought about brute-forcing with every || year between 200-2025|| but without being 100% sure about every other answer it's pointless.

I wonder if what @fiery viper said about|| Greek Ionian numerals|| has something to do with the answer. Like if we need that version of||1999|| which would be||͵ΑϠϞΘ´|| but strictly speaking that's more than 4 characters

[frail galleon]

If this is what has been changed perhaps we need ||16|| instead of||9|| but like I said I'm not sure about the angle from which the photo was taken. I tried using suncalc and I think it's one of these two|| hours|| though

[frail galleon]

So about the coordinates, does the dot count as a character? And if so, do some of them require more or less decimals?

[royal ridge]

That's the question indeed
#1445785367482732722 message

[frail galleon]

I figured we should ask again since we didn't get an answer last time 😭

[royal ridge]

I just finally found a satisfactory picture of the central pedestal in #14

[royal ridge]

And also finally found the exact location of #12, although it's not exactly high res

Can at least confirm it's ||a 2 and not a B, and that the Greek word is at least as long as the English 'ASSEMBLY POINT' ||

[empty grail]

You're right, there's definitely something wrong with some of these coordinates! I've double checked the files and analyzed the checksums, but I can't locate a decimal point in the coordinate metadata.

No second guessing!

[royal ridge]

Thanks! If no decimal points, do they follow DMS convention, and if so, do we exclude/include °,'," ?

[empty grail]

I can confirm that 66 characters appear to be corrupted in the tranche of images I've shared with you.

Corruption takes many forms, if a year was corrupted from 2007 to 2013 with only the last two characters affected, then it would be 2 characters. But if the entire year was corrupted (changes in size, font etc) that could well show up as a 4 character corruption.

For a coordinate following DMS, only the altered characters should be recorded, so ° ' " should be safely ignored.

[frail galleon]

Thanks!

[royal ridge]

Thank you. And importantly, is a corruption always based on consecutive/adjoining characters or can it 'skip' one or more characters?

(The DMS example suggests the latter is the case, but that would also mean that e.g. a corruption from e.g. 1007 to 2013 would involve 3chars only, assuming no perturbation in size/position/font etc)

[empty grail]

From my own tests, it seems that a corruption can skip characters, but that it isn't a frequent occurence.

[royal ridge]

oh boy. And, say, 28.20->27.12 counts as a 3 char diff, correct?

[royal ridge]

@empty grail Just one more clarifying question: with 8 corrupted characters for #11, there are a very large number of possible solutions that are all approximately equally 'correct'. || if we're talking about GPS coordinates in decimal notation with 6 decimal places, we're looking at placements at around 10cm accuracy. I can imagine getting closer by incorporating e.g. camera/zoom settings, but it's still a bit of a long shot. || Can we assume there is one unequivocal correct answer?

[frail galleon]

So is this the correct format? I'm hopeless when it comes to formatting so I'm sharing the correct locations too for those you can handle it better

Statue: ||23433868|| --> coords || https://maps.app.goo.gl/VUZGZCMQ8VLbFrcM9 ||

Phonebooth: ||3759|| --> coords || https://maps.app.goo.gl/9XNj3HX7XU2CJDoL6 ||

Also, do spaces count as characters?and if a change in font also counts, then perhaps for the church we just need || 1999|| ?

[royal ridge]

This perfectly illustrates my point, thanks for making it concrete @Claire!

Statue: ||yes could be 23°43'38.68"E, but could just as well be, e.g. 23°43'38.73"E. Or, with a decimal notation, 23.727444, or 23.727440, etc :-)||

Phonebooth: I agree that that's the cleanest solution, where it not for the fact that ||3800 -> 3759|| is a 3char corruption

[frail galleon]

Also, I looked around || Akropolis|| for similar signs. Since we can see a 2 I figured there'd be more of them and I found another one which looks a lot cleaner. I think the Greek word is || ΣΥΓΚΕΝΤΡΩΣΗΣ||

[frail galleon]

Damn! This is the hardest part of these challenges

[royal ridge]

Excellent sleuthing! But the solution can't be that as it's 12chars?

