Will anyone solve this crossword puzzle in one week?
21
370
390
resolved Sep 12
Resolved
YES

Unlike most crosswords, there are no blanks in the grid. Each hint clues two distinct words. My friends who helped playtest this are asked to not post the solution or bet. Market resolves YES when someone posts the correct solution in the comments, or NO if no one has done that by market close.

Get Ṁ200 play money

🏅 Top traders

#NameTotal profit
1Ṁ65
2Ṁ59
3Ṁ20
4Ṁ11
5Ṁ9
Sort by:

I'd love to hear your thoughts on this puzzle below!

@Shelvacu My thoughts are that clues 4 and 5 are ungrammatical. A clue should not indicate a plural when the answer is not a plural, and vice versa.

@asmith I struggled with 4 for a while, this was the best I could think of. Should I just add "(different plurality)" or something at the end?

I never thought about 5… Isn't it grammatical because "bath salts" is uncountable? Not sure

@Shelvacu You might be right

predicted YES

@Shelvacu Bath salts is fine. I don't know how dress fits as an answer though?

You could make the clue for 4 more abstract/cryptic, like "4: keep starting over"

Then it doesn't imply a grammar. Makes the clue a bit easier though but maybe that's okay; it's a difficult puzzle.

predicted NO

@Mqrius a better clue for 5 might be: "used by Victorian ladies"

@Odoacre Now we're talking! Except you would never say "dress are used by Victorian ladies", nor even "dress is used by victorian ladies". But good thinking.

predicted NO

@asmith maybe "associated with fancy Victorian ladies" ?

predicted NO

@Odoacre I dunno. There is definitely a thing as a Victorian lady dress

@Mqrius "gender euphoria" is something sometimes experienced by trans women when they put on women's clothing, like a dress. It's a stretch, like everything lol

@Shelvacu I'm pretty sure that the phrase gender euphoria crossed my mind at some point while trying to solve that clue, so it wasn't that much of a stretch.

predicted YES

@Shelvacu Oh right, yeah makes sense! That's fine then imo

bought Ṁ100 of YES

So this is my attempt, @Shelvacu can you say if this is correct or not?

bought Ṁ100 of YES

@JoaoPedroSantos dang this is good. How did you do it?

@JoaoPedroSantos That's correct! Congrats!

predicted YES

@Mqrius A mix of reading the community progress using bing (GPT-4) and old fashion puzzle solving. With the clues bing figured out bored, upper/opera, reset/tries. Then I fitted april, butts, salts/dress, at the end I had a word I didn't know "terse" but it was a word that existed and its description matched the clue.

predicted YES

alright i wrote a more heavily optimized solver, but without more guesses it's still not finding a good solution quickly. with no guesses it can churn out solutions like nobody's business, but they usually do not fit the clues. https://gist.github.com/brubsby/c92a3b530da5d5620fdc4e03822219d4


and here's the dictionary i've been using, i started limiting the words that came out of it to see if that helped, but no dice https://gist.github.com/brubsby/596e7a5ce57c14a4d9f7912c816bcc21

more guesses for positions would be most helpful from the crowd, and i might need to refactor for it to first attempt to exhaust all combinations of guesses first, before moving on to trying random words that fit. currently it starts on the row/col with the least amount of guesses, and finishes dfs-ing through all of the words there (including all words that aren't guesses), and then tries to fill in the rest of the puzzle dfs style (prioritizing guesses first). so it can get stuck in no guess land pretty quick. although if we had all of the words used in the guess list, it would find the combo pretty quick.

@brubsby So uh, I've done this before https://github.com/shelvacu/fwrf

I can finally say that tries are like linked lists; good in theory, terrible in practice on modern CPUs. Just use a hashmap. Unfortunately I think "reset" was guessed but not included in your list of guesses. And I tried to say it without saying it outright, but nvmbr and dcmbr were not possible because they aren't words in the wordlist I used.

predicted YES

wrote a program to try to brute force this from word lists, but not getting too many promising results, if anybody has a better word list you can try to plug it in, i think mine was from a wiktionary dump https://gist.github.com/brubsby/258107611dae337ef57145097584b1b3

predicted YES

@Shelvacu

Can you confirm all the words are real words in the dictionary?

predicted YES

https://gist.github.com/brubsby/acd4cbe0165596ef2329e635aca142a3 the candidate solutions, not great. i'll try not hardcoding the months maybe tomorrow

predicted YES

@Joshua hard to define what a "real" word is, so some overshoot is probably best when choosing a dictionary

@Joshua As brubsby said hard to define "real", but I am virtually certain that all of you will recognize all of the words and know what they mean.

predicted YES

https://gist.github.com/brubsby/a9105a393a6aea18ade068d7e1998129

version with less hardcoding, probably should put the months first though, running now but it's slow

predicted YES

anybody else feel free to use this code to solve it lol, a version that uses individual tries for each set of possible guesses might be good too.

predicted YES

or maybe trying guesses first and then the big trie of words

predicted NO

Edit: nvmd

Comment hidden