List animals until failure
112 points
by l1n
6 hours ago
| 38 comments
| rose.systems
| HN
monopoliessuck
39 minutes ago
[-]
I added Jellyfish and then Portuguese Man-o-war.

It took the man o war, but crossed out Jellyfish and said "added a vaguer term", but a jellyfish and a man-o-war are discrete animals.

The man-o-war is a colonial siphonophore composed of zooids, while a jellyfish is a singular marine organism.

They're both in the phylum Cnidaria, and that would have been a more vague term had I entered it.

reply
themk
1 hour ago
[-]
One of the few sites with a fun "you have javascript turned off" message.

> This game requires JavaScript. Or, if you've superior taste, take out a pen and paper and start listing animals.

reply
sublinear
4 minutes ago
[-]
I think that's meant to be a gentle insult, but I'm glad it had it's intended effect!
reply
kyle-rb
2 hours ago
[-]
reply
vunderba
2 hours ago
[-]
That and probably Sporcle. Name X from {group Y} is a very popular quiz archetype.

https://www.sporcle.com/games/jjjjlapine2nd/name-every-anima...

reply
apt-apt-apt-apt
1 hour ago
[-]
Lazy daisy:

  (async () => {
    for (c of 'red black white brown blue green yellow golden grey arctic mountain forest spotted striped'.split(' '))

      for (a of 'bear lion tiger wolf fox eagle shark whale snake frog cat dog horse bat rat mouse owl hawk duck crab ant bee spider deer penguin elephant rabbit'.split(' ')) {

        guessbox.value = c + ' ' + a;
        uncomment(); attempt();
        await new Promise(r => setTimeout(r, 75));
      }
  })();
reply
tptacek
2 hours ago
[-]
No LLMs is impressive. Also recognizes "drop bear". Well played.
reply
defrost
1 hour ago
[-]
The Australian Museum recognises them also: https://australian.museum/learn/animals/mammals/drop-bear/

Somewhat more impressively, it recognises bungarra .. although it stalls out and fails on other similar words for various local animals.

reply
wincy
1 hour ago
[-]
Hahah in a moment of desperation I put “unicorn” which its response made me laugh out loud.
reply
themanmaran
2 hours ago
[-]
68. The unique title texts are really fun. But I strongly disagree that "chipmunks are squirrels".
reply
jmtulloss
2 hours ago
[-]
This is what I'm saying. Chipmunks are not squirrels. I will do my best on this hill.
reply
pdabbadabba
1 hour ago
[-]
Similarly, it insisted to me that a pigeon is the same thing as a mourning dove. Not true! But your case is even more egregious.
reply
meatmanek
56 minutes ago
[-]
"pigeon" and "dove" are both words for the same family of birds. The bird most people think of with the word "pigeon" is the rock dove (https://en.wikipedia.org/wiki/Rock_dove) or domesticated / feral variants of it.
reply
skylurk
15 minutes ago
[-]
Yep, but importantly "pigeon" and "dove" are not exactly interchangeable words, there is just no consensus for which is which.

https://en.wikipedia.org/wiki/Columbidae

reply
bmitc
1 hour ago
[-]
There are some weird ones. It knew red-bellied woodpecker but not yellow-bellied woodpecker.
reply
samename
14 minutes ago
[-]
Fun results: outputted as emoji

30 animals listed 𓃱 𓃸

Edit: weird... emojis don't work here? how have I never known this...

reply
kevin_thibedeau
2 minutes ago
[-]
We can at least practice egyptology.
reply
the_fall
2 hours ago
[-]
It might be an interesting LLM benchmark: how many can they list without breaking the rules (repetition or non-animals). Although I bet that big bucks would be then thrown at pointlessly optimizing for that benchmark, so...
reply
bronco21016
2 hours ago
[-]
Might be an interesting problem for understanding how various models perform recollection of prior tokens within the context window. I'm sure they could list animals until their window is full but what I'm not sure of is how much of the window they could fill without repeating.
reply
helloplanets
2 hours ago
[-]
I guess it could be generalized to filling up the context window with any token, but just making sure none of the tokens repeat.