[frail galleon]

I know. ||Perhaps the number changed too? || Did you find the exact location of the sign in the photo? I really don't know what to make of this one

[royal ridge]

yes I found the exact location and it's 2. Perhaps we can find others too (3-8) 😉

[frail galleon]

Probably. Also I kinda think there are more letters between the two last Σ. The space between them looks larger I think? So could we be looking for a variation of the word with more letters?

[royal ridge]

@frail galleon I found the sign for 12 @ ||37.9718313, 23.7276947||, can you share where you found the other one? I'd like to see if we can find some other ones.

[royal ridge]

Re #5, ||the more I look at it, the more I think it could be a corruption in timestamp. I see the shadow going towards the NE which (I think) would indicate that it's in the afternoon. This would also fit with the streetlight being on. HOWEVER, the instructions say that each image is corrupted in exactly 1 way, and this particular corruption would involve changes to both the EXIF metadata and the filename, so that would be two corruptions...||

[celest wraith]

Hey everyone! I'm a abit late to the party from the looks of it. I've managed to get the ||embedded link in the pictures, but once I try to access it I get a 404 from Github. Did the website close or am I doing something wrong?|| Thanks

[royal ridge]

Website is still up and running! Perhaps you got something wrong in the final encoding?

[celest wraith]

can I just send the link that I got here or is it against the rules? You can just confirm is it correct, if its wrong dont tell me what. Would like to find myself haha

[royal ridge]

sure, just make sure to mask for spoilers!

and can always delete after confirmed

[royal ridge]

you are insanely close. But I think you copied over one character incorrectly. Go back and check your input and you'll get it!

[celest wraith]

haha, thanks for cheking!

[frail galleon]

Here || 37.970331,23.728324 ||

[frail galleon]

That would fit with the shadow I found on suncal. It was at around || 16:00 || just before the sun started to set. But yeah, that would mean both the name and the metadata were changed. Unless it counts as one change because it's the same one technically

[frail galleon]

I think this is 2 again?

[royal ridge]

yes, it is the one from the corrupted pic. Am just trying to find some others now

[frail galleon]

Oh I thought you said you found 12

[royal ridge]

No no, challenge/pic #12! 🙂

Here's sign #7 I think, but it's completely illegible

what doesn't help is that these signs have been up for what seems to be less than a year, so most photo material doesn't even have them 🙂

OK so I think challenge/photo #9 is really the only one that remains a complete mystery (at least to me)

[royal ridge]

Re 12, it's unfortunate that ||https://mysafetyplan.gov.gr/|| is essentially useless. Am also going over academic papers that study optimal evacuation procedures to see if they recommend specific locations for such points, but nothing much came out of that

[frail galleon]

I was just looking at this and had the same reaction. I think the signs are a recent addition too. Otherwise we would have found more by now.

[frail galleon]

I think it's about the || date. It looks kinda different than the rest of the lettering||. Then again we can access the metadata so perhaps there's something there.

The last one is a bit confusing too. The photos I found didn't have the || year || and I don't know if||spaces count as characters||

[royal ridge]

yeah but I think it's like you said: ||the publisher just shows a 'vanilla'/undated version, and once there's a big order they customize it for the desired year. We can at least assume that this particular calendar is instantiated for a particular year, and that it is likely going to be at least fairly recent (and perhaps even as 'recent' as 2026, although that is a perhaps unlikely given the discounted(?) price).||

[frail galleon]

Shoot! The ||dimensions|| of the ||second calendar|| are wrong. Unless there are more than one ||sizes||
||https://www.triperinas.gr/product/hmerologio-spiral-ellada-25x35/||

We should check if the || number on the side corresponds to the size of the image||

[royal ridge]

That's the same one I had found before. Doesn't change much, right? Given how the pic was (presumably) taken recently.

