Solution: "True" Friends
Answer: LEMONADE
Author: Peter Vera
The solver begins, presented with three obvious banks of information - a group of 8 seemingly unrelated images labeled “Top Friends”, a list of 8 Google Street View images, each paired with a MySpace comment, and some text presented as a MySpace “interests” section.
The most obvious step to take is to identify the words clued in the comments section, first we find the location of the street view, and pan the camera around to find the word clued in the comments. Along the way, you’d read that these pictures represent 8 countries in a tour, so we will list those as well:
| Clued Word | Street View Link |
|---|---|
| BENCH | Philippines |
| COTTAGE | Poland |
| DOOR | Sweden |
| FIRE | Turkey |
| LOBSTER | Indonesia |
| PEACOCK | Ireland |
| ROOF | Estonia |
| WATER | Czechia |
A solver might consider translating these words into the languages they’re found in, but noticing that Ruby describes themself as "Always wondering about where I'll go next" and apologizes, saying “Sorry I get ahead of myself sometimes!”, the words are not translated into the languages of the countries they’re found in, but rather the language of Ruby’s next destination.
Taking COTTAGE as an example, a solver can translate each word into the other languages in search of something meaningful:
| Language | Translated Word |
|---|---|
| English | cottage |
| Tagalog | maliit na bahay |
| Polish | domek |
| Swedish | stuga |
| Turkish | kır evi |
| Indonesian | pondok |
| Irish | teachín |
| Estonian | suvila |
| Czech | chalupa |
CHALUPA is noteworthy, as it is a product available at Taco Bell and there’s a Taco Bell logo in the “Top Friends” section. This is an example of the linguistic phenomenon of “false friends” where two languages contain a word that is spelled or pronounced the same, but has a different meaning in each language.
We can find similar matches to the other Top Friends.
| WATER | Indonesian | AIR |
| COTTAGE | Czech | CHALUPA |
| ROOF | Turkish | DAM |
| LOBSTER | Swedish | HUMMER |
| PEACOCK | Polish | PAW |
| BENCH | Estonian | PINK |
| DOOR | Tagalog | PINTO |
| FIRE | Irish | TINE |
Using this mapping, we can determine the order in which Ruby visited these countries, recalling that Ireland was stated to be the final location.
| COTTAGE was found in Poland | And CHALUPA comes from Czech |
| WATER was found in Czechia | And AIR comes from Indonesian |
| LOBSTER was found in Indonesia | And HUMMER comes from Swedish |
| DOOR was found in Sweden | And PINTO comes from Tagalog |
| BENCH was found in the Philippines | And PINK comes from Estonian |
| ROOF was found in Estonia | And DAM comes from Turkish |
| FIRE was found in Turkey | And TINE comes from Irish |
| PEACOCK was found in Ireland | And PAW comes from Polish |
And we can index using the number of question marks at the end of Ruby’s questions:
| CHALUPA | 4 | L |
| AIR | 2 | I |
| HUMMER | 4 | M |
| PINTO | 5 | O |
| PINK | 3 | N |
| DAM | 2 | A |
| TINE | 1 | T |
| PAW | 2 | A |
One more translation yields the final answer - LEMONADE.
Author’s Notes
This puzzle is about false cognates, words that are common across two languages, but with different meanings. I first got the idea for this puzzle after seeing this Tumblr post:
Investigating, I pulled a Czech dictionary from github. and wrote some code to compare it to an English dictionary from NPM. When I found out that CHALUPA was a false cognate for the Czech word for "Cottage", I knew I had to construct the puzzle.
I couldn't collect false cognates entirely automatically. I could pretty easily identify words that were shared between a language and English, but in a fairly high percentage of cases, those represented a true cognate or loanword. The easiest way I found to find false cognates from the target language list was to paste the overlap between the two dictionaries into a Google Sheet and use Sheet's GOOGLETRANLSATE function to translate each word back to English, then eliminate the rows where both values were the same.
I was pretty terrified to either get a translation wrong or accidentally include rude slang from a language I don't speak, so I spent a lot of mental energy on validating what I had. I used Glosbe to validate that not only was the translation present, but also that it had at least one example sentence with the translation and conjugation I was using.
Utlimately I went through the exercise of trying to find false cognates for Afrikaans, Czech, Estonian, Hungarian, Icelandic, Igbo, Indonesian, Irish, Norwegian, Polish, Tagalog, Turkish, Vienamese, Swahili, and Zulu. For some of the languages, particularly the African languages, I did find it a little saddening that it was hard to find a comprehensive dictionary on github.
Through the process, I did notice a couple commonalities across languages that I found really cute. It seemed like nearly all languages had some use of the word "bonbon" to mean a small candy, and nearly all language dictionaries had a reference to "manga".
Lastly, here's a few false cognates I wasn't able to use in the puzzle. I am being a little less rigorous about being certain they're legitimate, no guarantee that they're conjugated correctly or that they're not uncommon translations. Grain of salt!
- In Afrikaans and Dutch, SLIM means SMART
- In Czech, POLICE means SHELF and STUD means SHAME
- In Estonian, HUNT means WOLF
- In Hungarian, ORCA means CHEEK and FARMER means JEANS
- In Icelandic, VILLA means ERROR and SOFA means SLEEP
- In Indonesian, EMBER means BUCKET
- In Igbo, EGO means MONEY
- In Tagalog, both HULA and HAKA mean GUESS (this is cool because these are loan words from Hawaiian and Maori respectively)
- In Turkish, GAGA means BEAK or BILL (like a bird's mouth)