3 ms·

I don't find the "limit possible solutions to the 2315-word list" problem interesting. Is there any reason to think this isn't just the first 2315 words of a la

I don't find the "limit possible solutions to the 2315-word list" problem interesting. Is there any reason to think this isn't just the first 2315 words of a lazily-evaluated ordered sequence of all possible solutions? Presumably the complete list of possible solutions would be "all 5-letter combinations one could reasonably argue are English words" minus any words considered profane or otherwise inappropriate.

jonathankoren 5y ago
The main problem I have with these optimal solutions isn’t that they’re using the target dictionary, it’s that they’re taking advantage of the fact that there is a dictionary of valid words, and smaller dictionary of target words. That’s why none of them solve it in one guess.
Thorrez 5y ago
The game itself prevents you from guessing words that aren't in the 12972-word list. It doesn't count as a guess, and doesn't go through. So if you want to try anything with a larger list, you're playing something Wordle-like, not actual Wordle. But I agree that using the 2315-word list is basically cheating. On the other hand, are words in that list more common English words than words in the 12972-word list? If so, then humans in a sense intuitively take advantage of the 2315-word list by guessing more common English words rather than infrequent English words. So since humans sort of use the 2315-word list intuitively, it sort of makes sense for a computer to use it as well.
- cookie_monsta 5y ago
  > are words in that list more common English words than words in the 12972-word list? Subjectively, yes - those are the 2315 words that the dev's partner recognised out of the 12972. https://pressnewsagency.org/wordle-is-a-love-story https://pressnewsagency.org/wordle-is-a-love-story
- NathanOsullivan 5y ago
  I agree, the answer list is clearly curated to avoid unusual words. This leaves a fuzzy middle ground for human players: there are accepted guess words like SOARE that can be good as a first guess, but is almost certainly not going to be the answer. Conversely, FAVOR is not a good first guess; but of the two words human players know which is more likely to be the answer. For computer implementations that want to avoid "cheating" (not use the answer list), there would still seem to be room for evaluating how likely a given guess is to be in the curated list. That could be done by examining a corpus for frequency of each guess word, which if done right should give the same insight we have as humans.

jonathankoren 5y ago

The main problem I have with these optimal solutions isn’t that they’re using the target dictionary, it’s that they’re taking advantage of the fact that there is a dictionary of valid words, and smaller dictionary of target words. That’s why none of them solve it in one guess.

Thorrez 5y ago

The game itself prevents you from guessing words that aren't in the 12972-word list. It doesn't count as a guess, and doesn't go through. So if you want to try anything with a larger list, you're playing something Wordle-like, not actual Wordle. But I agree that using the 2315-word list is basically cheating. On the other hand, are words in that list more common English words than words in the 12972-word list? If so, then humans in a sense intuitively take advantage of the 2315-word list by guessing more common English words rather than infrequent English words. So since humans sort of use the 2315-word list intuitively, it sort of makes sense for a computer to use it as well.

cookie_monsta 5y ago
> are words in that list more common English words than words in the 12972-word list? Subjectively, yes - those are the 2315 words that the dev's partner recognised out of the 12972. https://pressnewsagency.org/wordle-is-a-love-story https://pressnewsagency.org/wordle-is-a-love-story
NathanOsullivan 5y ago
I agree, the answer list is clearly curated to avoid unusual words. This leaves a fuzzy middle ground for human players: there are accepted guess words like SOARE that can be good as a first guess, but is almost certainly not going to be the answer. Conversely, FAVOR is not a good first guess; but of the two words human players know which is more likely to be the answer. For computer implementations that want to avoid "cheating" (not use the answer list), there would still seem to be room for evaluating how likely a given guess is to be in the curated list. That could be done by examining a corpus for frequency of each guess word, which if done right should give the same insight we have as humans.