Here's the 'correct' one: || https://www.triperinas.gr/product/imerologio-toixou-topia-megalo/ ||

Oooohhh actually, || 4 euro might not even be that unrealistic, given how this one is discounted to 6 euro! https://arhetyposhop.gr/hmepologio-toixoy-spipal-ellada-21x29-16fylla/ ||

[frail galleon]

Yeah, I think the second calendar has || the correct barcode for it's dimensions||. I really hope it's not about the ||name of the street|| because I really can't make it out in the photo to compare it with the store's

I'm contemplating whether I should create a fake account to check the price of the cat calendar. I don't really want to 😢

[royal ridge]

Kiosk price will be different though, so no real point. I'm mostly thinking it's ||not 2026 because of how weathered the price label looks||

[frail galleon]

Yeah but we could see if it's on discount

I mean they wouldn't sell it at a lower price if it were new

Regarding the sign. I think we need || half the Greek word and half the English one || The Greek one has ||some letters that are the same || but I think the English one seems a bit different in the picture I sent

[royal ridge]

And yet again, this comes to down to whether if any characters in a word/string change, all characters are considered to be a 'corruption'. In this case, it could be justified by (slight) changes in position maybe, but in case of e.g. || GPS coordinates || I don't see the justification 🤷

[frail galleon]

I think it has to do with the font as well like the @empty grail said. So for the sign for example, we'd need ||the entire word|| but this leaves us three letters short. Do any of the letters of the word above seem off? Like their font has changed too?

I've also uploaded the calendar picture on || forensically ||. I'm not entirely sure how their error analysis tool works but it seems that something has been changed.

[royal ridge]

Almost looks like ||2026->2025||

[frail galleon]

Yeah. ||5 seems to be what's changed||

We should make an updated list of the answers soon

Once we clarify how the coordinates should be formatted

[royal ridge]

The calendar ||(and price tag!) looks so poorly/weathered though, and combined with the low price I have a hard time believing it would indeed be for 2026, so 2024 might be more likely after all. (then again, perhaps it is indeed 2026 but just discounted because it's in bad shape lol||

[royal ridge]

...||file on the publisher site says it was uploaded in 2025, and it also lists the calendar as 'NEW', suggesting it came into existence this year. So then it must be 2026, and this particular instance of the calendar is just discounted cause of the poor condition it's in.|| Although it violates the 'strict' interpretation of character corruption, my guess for the solution of this one is therefore... ||2026|| after all.

[frail galleon]

I found sign No 4 too but it's not legible
||https://maps.app.goo.gl/bWwezJapBy82d5Ft5?g_st=ac||

[frail galleon]

Yes. Probably that's what we need. I agree.

[royal ridge]

jean-louis Quintana delivers once again! (albeit suboptimally)

It occurred to me it may be a good idea to at least map the locations of these assembly points, as it may make it easier to find the others. For instance, points 7 and 9 are quite close together, suggesting that 8 may be inbetween them

[frail galleon]

Yes! That would make sense. The only problem is that they were installed in 2025 it seems

This is what we have so far right?
1.||haus||
2.||50||
3.||ε||
4.||sra||
5.||16|| (?)
6.||𐀯||
7.||6||
8.||2||
9.
10.||TARTA||
11.||23433868|| ( format?)
12.||ΣΥΓΚΕΝΤΡΩΣΗΣ||
13.||Ямбол||
14.||ΣΙΛ||
15.||ΚΙΑ||
16. ||759||
17.||Ω||
18.||2026|| (?)

[royal ridge]

Yes that's it. Except that personally I am still working with ||ΣΥΝΑΡΜΟΛΟΓΗΣΗΣ|| for 12 (despite your very valid concerns about that).

[frail galleon]

We need to find a clear picture for 12

[royal ridge]

Or have @empty grail tell us there has been a terrible mistake and the number of corrupted chars is 12 🙂

[frail galleon]

What I'm hoping is that there aren't any changes in fonts. Because if the letters themselves are the same but the font is slightly different and it still counts as a change then I don't think I'll be able to find it

[frail galleon]

Found a video
||https://youtu.be/2p17FKPov8Y?si=ilzt3_HeDOMUcf-F||

[royal ridge]

Re photo number 9 (church) I have a nagging feeling that ||while it is indeed the date/year, it may be a very recent one (2000+)||

[frail galleon]

At~ 2:30 you can see sign No 9

[royal ridge]

lol that is exactly what I did for hours yesterday: go through walkthrough videos on youtube 🙂

[frail galleon]

Yeah, it's either that or we need the verse number? (not sure how they are called)

[royal ridge]

yes, in which case it would be ||2217||

haha I love how the guy in your video is actually tapping on the sign and having his pic taking with it! It's exaclty the type of footage we would need 🙂

That's a bit better!

