Wordle Is a Love Story | Hacker News Reader

Wordle Is a Love Story

nytimes.com

140 pointsjsvine122 comments

Open articleView on HN

- Fun

- Simple

- One puzzle a day, and everyone plays the same one so you can share and compare scores

- No account creation, logins, ads, subscriptions, data mining

All the right ingredients for going viral. I just hope the last point stays unchanged.

I hope the third point stays unchanged too. Too many services out there optimize for engagement, the expense being the encouragement of addictive behaviors.

Everything to do with the game is hard coded into the client. I doubt there's any plan to change that. But I wouldn't be surprised if clones popped up that tried to maximize engagement/revenue.

I've already found a clone that lets me play as many words as desired with as many characters. The only downside is that I've burnt myself out two days after discovering Wordle.

Agreed. Only one word a day makes me take my time and use the fewest guesses possible instead of brute forcing it by reusing dead letters. It feels more like a puzzle than a game, and the novelty lasts longer. I don’t think it would be getting the traction it has if it allowed endless play. It’s a throwback to the daily rebus/crossword/quiz newspaper page.

Without ads or logins to sell information from, what’s the need to maximize engagement?

I've seen at least one FOSS project chasing engagement even with little to no financial incentive.

Notoriety and "number go up" is enough incentive sometimes.

- enriching for non native speaker. (Not so easy to come up with 5 letter correct words.)

- Not banal

4 guesses today. First time it hasn't taken 6 lol

I feel bad that he isn't making a dime from this. I think people should be able to get something in return for creating value for others.

Maybe they enjoy non-monetary rewards? One of the joys of the internet for me is experiencing peoples' self-less creations. More of that please!

I think the inventor of the internet shouldn't have to work another day in his life. In our society that means money.

This sounds like the writing prompt for a Onion.com article that begins with "local man".

Did you actually read the article? It sounds like he's getting much more of a reward than money.

Is this the setup for a new genre of joke, the punchline being The Capitalists!” instead of “The Aristocrats!” ?

So when is Scott from HQ Trivia gonna start emceeing this?

The game's great but I'm puzzled why no one mentions that this is a verbatim copy of Lingo, a 1987 tv show that's been adapted in many countries. I used to watch this in France where it was very successful. Once people got too good at 5-letter words, they ran 6, then 7, and finally 8 letters.

It's obviously very standard (and has been done a million times before and after Lingo... probably as long as language has existed) but it's all about the execution here.

Nine days ago :)

> Fun fact: this concept is also the basis for a game show on Dutch national TV. I took part two years ago, and spent way too much time optimising our strategy: https://vincenttunru.com/hacking-a-gameshow/

https://news.ycombinator.com/item?id=29713212

And we even went up to ten letters! (Albeit in a slightly different format.)

It goes back even further, to a word game from 1955, though I'm sure it's been invented earlier. https://en.wikipedia.org/wiki/Jotto

Everything is a remix!

What’s your optimal 5-letter starting word?

Mine is HOUSE. It has 3 vowels (including E which is the most frequent in the English language) and S which helps test for plurals.

I always begin with PENIS. Maybe not optimal statistically, but emotionally.

In the 1960's my dad programmed Jotto (a simpler five letter secret word game) on Kodak's computers. Entropy became the first interesting mathematical concept that I learned.

Log base 2 of the remaining words is a measure of how many yes/no questions it would take to identify the word. An entropy strategy looks for a clue word that minimizes the expected value of this measure. One optimizes sum p log p over the pile sizes.

Pure mathematicians prefer certain concepts with a religious fervor. Often this has been informed by a reasonable number of problems where a concept has been proven optimal. The best applied mathematicians understand pure math but prefer practical work. To a pure mathematician, the rest are just guessing.

Here, one needs a clearly stated objective function for measuring success. Entropy strategies are often optimal for simple objective functions.

A critical detail for this game: The secret words come from a shorter list than the valid guess words. One wants a guess word that best partitions the shorter list of secret word candidates, not the full list of valid guess words.

Hah funny to see other people strategizing like this..

I did some research and (from what I found) the most common letters in English are:

E, O, T, A, I, R, N, H, S (not in any order)

So I came up with TRAIN and SHOVE as 2 starting words that use all those letters without repeats, plus V.

WORLD is also a good one because it uses one of the most common starting letters (W, T, A, O, D) and one of the most common ending letters (E, D, S, T).

Some other good starting words for me have been RIVET, CANDY, PLUCK and BASTE.