An interesting twist could be making sure a specific token is an anagram of the token N tokens back. This could possibly measure how much a model can actually plan forwards.

reply
OxfordOutlander
2 hours ago
[-]
reply
helloplanets
2 hours ago
[-]
For anyone wondering: This is based on basic text parsing and a key-value lookup table, no AI involved whatsoever.

Here's the table: https://rose.systems/animalist/lower_title_to_id.js

reply
kranner
2 hours ago
[-]
There must be another table. I got "Are you Australian?" for "dingo" and for "cicada" "don't you love their songs?"

edit: https://rose.systems/animalist/eggs.js

reply
jml7c5
2 hours ago
[-]
Hmm, what's this one?

  var h = hash(guess);
  if (h==7182294905658010 || h==6344346315172974) { return "Adorable guess, but it's spelled “rosy”."; }
I'm guessing they're hashes for "<something> rosie" or "<something> rosey", but what?
reply
moeffju
16 minutes ago
[-]
7182294905658010 "rosey maple moth" 6344346315172974 "rose maple moth"
reply
helloplanets
2 hours ago
[-]
Yea, there's a bunch of easter eggs. And then there's the table for the taxonomy tree.
reply
Retr0id
2 hours ago
[-]
There's a long Indonesian string referencing "weasel" - weird https://www.wikidata.org/wiki/Q28521

Edit: someone edited to remove it just this minute!

Introduction: https://www.wikidata.org/w/index.php?diff=2353193943 - just random vandalism I suppose.

reply
lepolas
2 hours ago
[-]
This was fun. I definitely could feel the fatigue slowing me down until the timer got me. I also wasted a bunch of time trying to spell specific animals like the wobbegong.

I like the emoji output as well: 203 animals listed:

𓃬𓆊 𓃜𓃘𓅱𓆉𓅃

reply
cellis
1 hour ago
[-]
79. I feel like i should have done better but got stuck in a local minima of "farm animals, which obvious farm animals haven't I said??", then tried thinking of names of fish which worked until it didn't.
reply
maxbond
2 hours ago
[-]
I got 42. I was very impressed by how it handled more and less specific categories. It also understood rotifers were a microscopic animal, which I half expected not to work. Great project.
reply
hyperno
33 minutes ago
[-]
110. my strategy is to use pokemon to remember animals. (i can list all 1025 pokemon from memory)

not a great strat, though. (tons repeated animals)

reply
jammaloo
1 hour ago
[-]
267, I was going pretty strong and had about 2 minutes racked up, until I hit a wall, and couldn't think of anything else. Thinking in groups helped the most, e.g. reptiles, flightless birds, african animals, etc.

Extinct animals also work, including the dinosaurs!

reply
adt2bt
2 hours ago
[-]
157. Very neat! Started a, b, c then found much more success when thinking about biomes (sea, mountains, forest, jungle, etc).
reply
locusofself
1 hour ago
[-]
nice score! I got 129, and yes, same experience.
reply
divbzero
2 hours ago
[-]
Instead of trying to think of just any animal, I found it easier to add a constraint…

1. Animal that starts with A

2. Animal that starts with B

3. Animal that starts with C

(I also appreciated the easter eggs: “Are you Australian?” and “You listed both dingos and dogs, so I gave you the benefit of the doubt, but there's disagreement on whether the dingo is its own species of canid, a subspecies of grey wolf, or simply a breed of dog.”)

reply
2muchcoffeeman
2 hours ago
[-]
Without considering if it’s a distinct species, a dingo is descended from the same wolf population as dogs.

They are feral dogs. IE wolf -> domesticated dog -> became wild again.

reply
bmitc
1 hour ago
[-]
I went by groups and families of animals.
reply
teaearlgraycold
6 minutes ago
[-]
Countdown timer seemed stuck at 1:16 and then was suddenly big red and almost out of time?