[frail galleon]

I know right? It's as if he knows what we're doing 🤣

[royal ridge]

still looks like 12 chars tho

[frail galleon]

Yeah I know. It's definitely|| ΣΥΓΚΕΝΤΡΩΣΗΣ|| I can't read the English word though. Maybe it's slightly different too?

[royal ridge]

That's possible, maybe he's pointing out ||a silly typo?||

[frail galleon]

Could be. I can't look at the whole video right now but perhaps there will be more signs. It seems like there are more letters than ASSEMBLY

[frail galleon]

This too || https://www.facebook.com/groups/271625487184735/posts/1533492274331377/|| I don't have a FB account so I can't see the group but the photo has No 3

[royal ridge]

Oh wow, about as clear as it gets. Great find.

[royal ridge]

How'd you find that without a FB account?

[royal ridge]

Astoundingly, it appears that in this video sign number 2 was not (yet) placed!!

[royal ridge]

Worth noting this is likely not part of the ||'series' of 1-9 assembly point signs at the Akropolis, as it is taken at quite a distant location from it (Pnyka)||

So there could still be differences with the one shown in the original photo assignment

[royal ridge]

In that case it would however also include a reference to the|| 'Book' i.e. Revelations||

[empty grail]

There has been a terrible mistake.

You are right that the intended solution ||ΣΥΝΑΡΜΟΛΟΓΗΣΗΣ|| has ||14|| characters, but is ultimately incorrect because the original sign has ||12|| characters ||ΣΥΓΚΕΝΤΡΩΣΗΣ||.

I have fixed the archive so it should ask for 12 characters now.

You will definitely get a special mention in dispatches for not only noticing the mistake, but finding the correct answer and the mistaken answer.

I can't apologize enough for this mistake!

[frail galleon]

Google image. I searched for|| " Akropolis" "Assembly point" sign|| Then I chose to open the link in incognito mode. Most of the time it works but you can't see much

[frail galleon]

Thank you for clarifying this!! It was driving us nuts!

[empty grail]

Appreciate @frail galleon and @royal ridge for calling this out! DId anyone else help point this out? (adding a special mention to the site)

[frail galleon]

Isn't always something like 12:42?

[empty grail]

This is on the right track - degrees and minutes, no second guessing

Where there is latitude and longitude, they should be specified in that order.

This should be a ||3|| character corruption as you point out - I have updated the site accordingly!

Once again, my sincere apologies for these errors!

[royal ridge]

The only thing is that I think you may have tagged the wrong Claire?

[royal ridge]

Ok so what we have now is:
1: ||haus||
2: ||50||
3: ||ε||
4: ||SRA||
5: ||16|| (?)
6: ||𐀯||
7: ||6||
8: ||2||
9: ||absolutely no idea!||
10: ||TARTA||
11: ||23433868|| (or some slight variation or different encoding of that)
12: ||ΣΥΓΚΕΝΤΡΩΣΗΣ|| ❤️‍🔥
13: ||Ямбол||
14: ||ΣΙΛ||
15: ||KIA||
16: ||759||
17: ||Ω||
18: ||2026|| (?)

I went outside for a bit today and all I saw was Greek characters (note, in reality there were none) 😄

[frail galleon]

We're getting close! For 9, the only thing I can think of is || the date ||. I don't know if I have been staring at it for way too long and I started seeing things but I noticed something||on the bottom left too. Probably the artist signature || but I can't make it out

[royal ridge]

Interesting. But regardless of whether it's that or the ||year||, we don't have a good enough view of the thing, unfortunately

[frail galleon]

Yeah. I wonder if the best thing right now would be to verify Numbers 5 and 18 and then "brute force" the ||date||.

