Talk:Wordle

Question about sources

I noticed some probably questionable sourcing in Wordle § Adaptations and clones:

• Source 76, after the sentence Nerdle, which involves solving an 8-digit mathematical equation, doesn't mention the game. Out of the three sources in Nerdle, two are unreliable and the other isn't listed at WP:RSP, so I'm not confident enough to change the source.
• The statement about Framed is unsourced, even before the GA review.

I wasn't able to find suitable sources to these after an admittedly cursory search.

ItMarki (talk) 14:53, 6 June 2025 (UTC)

I think these are fair concerns which I'm honestly surprised I never caught. I've removed both parts as unnecessary and unfairly sourced. Nub098765 (talk) 08:52, 8 June 2025 (UTC)

"Found"

She also avoids words that are particularly difficult to deduce due to having too many common letter combinations—such as found, where the first letter could be one of eight possibilities.

In the interview cited, the editor doesn't say she avoids those words, and in fact if you play Wordle you know that such words are not avoided. (Good morning! See today's puzzle.) Look up how many times the answer has been an -ATCH word. What she says is that for word groups in which changing only one letter produces a number of answers that exceeds the number of guesses, if you get it right, "that’s based on luck, rather than deduction." Cynwolfe (talk) 14:12, 12 October 2025 (UTC)

The sentence pretty much captures the source, but the only thing I've changed is to say "tries to avoid" as opposed to "avoids". Today's is certainly not the first such case. Masem (t) 14:33, 12 October 2025 (UTC)

Apologies, but it doesn't capture the source.

I saw this statement in the article the other day and found it puzzling, and this morning when my routine guess 1 yielded no yellows, I mischievously used one of the -OUND words on guess 2 with this claim in mind because it was so obviously false. (So thanks for the misinformation, because otherwise I probably wouldn't have got it right on guess 3 today.)

Could you quote the sentence that says she avoids or even tries to avoid such words? It's a longish interview, so perhaps I'm skimming over it. Cynwolfe (talk) 15:34, 12 October 2025 (UTC)

This is the sourced paragraph: Q: What are you thinking about when you choose each word? A: It’s more that I’m authorizing them. First of all, I usually use a random number generator, some kind of random process to pick the words. And then I research them, so I will look through dictionaries and Google, and I’ll try to find out if there are any, like, secondary meanings that are maybe either profane or derogatory. And I wouldn’t want to run a word like that. I also run through letters, so if you have a word that has very common letters, and you have four letters in place — like, for example, blank-o-u-n-d. That fifth letter — found, mound, round, sound — there’s like eight of them, I think. There are only six guesses so even if you have already figured out four correct letters, you still have more letters to guess than there are. So, that’s based on luck, rather than deduction. Its clear this is a process to avoid certain words, but as you state, its by no means an absolute. Masem (t) 15:53, 12 October 2025 (UTC)

Well, I read that paragraph, and I had hoped not to have to spill a lot of words to point out that it doesn't actually say that she avoids those word groups. If it did, it would be counter-factual because words formed on -OUND, -ATCH, and the like are in fact used regularly. She's describing the process as basically random, with some editorial discretion. The only commonly used words she says she avoids are those that may be considered "profane or derogatory." She says she "wouldn't want to run a word like that." That is a clear principle that she states with surety, though of course it can be subjective as to whether a word is "profane or derogatory." (I don't expect "bitch" to turn up in Wordle even though it has a non-derogatory meaning, for instance.) Those words are indeed avoided. They don't turn up as answers. But then she follows with a transitional clause: "I also run through letters." This is a return to the initial inquiry about her thought process in authorizing (not choosing) words. What may be implied (and not stated, so I wouldn't put it in the article) is that if one of those kinds of words turned up after another had been used recently, she might bump it to another time so as not to cluster-frustrate players. But these words are obviously not avoided, as any list of previously used Wordle answers shows. So there are two sourcing issues here. If the editor had actually said, "I try not to use those words", the statement would be counter-factual, as could be proven with any number of sources listing the answers or discussing the frustration of such answers. The editor is a primary source speaking about herself, not a secondary source — the interview simply reproduces what she says without further context, verification, or clarification, as should occur in a better-developed narrative article. However, the second issue is that she doesn't actually say she avoids those word groups. She acknowledges that those answers depend on chance rather than skill. Wordlebot itself attempts to factor "luck" into the stats (I would call it intuition), so the greater role of luck in guessing certain word groups is not an exclusionary principle in the way that vulgarity is. My conjecture (and of course it's hazardous to bring our own assumptions into reading a source) is that she considers whether it's time to let one of those in to keep it challenging. Like today, or like when you get the feeling that it's been a long time since the answer had a Q or J, so you start expecting one to turn up. I wouldn't be surprised if today's answer was a little dig at Wikipedia for misleading readers into thinking that this particular word group was somehow verboten – despite the ascertainable fact that FOUND, the word given in the 2023 interview as an example, had been an answer on March 28, 2022. I would expect a GA article not to contain a statement that's so wobbly on both accuracy and factuality. Cynwolfe (talk) 17:02, 12 October 2025 (UTC)

Hi! I'm the GA nominator. Apologies for opening this discussion back up—it fell through the cracks of my ever-flowing watchlist, and I feel the need to write a response. I quite agree with your assessment. Perhaps it was my own lack of rigor that produced such a "wobbly" statement. I'm in accordance with changing it somehow (though I'm not exactly sure how you would without making it a mess of a sentence). I would, though, like to defend my original assessment; it is exceedingly obvious (to me, at least) that under her diplomatic tone is an implicit "I try to avoid those words" (because why, then, would she even bring it up?). But of course, personal evaluation is not enough for Wikipedia. Thank you for the feedback, and if no one objects, I will try to amend the inaccuracy. nub :) 23:01, 19 October 2025 (UTC)

Spelling variants

Can it be made clear in the text that the original version was with British English spelling, as would be expected for a game developed by British people. It changed to US spelling when it was bought by the NY Times. This isn't made explicit - just a reference to 'some British spelling variants' being removed. Stronach (talk) 10:26, 4 April 2026 (UTC)

Were American-spelled words ever not in the list of possible answers? This article mentions British words being removed, but not American words being added and I can't seem to find any evidence that states American words were added when NYT bought the game. I think saying that some British spelling variants were removed is correct, and additionally we should revert edit 1353781216. Purpleduckfish (talk) 23:49, 13 May 2026 (UTC)

The NY Times has never added a "guessable" word (correct + incorrect but real lists) and they (or more accurately she, Tracy Bennett) did not explicitly add any new words when removing the British ones as American spellings were already in the "correct" list, though since a few months after their takeover words are no longer from a predefined list but rather a dynamic API (and they have used words not in the previous "correct answers list")

I've talked about this in the release notes of Wordle DS (my clone for the Nintendo DS), but I don't know if that'd be considered a reliable source. Pk11 (talk) 03:11, 14 May 2026 (UTC)

5th anniversary and start date

On June 19, 2026, the New York Times is offering a special 5th birthday badge. The puzzle number is listed as game 1826, and accounting for 2024 being a leap year, 5*365+1=1826. So puzzle number 1 would have been June 20, 2021. However no date in June is currently mentioned in the article. According to Wordle answer websites, Wordle 0 was on June 19, 2021. The issue here is how to source this. Larrybob (talk) 14:44, 19 June 2026 (UTC)