Thanks for the hint, I used PLUCK for wordle 202 and got a very good start

I used ORATE which was a hilariously bad guess, but surprisingly effective at pruning the search space in later rounds:

Wordle 202 3/6

  00000
  0?0??
  11111

No kidding, PLUCK has LUCK in it today. I went with PIANO first because I like to constrain the vowels, and got it in three.

The optimal starting word is ARISE, as it partitions the possible words most evenly across the different green/yellow/grey colour combinations

ARISE is good, but AROSE is better, and SOARE is the best: On average it eliminates all but 2264 out of ~12000 5 letter words.

https://github.com/lalaithion/wordle

I wrote a brute-force minimax solver (minimizing expected guesses) which tracks with this information:

  SOARE 3.45
  RAISE 3.46
  ARISE 3.47
  SERAI 3.52

Most 'reasonable' words seem within 0.1 or so of the optimal strategy. I think the second word is likely far more important than the first.

but soare isn't a word

It depends on your word list, are you using the same word list as wordle?

I do not. According to the original article, Wordle uses[1] 2,500 common words out of the 12,000 5-letter words in the english language[2]. I use the 5 letter words in the collins scrabble dictionary (which is about 12,000 words).

The assumption you need to make for my analysis to be correct is that the letter patterns in the 2,500 possible answers is statistically similar to the distribution of letter patterns in the original 12,000. There are probably some differences between the distributions, and I'd love to rerun my code with the actual word list Wordle uses, but in the absence of that list, I think that my code does about as good as possible.

[1] uses for the answers; I assume it allows all 12,000 for guesses. [2] NYTimes does not specify which source they used

Read the JavaScript? It contains both lists. Training on the wrong dictionary, tomorrow you might find yourself in a slump.

The word list is in Wordle code, so you can just grab that.

The answers are also in the code which opens the door to speedruns.

I like TRAIL - no "E", but the next two big vowels and TRL are very good consonants

What a coincidence. I played this for the first time today and started with TRIAL.

I did some analysis on this last week [https://noxville.medium.com/raising-the-wordle-first-guess-b...], ROATE left the lowest average valid answers (60.424) but 195 in the worst case. RAISE was the second best average (61.0) but a much better worst case of 168 words. Both are just heuristics, there might be a better word than either if the game were solved.

I wonder if ROATE is even in the list of words. I'd never heard it and I'd like to think I have a decent vocabulary.

The article says they whittled it down from 12,000 words to around 2,500 words, aiming for words that most(?) people would be familiar with.

Yes, it's in the list of guessable words - however it's not a valid answer word. You need to use the WORDLE word list in order to evaluate how good or bad your initial guess is, using another word list will provide distorted results.

Yes that's what I meant, is it valid in Wordle specifically.

That's what I was clarifying with 'Yes, it's in the list of guessable words - however it's not a valid answer word'.

They have two lists of words: one is a list of possible answers, and another is a list of extra (valid) words which can be guessed (in addition to the words in the answer list). Sometimes it might be better to use a non-answer word as a guess: the best case gets worse (since you cannot win immediately) but the average case and worst case both get better.

Ah I understand, thank you for clarifying!

I use "RENTS" which puts the S at the end as a quick test for plurals.

I've been using that as well, just one letter off of the classic R S T L N E

I coded some heuristics and ran them on a Scrabble dictionary: https://github.com/lalaithion/wordle

With the first word you aren't necessarily trying to get the most number of matches.

You are trying to use the word that once you get the match result back, it discards the most number of words.

These two are not the same thing.

> You are trying to use the word that once you get the match result back, it discards the most number of words.

This isn't strictly true either. Two N-sized subsets of words from a common initial set might have completely different difficulty in reducing further, because in the worst case there might not be a valid guess which nicely spreads the remaining words out among the 243 possible outcomes for that guess.

Set-size is a good heuristic, but it's just that - a heuristic.

Yes, as you'll see in the readme that I have numbers for both approaches. The first section is the average number of remaining words after getting the match result back, and the second section is the average number of yellow and green squares.

You might also try maximum instead of average. This is minimax and represents worst case scenarios for each guess.

This is mostly useful for optimal play against an opponent (which is not the case here). Imagine an adversarial version where the opponent doesn't have to commit to a word at the beginning but must reveal one matching all clues if you can't get it in 6 guesses (basically, they can change their word when you guess and you are trying to make that impossible).

As a handicap, I'm now beginning each new day with the prior day's answer word.

I start with ADIEU, personally.

the optimal 5 letter starting work is SERAI (as computed by someone's AI)

There was HN post a few weeks ago to a blog where someone computed SOARE as the optimal word.