Re 5, I'm getting more and more convinced that the ||hour|| is what has been changed. Your observation about the|| street lamp|| was spot on. The only problem is I'm not confident that I got the right ||time using suncalc||. We could also try to || map the journey of the photographers to see if being at the park at 9 makes sense. We have timestamps for many of the locations||

[frail galleon]

It seems that the photos taken on || Nov. 10 (excluding 5) were all taken between 17:01 and 18:26 except number 15 which was taken at 12:02||. So probably 5 was also || taken around that time||? We probably need || 18 || though because of || time zones||. Does this count as 2 changes?

[royal ridge]

hmm interesting. I had already mapped all (well, most) of the photo locations to see if there were any patterns, but had not yet related it to the dates/times.

Looking at the dates/times of the file(name)s, it appears that photo 18 was also taken around there. I wonder if this is the stall! (not that it helps a lot)

[frail galleon]

Nice find!! I think you're right about the calendar being || on discount because of it's condition||. I can see ||wrapped calendars behind it|| But if the answer is||2026|| wouldn't this mean again that the characters we need are not|| 4 but 1?||

[royal ridge]

yep. unless it's a different ||font||

Note that we're also not quite yet sure about 11

at least I'm not

[frail galleon]

Yeah, I was going to mention that. We need to figure out the correct format because technically some || characters are still the same right? It's 2 19 38.79 vs 2343386*. || And I think the final char is || 5 || after all.

Like no matter how you write it, at least one character remains the same AND at the same spot. Example:

||
02193879
23433865

or

21938790
23433865

||

[royal ridge]

yep

and note that it could be ||DMS or decimal notation||

Oh, I just spotted a potential error with the decryption check

[frail galleon]

I thought it was confirmed that all ||coordinates are DMS||?

[frail galleon]

What error?

[royal ridge]

will need to figure something else out first:

I just re-read the note by the Archivist, and I had been thinking about their remark of 'no second guessing'; of course that was intentional! lol

That suggests that the solution for 11 could be ||37582343|| (again, not even nearly all corruptions, but it would fit their remarks)

[frail galleon]

Maybe the last character of the || latitude is also wrong ||

[royal ridge]

Not sure I follow

[frail galleon]

The Archivist also said that "where there's latitude and longitude they should be specified in that order"

[royal ridge]

To my earlier remark about a potential error, I just want to check with @empty grail :
Are you absolutely positive we're looking for "CORRUPTED_ARCHIVE_RESTORED_8dh36rhhjy31tu45" as the decrypted message (INCLUDING quotes!) and not CORRUPTED_ARCHIVE_RESTORED_8dh36rhhjy31tu45 (without quotes)? Because the former is currently being used, and that seems a bit odd

[frail galleon]

You kinda lost me there 😆

Oh you mean after we find all of the answers

[frail galleon]

No second guessing could also mean that we got it right

[royal ridge]

Yeah but not in the context of the rest of their sentence I think

[trim blade]

Hi, I didn't follow the challenge much over the weekend. Any important updates?

[frail galleon]

Also, is No 5 corrupted in 2 ways? || Both the file name and the exit data ||?

[frail galleon]

Lots! 😂 I'm not sure I can narrate everything but you can find what we have so far in one of the previous messages. Basically we know nothing about 9 and we're not quite about 5, 11 and 18.

[trim blade]

I'll try to reread the chat and catch up

[royal ridge]

I got it!

[frail galleon]

You found 9??

[royal ridge]

got it all 😄

Already knew what the animations would look like upon completion (they're really cool), but it was great to see it 'as intended'

[royal ridge]

I ended up brute-forcing it

[frail galleon]

What's the answer then? Did we have any of the others wrong too?

[royal ridge]

yeah

So we were not sure about 5, 11, 18. And of course 9 was a complete mystery.