66 animals listed 𓃬𓃰

reply
locusofself
1 hour ago
[-]
129 here, not bad for the end of the day. I listed a surprising number of dinosaurs, and of course, edible animals, and Lion King stuff.
reply
khazhoux
46 minutes ago
[-]
Warthog! Meerkat!
reply
yuppiepuppie
1 hour ago
[-]
reply
dudewhocodes
2 hours ago
[-]
Accepts the word "human" as well.

update: Start with "human" or "homo sapiens" and the website keeps changing as you add new words.

reply
troyvit
1 hour ago
[-]
I said "ape" after and it wouldn't take it because I'd said human already :)
reply
hillcrestenigma
2 hours ago
[-]
I like how it triggers an overlay when you try "human", it's a nice touch
reply
nxtfari
2 hours ago
[-]
there are so many special cases that this game responds to, i was shocked that it didn’t involve an llm at all at the end. very cool.
reply
ViscountPenguin
2 hours ago
[-]
Possum => opossum is erroneous. I was planning to list both species, it wasnt a typo.
reply
khazhoux
44 minutes ago
[-]
130

Fun game, but also a fascinating brain-probe. We're not used to reversing our internal classifiers.

reply
echelon
2 hours ago
[-]
106, I feel like I should have done much better. (I feel like I cheated by naming a lot of dinosaurs and insects.)

The clown emoji is great. :)

reply
alterom
2 hours ago
[-]
Another 106 here. I got to the point where I was just blanking out looking at the countdown timer.
reply
schnaars
1 hour ago
[-]
This is awesome. I made it to 100, but I know jack about animals.
reply
rogual
2 hours ago
[-]
140. Good fun. I like how it teaches you things, too. I learned that toads are considered frogs, axolotls are salamanders, and that it's "anemone" not "anenome". If you type in Unicorn it accepts it as "Unicorn spider" with a fun message. Don't forget to think of insects, birds and fish too, all of which it accepts. I love this kind of detailed, handcrafted thing that someone put a lot of time and effort into.

If you wanted to develop this more, some fun features might be telling you the most commonly entered animals you missed and the most unusual ones you thought of. Appreciate you probably want to keep it a static site though.

reply
Jordan-117
58 minutes ago
[-]
205! The running commentary was fun. And I love how permissive it is -- it was fun stumbling into a new category that you wouldn't necessarily expect to qualify. I do wish that there was an option to see a list of the most popular ones you missed (based on traffic to the article or similar).

For a similar brain exercise, try to Name Every City:

https://cityquiz.io/

reply
cvhc
2 hours ago
[-]
49 as ESL speaker. Fun little game to practice English words :)
reply
agnishom
52 minutes ago
[-]
I got 48 animals
reply
ivanech
2 hours ago
[-]
this was really delightful. The Easter eggs in particular made it feel like someone was actually on the other side
reply
TZubiri
2 hours ago
[-]
The background is alternating between cyan and black, which is very distracting. Not sure if that's on purpose.
reply
disillusioned
1 hour ago
[-]
205 and I very much was scraping the bottom of the barrel at the end. Starting a bit generic and adding specificity helped a lot. The little meta-commentary was great. "you already said dogs. dogs are dogs." when I tried "golden retrievers" after already typing dogs.
reply
Buildstarted
2 hours ago
[-]
drop bear => Already said Koala. but if you type it before you say koala the answer drops from the top of the page. so many great easter eggs. got 92 in the end
reply
pilaf
2 hours ago
[-]
I entered plankton, which technically isn't an animal and so it rejected it like any other random word, but then after I lost it offered me a link to the Wikipedia article on plankton. Very thoughtful.
reply
jojobas
1 hour ago
[-]
Some animals replace vaguer definitions, but say "snake" is there forever, no cobra or viper for you. Strange.
reply
bitwize
2 hours ago
[-]
ARE YOU THINKING OF AN ANIMAL? yes

DOES IT SWIM? no

IS IT A BIRD?

https://www.atariarchives.org/basicgames/showpage.php?page=4

reply
mberning
2 hours ago
[-]
It accepted tardigrade which I thought was interesting.
reply
pilaf
2 hours ago
[-]
It pulls its data from Wikidata, which is very thorough, so I'd be more surprised if you managed to enter an animal it didn't know.
reply
squibonpig
1 hour ago
[-]
Lmfao black widow turns the title black I think
reply
kalmyk
2 hours ago
[-]
lots of animals
reply
Cyphase
2 hours ago
[-]
95. Definitely feel like I should have done better; will try another time with more sleep.
reply