Correction: a month ago - https://news.ycombinator.com/item?id=29439191

https://twitter.com/jgrahamc/status/1479189616846639110

I coded a small program, and the sequence of words that I use (not interchangeable, you'd use them in sequence to gather more info) is:

AROSE UNTIL DUCHY BLIMP GAWKS

I also really like JUMPY and INTER

STEAR - not technically a word, but it is accepted, and hits the most common vowels and consonants.

I think it's an archaic spelling of the verb "steer". I remember the edition of Treasure Island I had as a boy has Long John Silver use it in the line "We can stear a course, but who's to set one?" when they are discussing whether to mutiny.

Though when I google that phrase, Silver uses the modern spelling.

I've been using TEARS by same logic.

Mine was ouija, but house is a lot smarter, d'oh!

SAUCE for me

I always start with UPDOG.

What’s updog?

NOT MUCH WHAT'S UP WITH YOU

thanks

I always start with HAOLE

Last couple times I started with the word which rhymes with "tennis".

The only word I can think of that rhymes with "tennis" is "menace". That's 6 letters. But I think I know what you meant.

What accent do you have that those two words rhyme?

“It’s not trying to do anything shady with your data or your eyeballs. It’s just a game that’s fun.”

Love it.

The eligible word list is an array of 5 char words embedded in the script. After digging a bit it looks like it gets the word like this (pseudocode):

dayOffset = floor((today - 19 April 2021) / 864e5)

index = dayOffset % wordList.length

wordOfTheDay = wordList[index]

edit: It's literally just picking the next item in the array in sequence.

Seems like it would be important to note in the rules that letters can be used more than once even if they're green.

I’ve played twice now and had trouble. Today I had to resort to “cheating” using a scrabble word search site. Because I didn’t know that was allowed/possibly required.

Looking forward to tomorrow.

LOL, this sounds exactly like the Dutch tv game show 'Lingo' that's been on tv since 1989. Everything old is new again!

https://en.m.wikipedia.org/wiki/Lingo_(Dutch_game_show)

Which I recently found out originated in the US: https://en.m.wikipedia.org/wiki/Lingo_(American_game_show)

In France, we had Motus [0], which is the same thing but with longer words — and it's actually an adaptation of Dutch version of Lingo! It ran on France 2 for 29 seasons and is widely known in French popular culture.

[0]: https://fr.wikipedia.org/wiki/Motus_(jeu_t%C3%A9l%C3%A9vis%C...

This has become one of my favorites since I learned about it from NYT. It's more fun if you use the "hard mode" where your next word attempt has to include the right letters/spaces found in the prior try.

http://web.archive.org/web/20220106042510/https://www.nytime...

I was scrolling for this. Thanks colega

https://news.ycombinator.com/item?id=30004743

Show HN: Wordlet - a Wordle Bookmarklet to add change day button (just in case you miss some days :P)

This is an easier version of Bulls & Cows [1] game I played in college during boring classes.

[1] https://en.wikipedia.org/wiki/Bulls_and_Cows

I built a version of Bulls and Cows at https://cowbull.co/, mainly for my kids to play. One difference in Wordle is that when there is a match, Wordle shows which letters matched. In Bulls and Cows, it just tells the number of letters which matched, which makes it a bit more difficult in some ways.

Thanks for this - it looks it'll be ideal for feeding the craving after my partner and mine's daily Wordle!

Tip: don’t look at the source if you don’t want to see the answers (I thought the words might be obfuscated but they’re not)

I was wondering about this but I haven't looked yet.

...

So I just took a look now. Are you talking about the huge list of ~12,974 words embedded right in the code?

There are two massive lists, one is the wordlist and one is the actual answers in date order.

Yes, if like me you search the source for the word you just won with, you will see future answers, as they are listed in order (at least the preceding word was yesterday's answer)

ohhhhh, ok I see it now. I wasn't paying attention to order or even any of the other words.

I made sure not to look at the words after today's as I don't want to cheat.

The NYTimes spelling bee, which they said they were inspired by, also is unobscured in the source

I’ve been able to play a handful of the past words with internet archive for anyone interested

https://web.archive.org/web/20220105234455/https://www.power...

I can only see the top 2 rows of the keyboard. Adjusting the window size doesn't fix it.

I used devtools to delete archive.org's header div. But also, you can just type with your keyboard, including Enter and Backspace.

You can play any day by changing the date on your device.

I've been playing this as a road trip game since a friend introduced me to it in high school. Depending on the age of the people playing (and/or the fraction of concentration the driver needs to devote to driving) we also play with 3- or 4-letter words. 6 is a bit much for me to keep track of in my head.