For 5, I extended the search range slightly (also because of your remark re timezone) and that ended up being ||15||

For 11, I do think it is a bit misleading / incorrect in the sense that the solution contains mostly uncorrupted characters: ||37582343||

For 18 it was indeed ||2026||

And 9... oh boy. I realized that if it would indeed be a 'full' 4 char corruption, then it could not be ||19xx||. So I tried ||2000+|| only and it ended up being ||2006||

[royal ridge]

And so this ended up being 'correct' in the code, i.e. including quotes (I don't remember this being the case in the past)

[frail galleon]

5 doesn't make much sense though. It should be either || 16 || or || 18|| no? Why do we need || +1 || instead? Unless local time wasn't||18||

Yeah, the coordinates one is definitely misleading. The formatting was a nightmare

Congrats on getting it right!! 🎉

[royal ridge]

Congrats to you, too! This was fun

(and I agree a bit frustrating w.r.t. formatting)

[royal ridge]

Sun sets at around ||5pm there in Nov I think? so 15:00 (3pm) + 2 = 17:00 (5pm)||, maybe? 🤷
Either that or the corruption was not ||the time|| at all and just got very lucky

[frail galleon]

Could be yeah. So local time ||17|| not ||18|| makes sense

[royal ridge]

(but yeah, like you said, that does mean ||there were two corruptions there...||)

[frail galleon]

Exactly. Assuming of course the ||time|| was the corruption but it fits besides this detail

We made a good team!

[royal ridge]

Absolutely! Was a pleasure

And thanks a lot to @empty grail / @glass wind / @amber fulcrum etc for organizing all of this!

[frail galleon]

Yes!! These challenges were fun! Thank you for this!

[amber fulcrum]

Wooo! YOU DID IT!!! So exciting to follow you solve these puzzles, it was hard to keep up! What a great team. Hope you enjoyed working on these 🙌 🇬🇷

[frail galleon]

We absolutely did!

[glass wind]

HOORAY! Congratulations 🎉🎉 Well done on getting it all and navigating the myriad imprecisions!

[trim blade]

Completed, what a great challenge, I can't wait to try the new ones in February

[flint schooner]

out of curiosity since you are the creators @amber fulcrum @glass wind looking back on all the challenges from the year, which one would you say was the most difficult?

[trim blade]

I didn't do them all, personally the most complicated ones were numbers 55 and 70

[frail galleon]

Personally, I haven't been able to solve only one throughout the year so I'm guessing that's the hardest one for me.

Door to Nowhere

[glass wind]

I think 'Door To Nowhere' takes it for overall difficulty, despite being conceptually straightforward.

For the challenges we often check that the solution path feels possible and might contain some useful or interesting skills or tools. Trying to make them possible is part of the editing process.

But for Door to Nowhere we checked that imagery was available on street view and left the checking there. Consequently there was very little to go on!

I think Hopping Around takes second place for difficulty for similar reasons!

But ultimately that's been one of my learnings over the last year - the difficulty of a challenge is extremely subjective based on people's knowledge of different topics, tools and approaches.

I'd love to hear others' thoughts on the challenges though (we even have a feedback form on the Corrupted Archive completion page... 👀). Particularly interesting would be ideas/types of puzzle that we haven't seen yet, or topics that we haven't included or are underrepresented!

[amber fulcrum]

Good question! Door to Nowhere definitely came to mind. I really thought someone must have posted a photo of that door somewhere, quite a surprise no one could find any reference material.

For me the explosive ordnance challenges were also a tricky one. But like Galen said, that's probably because that was a very new field of research for me.

[frail galleon]

I've spent two days trying to see if there's a hidden message in the video. Now I'm suspicious of everything.

[trim blade]

Another rabbit hole? XD

[frail galleon]

Who knows?

[alpine condor]

@glass wind Maybe we should have a separate thread to discuss this

[amber fulcrum]

We can open one but I think using the feedback form will be most effective. I'll share it here, and in the next newsletter as well so you don't just get access to it when you have solved the archivist challenge! https://docs.google.com/forms/d/e/1FAIpQLSdv5QcWke5GOu6GT3ZyOdb6BmfMncDwDLk_9cwckJ25cQ8f9A/viewform?usp=sharing&ouid=109837718843509998774

[tacit socket]

I was so gutted I lost internet for over a week at the release of the Archive challenges 😓 going to try to catch up from that point now