12
$\begingroup$

I know y'all want to be generalists or something, but here's a little sudoku for (geography) specialists. Enjoy.

enter image description here

ABRAU ⋅ AGANTON ⋅ AKUSEKIJIMA ⋅ ALICE HOLT ⋅ ALTENBURG-NOBITZ ⋅ ANJEDIVA ⋅ APULIA ⋅ ASHTAMUDI ⋅ ATTU ⋅ CAMPO DEL MORO ⋅ CARLHAUSHÖHE ⋅ CASA DE CAMPO ⋅ CHAMP DE MARS ⋅ CHANYCHULMAN NERYUNGRICHUSOVSKOYECOCHIN INT. ⋅ CROTONE-SANT'ANNA ⋅ GELSENKIRCHEN ⋅ GIARDINI MARGHERITA ⋅ GILA ⋅ GIRONA-COSTA BRAVA ⋅ GRANDE CASSE ⋅ GREAT OUSE ⋅ GROßE LABERGÜELLGWYDIR ⋅ LA FLORIDA ⋅ LAOUZAS ⋅ LAUVITEL ⋅ LEIPZIG ⋅ LEMMON ⋅ LOIRE ⋅ LOMOND ⋅ LOUISIANA ⋅ LUCRINUS ⋅ MAES-YR-UCHAF ⋅ MAHANADI ⋅ MAJULI ⋅ MÁLAGA ⋅ MALAGUNI ⋅ MANZANARES ⋅ MAUNA KEA ⋅ MILAN ⋅ MOSSÈN COSTA I LLOBERA ⋅ NAPLES-CAPODICHINO ⋅ NARMADANATURE ET PAYSAGESNEMADJI ⋅ NÉOUVIELLE ⋅ NETLEY HEATH ⋅ NORMANDY ⋅ NORTH OSSETIA-ALANIA ⋅ NUREMBERG ⋅ OBERHAUSEN ⋅ ODISHA ⋅ OHIO ⋅ OKINAWA PREFECTURE ⋅ OLDMAN WOOD ⋅ OMODEO ⋅ OMSK OBLAST ⋅ OREGON ⋅ OSMANABAD (OMN) ⋅ SAGANO BAMBOO ⋅ SAKHALIN ⋅ SANCTI PETRI ⋅ SANTERNO ⋅ SHEREMETYEVO ⋅ SHIKOKU ⋅ SOBRETTA ⋅ SOIERNSPITZE ⋅ SPREE ⋅ TADASU NO MORI ⋅ TAIHEIYO EVERGREEN ⋅ TIREE ⋅ TOKUSHIMA AWAODORI ⋅ TOKYO ⋅ TOM A' CHÒINICH ⋅ TOMSK ⋅ TOYAMA ⋅ TRESIDDER

$\endgroup$
  • 1
    $\begingroup$ (geography) specialists or (Google) specialists? $\endgroup$ – m4n0 Feb 17 at 12:46
  • 2
    $\begingroup$ @ManojKumar A strong Google-fu certainly helps! $\endgroup$ – jafe Feb 17 at 12:50
  • $\begingroup$ What's the criteria to fill in? I thought a character cannot take the same position of the block in another grid. For eg: C is repeating in Cochin and Chulman. Also G is repeating in bottom right. $\endgroup$ – m4n0 Feb 17 at 16:08
  • $\begingroup$ @ManojKumar The criteria are not explicitly specified on purpose. All tags are relevant, though :) If nobody gets it in a day or two I'll add a hint. $\endgroup$ – jafe Feb 17 at 16:28
  • 1
    $\begingroup$ I find no solution, which I assume means I've miscategorized something. (There are a number of names that correspond to multiple categories, and I didn't by any means prove that the specific way I classified them was the only way to get the numbers right.) Of course, I might also have misunderstood what it means to solve the puzzle, but I'm pretty sure I've got that right. $\endgroup$ – Gareth McCaughan Feb 18 at 17:35
5
$\begingroup$

The completed grid:

Omsk Oblast         Gelsenkirchen      Shikoku                | Alice Holt        Tresidder          Lauvitel             | Mahanadi           Casa de Campo    Naples-Capodichino
Nemadji             Manzanares         Ashtamudi              | Chulman Neryungri Okinawa Prefecture Santerno             | Tiree              Grande Casse     Leipzig
Tom a' Choinich     Loire              Crotone-Sant'Anna      | Majuli            Nuremberg          Guell                | Sagano Bamboo      Oregon           Abrau
--------------------------------------------------------------+-----------------------------------------------------------+-------------------------------------------------------
Spree               Tokushima Awaodori Neouvielle             | La Florida        Gwydir             Attu                 | Chany              Milan            Odisha
Malaga              Anjediva           Louisiana              | Omodeo            Sheremetyevo       Carlhaushoehe        | Nature et Paysages Tadasu no Mori   Great Ouse
Giardini Margherita Chukovskoye        Oldman Wood            | Normandy          Malaguni           Toyama               | Lemmon             Altenburg-Nobitz Sancti Petri
--------------------------------------------------------------+-----------------------------------------------------------+-------------------------------------------------------
Akusekijima         Ohio               Tomsk                  | Soiernspitze      Lucrinus           Maes-yr-Uchaf        | Girona-Costa Brava Narmada          Champ de Mars
Cochin Int.         Sobretta           Mossen Costa i Llobera | Gila              Aganton            North Ossetia-Alania | Oberhausen         Lomond           Taiheiyo Evergreen
Laouzas             Netley Heath       Grosse Laber           | Tokyo             Campo del Moro     Osmanabad (OMN)      | Apulia             Sakhalin         Mauna Kea

So, first of all,

we can classify each of our 81 names in three ways: by first letter; by the country where the thing is; and by what sort of thing it is. For some of them there are multiple possible classifications or multiple possible countries; there are a few different ways to pick and choose so that everything is consistent; both Jens and I made plausible but wrong choices, and jafe kindly put us out of our misery in the comments. The correct assignments are below. I hope.

So:

ALR ABRAU · karst lake in Russia, spur of Caucasus mountain range, winery, peninsula
AIF AGANTON · tidal island off north coast of Brittany in France
AIJ AKUSEKIJIMA · Japanese island, Tokara group in Satsunan Is. in Kagoshima Prefecture
AFK ALICE HOLT · royal forest in Hampshire, UK, formerly oak, now conifers
AAG ALTENBURG-NOBITZ · airport in Thuringia, near Leipzig, Germany
AII ANJEDIVA · island in Arabian Sea off Canacona in Goa, India; fort
ASY APULIA · region of Italy, southern peninsular section, eastern coast, Adriatic/Ionian/Otranto
ALI ASHTAMUDI · lake in India; Kerala
AIU ATTU · US island, westernmost point of Alaska, in the Near Islands, uninhabited
CPS CAMPO DEL MORO · park in Madrid, Spain; gardens of Royal Palace of Madrid
CMG CARLHAUSHÖHE · mountain in Harz mountain range in Northern Germany
CPS CASA DE CAMPO · park in Madrid, Spain --or-- seaside resort in Dominican Republic
CPF CHAMP DE MARS · park in Paris, France, near Eiffel Tower
CLR CHANY · lake and municipality in Novosibirsk, Russia
CAR CHULMAN NERYUNGRI · airport in Yakutia, Russia
CLR CHUSOVSKOYE · lake in Perm Krai, Russia
CAI COCHIN INT. · airport in Kochi, Kerala, India
CAY CROTONE-SANT'ANNA · airport in Crotone, Calabria, Italy
GCG GELSENKIRCHEN · city, in Westphalia, Germany
GPY GIARDINI MARGHERITA · park in Bologna, Italy
GRU GILA · river in New Mexico, US; some places in Arizona; = Sonoran Desert; mountain range in US
GAS GIRONA-COSTA BRAVA · airport near Girona, Catalonia, Spain
GMF GRANDE CASSE · mountain in Vanoise Massif, Graian Alps, Savoie, France
GRK GREAT OUSE · river in UK
GRG GROßE LABER · river in Bavaria, Germany, tributary of Danube
GPS GÜELL · park in Barcelona, Spain
GFK GWYDIR · forest and castle in Conwy in Snowdonia in Wales, UK
LPS LA FLORIDA · airport in Chile --or-- wetland park in Colombia --or-- US state --or-- commune in Chile; but treating as Spanish, which it apparently must be
LLF LAOUZAS · lake in France
LLF LAUVITEL · lake in Isere department of France
LCG LEIPZIG · city and region in Saxony, Germany
LMU LEMMON · mountain in Arizona, US --or-- various US towns
LRF LOIRE · river in France
LLK LOMOND · lake in Scotland, UK
LSU LOUISIANA · US state; various regions and towns
LLY LUCRINUS · lake in Campania, Southern Italy, near Avernus
MFK MAES-YR-UCHAF · forest in Monmouthshire, Wales, UK
MRI MAHANADI · river in East Central India
MII MAJULI · island in Brahmaputra River in Assam, India
MCS MÁLAGA · airport, province, municipality in Spain
MRI MALAGUNI · river in Khurda, Odisha, India
MPS MANZANARES · river in central Spain
MMU MAUNA KEA · mountain (volcano, site of telescope) in Hawai'i, US
MCY MILAN · city in Italy (airports have other specific names)
MPS MOSSÈN COSTA I LLOBERA · park (botanical garden) in Barcelona, Spain
NAY NAPLES-CAPODICHINO · airport in Naples, Italy
NRI NARMADA · river in central India
NPF NATURE ET PAYSAGES · park (botanical garden) in Gers, France, specializing in carnivorous plants
NFU NEMADJI · river and forest in Minnesota, US
NMF NÉOUVIELLE · mountain group (massif) in Pyrenees, France --or-- park (nature reserve) there
NFK NETLEY HEATH · forest in Surrey, England, UK
NSF NORMANDY · region of France (what else?)
NSR NORTH OSSETIA-ALANIA · republic in Russia
NCG NUREMBERG · city in Germany
OCG OBERHAUSEN · city in Germany
OSI ODISHA · state in India
OSU OHIO · state in US
OSJ OKINAWA PREFECTURE · prefecture in Japan
OFK OLDMAN WOOD · forest in Scotland, UK
OLY OMODEO · artificial lake in Sardinia, Italy
OSR OMSK OBLAST · federal subject of Russia
OSU OREGON · state in US
OAI OSMANABAD (OMN) · airport in Maharashtra, India
SFJ SAGANO BAMBOO · forest in Kyoto, Japan
SIR SAKHALIN · island in Russia
SIS SANCTI PETRI · island in Cadiz, Spain
SRY SANTERNO · river in Romagna, Northern Italy
SAR SHEREMETYEVO · airport in Moscow, Russia
SIJ SHIKOKU · island of Japan
SMY SOBRETTA · mountain in Lombardy, Italy
SMG SOIERNSPITZE · mountain in Soiern Group, Bavaria, Germany
SRG SPREE · river in Saxony, Brandenburg & Berlin, Germany and Czech Republic
TFJ TADASU NO MORI · forest in Kyoto, Japan
TFJ TAIHEIYO EVERGREEN · forest region in Japan
TIK TIREE · island and airport in Hebrides, Scotland, UK
TAJ TOKUSHIMA AWAODORI · airport in Japan
TCJ TOKYO · city in Japan
TMK TOM A' CHÒINICH · mountain in Scotland, UK
TCR TOMSK · city and oblast in Russia
TCJ TOYAMA · city and prefecture in Japan
TMU TRESSIDER · misspelt mountain in Yosemite, California, US

Here

we have all those place names, in their originally-given alphabetical order, one per line; the three-letter code at the start of each line consists of (1) the first letter of the place name, (2) the kind of place it is (Airport, City, Forest or wood, Island, Lake, Mountain, massif, etc., Park or gardens, River, State, prefecture, etc.), and (3) what country it's in (France, Germany, India, Japan, the United Kingdom, Russia, Spain, the United States, ItalY). My annotations mention other possibilities for some of these places, and in fact in one instance they fail to mention what jafe has now revealed was intended as the correct assignment :-).

And now

it seems that the obvious thing to try to do is to put the places into the grid so that no row, column or 3x3 box contains two with the same initial letter, the same feature-type, or the same country.

It's OK that

some (letter, feature, country) triples are repeated, because in each case one of the repeated ones is already in the grid so it should be possible to disambiguate them. (And it turns out that it is.)

Here are

the three component sudoku (initial letters, place-types, countries):

O G S | A T L | M C N    S C I | F M L | R P A    R G J | K U F | I S Y
N M A | C O S | T G L    F P L | A S R | I M C    U S I | R J Y | K F G
T L C | M N G | S O A    M R A | I C P | F S L    K F Y | I G S | J U R
------+-------+------    ------+-------+------    ------+-------+------
S T N | L G A | C M O    R A M | P F I | L C S    G J F | S K U | R Y I
M A L | O S C | N T G    C I S | L A M | P F R    S I U | Y R G | F J K
G C O | N M T | L A S    P L F | S R C | M A I    Y R K | F I J | U G S
------+-------+------    ------+-------+------    ------+-------+------
A O T | S L M | G N C    I S C | M L F | A R P    J U R | G Y K | S I F
C S M | G A N | O L T    A M P | R I S | C L F    I Y S | U F R | G K J
L N G | T C O | A S M    L F R | C P A | S I M    F K G | J S I | Y R U