I wish I could play a few more games at once and not be metered 1 per day.

https://www.fiveletters.xyz/

Allows for continuous play

Repo here https://github.com/sigvef/fiveletters.xyz (not mine)

Not mine, but see https://foldr.moe/hello-wordl/

Reset your clock to tomorrow's date, and play again.

That's how I confirmed that the solutions are in order, in the JavaScript.

Hah, how amusing. Client side self metering

Strangely enough, this is not the game for you: https://store.steampowered.com/app/1334000/Wordle/ However it looks fun.

If you happen to be on an Apple device, may I humbly suggest you look up "Three Magic Words" on the app store. It's a different take on the same root game, sometimes called "Jotto" or "Ghost."

Interestingly, according to the article, it was the decision to limit it that led to its popularity.

If you change your device's system clock then you can play as many times as you want.

If you want more than 1 Wordle a day, I forked his code and added a randomizer.

Same daily word, same dictionary of words, same code.

wordle.berknation.com

I tried this game the last couple of days and I don't get the appeal. There is some skill involved, but there is also a huge amount of luck. For example yesterday I had "ti_er". I guessed "timer" but the actual word was "tiger". That left me with a score of 5 instead of 4. Does that say anything about my skill level as a player? Honestly, this seems to mostly be a marketing success story about the sharing score functionality and how that led to virality on Twitter.

The appeal of vocabulary games is that they encourage you to flex your memory. They remind you of words you know that you might not use under normal circumstances.

As for the luck, sure, there's some of that. But you have plenty of guesses to get to the word if you play according to letter frequency, and always maximize the information you'll get out of a guess. Think of it like counting cards. The goal is to shift the odds as much in your favor as possible. Most games are like that. Deterministic, perfect-information games are a just a small subset of games.

But the goal isn't just to complete the puzzle. At least in my social circle, there is much more focus on the score rather than on just completing it. There are other word games that have a much better balance of skill and luck. The NYT Spelling Bee game mentioned in the article is one example.

My friend group doesn't focus on score, we just celebrate getting lucky or lament if we take until the last try.

There is a point about luck taking away from the longevity of the game. Unless the developer adds more game dimensions (in the same minimal, non-intrusive way) I kinda think most people will stop playing in a few months

Of course it's luck based - it's mastermind with a preset number of mnemonic arrangements. I struggle with it too, but the problem with systems thinking is that you don't get to enjoy the magic which drives the fad.

If one must judge performance in this game, then they shouldn't put much weight into the final score. That's why the share game button shows your progression. It tells a story that other players can understand. I learned to guess without using known correct letters by seeing games shared by friends. I know counting theory, I know Bayes, but it still wasn't a strategy I came up with on my own. The game is just something fun to do at a time where that is in short supply.

Put another way:

https://imgur.com/a/AkPPuYH

The skill is in not leaving yourself with two words that would both fit. If earlier you had tried a word with both an 'm' and a 'g' such as 'image' you would not have been left in a position where you would have to have a lucky guess

It is much easier to work backwards with that logic than forward. It is impossible to both try to guess the word while also trying to eliminate all possible overlaps that may occur with whatever results you get from the previous word. In this specific instance, I would have never played "image" because my first word was "tears" so I immediately knew there was a t, e, and r. There was no reason to try to determine if there was a g or m until I was presented with "ti_er".

It's only "luck" if you actually know all the words - most people do passively, but can they find the right one actively? That's part of the 'skill'. The other part is constructing a mental graph that allows you to find your way to the correct word within 6 steps, regardless of the word. An interesting question is whether there is actually any luck involved and how much. i.e. What is the optimal graph reducing the luck factor to a minimum, and is there a perfect one, if you know all words in the game?

People who say "it's just luck" either don't know, or don't care - but it's neither a correct, nor a clever position.

I went TIMER, TILER, TIGER and scraped in. I had been getting most in 3-4 attempts prior to that.

The aspect I don’t love is that in forcing dictionary words to be attempted, people can rely on that at the end rather than winning through word knowledge. My 9yo plays it and generally does well but was bailed out once by this.

I attempt to get at the fourth try. Before that is luck, after that is guesswork I like to believe that four is the skill sweet spot.

see also: Josh Wardle's talk on "Why April Fools' Day sucks and what we did about it at Reddit" – https://www.youtube.com/watch?v=OXzfvYoFQFo

naturally, clones have emerged

here is one I found today, you can have multiple tries, and chose word length

http://foldr.moe/hello-wordl/

wonder what other innovations occur before we round back into scrabble

https://octokatherine.github.io/word-master/

Another clone

https://news.ycombinator.com/item?id=29784601

Reply on news.ycombinator.com