Credit where due:

Jens found essentially all the same geographical information as I did, and his assignments were a little closer to the intended ones than mine. He could have done the last step just as easily as me. (Our work was independent, though.) Go upvote something of his if you liked this.

| improve this answer | |
$\endgroup$
  • $\begingroup$ Given your observation that having a few "triplets" doesn't matter if one of them is already in the grid (should have seen this myself), I suspect we are on the right track. It only remains to fiddle the geographical assignments correctly. As you say, a rather tedious endeavour. ;) $\endgroup$ – Jens Feb 18 at 18:34
  • $\begingroup$ I'm tempted to write a program to do it. And yes, I'd be extremely surprised if this weren't the right track. The only reason I haven't written the program already is the thought that maybe jafe intended the assignments to be "obvious" and didn't intend solvers to have to try lots of possibilities and attempt to solve the sudokoid for each one, which does seem like a lot of fairly unrewarding work. Maybe there's something clever one can do to reduce it, though it's not clear what. $\endgroup$ – Gareth McCaughan Feb 18 at 19:03
  • $\begingroup$ Yep, you're on the right track. This is definitely meant to be human-solvable and not require any deep try-this-if-not-then-roll-back chains, though. I commented about the intended labellings under Jens's answer. $\endgroup$ – jafe Feb 18 at 22:27
  • $\begingroup$ OK, very solvable with the intended labellings :-). $\endgroup$ – Gareth McCaughan Feb 18 at 23:55
  • $\begingroup$ All correct, nice work! Sorry about the ambiguity headaches, I did mean to leave some degree of freedom there but it turned out to be too much because I didn't check carefully enough. There are a lot of places in the world with the same name! $\endgroup$ – jafe Feb 19 at 0:28
5
$\begingroup$

Partial answer

I hesitate to call this a partial answer as it more of lack-of-progress report, but I do have at least one observation which must be relevant.

From the OP we see that there are 81 words divided into groups of 9 words starting with one of the letters A, C, G, L, M, N, O, S, T. Assuming the starting letters are relevant for the sudoku, and given that the solution is unique, we have the problem of which e.g. A word to place in a spot requiring an A word. There must be another selection criteria.

I therefore looked at what type of "thing" each word was. After a bit of squeezing and tucking (some words fit more than one category) I managed to fit them into one of 9 categories, with 9 in each.


enter image description here

The problem though, is that some words starting with the same letter also have the same categorization. How then to chose?

I then looked at

which countries each word was located in. Turns out the words are neatly divided into 9 words in each of 9 countries!

Expanding my table above, we have


enter image description here

However, again we have the problem that

some words starting with same letter are also from the same country. Again, how to chose?

And even if we decide that all three categorizations are relevant, we have several cases where all three are the same, e.g.

CAMPO DEL MORO and CASA DE CAMPO both start with the letter C, are both in Spain and are both parks.

So, to sum up:

The observation that all words can be neatly divided between 9 countries cannot be a coincidence. Whether the geograhical categorization is relevant is debateable, but it is striking that such a categorization is possible. Perhaps there is a fourth way to categorize?

| improve this answer | |
$\endgroup$
  • $\begingroup$ Heh. I am just in the middle of writing up a similar lack-of-progress report. It will be interesting to compare your assignments with mine. $\endgroup$ – Gareth McCaughan Feb 18 at 18:02
  • $\begingroup$ (On the one I'm most doubtful about, we agree.) $\endgroup$ – Gareth McCaughan Feb 18 at 18:02
  • $\begingroup$ @Gareth I'm also looking forward to your lack-of-progress report! $\endgroup$ – Jens Feb 18 at 18:07
  • $\begingroup$ Aha, we do have some differences. Give me a couple of minutes and I'll try it out with your choices... $\endgroup$ – Gareth McCaughan Feb 18 at 18:13
  • $\begingroup$ Ha, I have a mistranscription in mine, which probably explains why I also have a different choice in one place. [EDITED to add:] No, wait, it's not a mistranscription, just a different legitimate choice, sorry. $\endgroup$ – Gareth McCaughan Feb 18 at 18:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.