JustOne/InterestingEdgeCases

The rules about what constitutes a clash in Just One have some interesting edge cases which can give rise to interesting linguistic and philosophical discussions and/or heated arguments.

Foreign languages

Just writing the target word in a foreign language is not allowed, but synonyms in English are allowed, and foreign words that clue the target word but are not a direct translation of it are allowed.
This isn't as straightforward as it first looks.

Because languages aren't one-to-one mappings of each other, there's room for debate about whether a word is a direct translation or just a related concept.

Alex's mum thought "H2O" for "water" shouldn't be allowed because it's writing the word in a different language: the language of chemistry. Not a perspective we'd considered, but fairly hard to argue with.

Anne says it clearly means just natural human languages. But Bethany pointed out the example of ciphers: you couldn't just write the word in Morse code or rot13, even though those aren't natural languages. (Morse might be diallowed by strict application of the hyphen rule, but rot13 still stands, and you could contrive examples that could be spelled with just dots in Morse.)

: Morse code and rot13 aren't languages at all: they're encoding mechanisms. If the rule is phrased something like "You must clue the word rather than just translate or encode it" then that would unambiguously rule out simple Morse or rot13 application, and would tend to rule out "H2O" while not being unambiguous about it. (Compare "NaCl?": pretty clearly cluing "salt", but the word "salt" is far less narrowly tied to a single substance than the word "water". If I were to ask someone to name five salts then I would expect answers like table salt, copper sulphate, silver nitrate; whereas if I asked for the names of five waters I'd expect answers like Evian and San Pellegrino. On the other hand, less obvious answers might include Coniston, urine, rosewater, etc.).
: Those are already mostly not allowed because a clue has to be a word or a symbol or a number or an abbreviation: you are not allowed arbitrary strings. This doesn't fully cover the issue, though: sometimes the ROT13isation will happen to be a word, while doing something like 0x7768617465766572 always passes that condition. -Nozaquirce

Someone might clue "salsa" for "sauce", just because salsa is a type of sauce (like you might clue "hollandaise"), but not realise that it's Italian and Spanish for sauce. Or sometimes people clue "Simba" for "lion", but since playing Just One I've been informed that "Simba" is Swahili for "lion". Should these be allowed? Some people rule that they're allowed if you didn't know the etymology. That's getting into infohazard territory: if you don't have that linguistic knowledge, you have an advantage that you lose if you gain the knowledge. Another possible solution is to say that the intended connection (salsa is a kind of sauce, Simba is a lion) overrides the no-foreign-translations rule, and the no-foreign-translations rule only applies if that's the only connection. I think that's a pretty good solution, but potentially open to abuse: someone could give an obvious direct-translation clue like "fromage" for "cheese", expecting the guesser to get it on those grounds, but keeping plausible deniability because they know of some cheese shop called Fromage or something. It wouldn't even have to be deliberate abuse: you might overestimate how widely known your favourite local cheese shop is.

Also, there isn't a clear-cut line between English synonyms, foreign loan words in English, and just foreign words. When the word was "goodbye" I avoided cluing "adieu" because I thought it would probably be ruled a foreign translation, but it's also an English loan word. There are a lot of French and Latin words that exist in English as obscure/fancy/technical synonyms for the English word or subcategories of it, like "chemise".

Clive recently challenged "buss" for "kiss" because he didn't know "buss" was a synonym for "kiss" in English, whereas I was completely confused by the challenge because I didn't know that "buss" was the word for "kiss" in German. Presumably the former comes from the latter, but to what extent is it an English word in its own right?

Acronyms

Acronyms clash if one of the constituent words clashes, so "BBC" would clash with "Broadcast".
Does this apply recursively? This comes up when playing with computer geeks. I avoided "TCP/IP" for "net" because the expansion contains "internet". (Please insert better examples of multi-level acronyms if you can think of them.)
What about ex-acronyms? I think it was Edwin who argued that "laser" shouldn't clash with "light", because the acronym is only the historical etymology of "laser", and it's now just a word in its own right.

Coincidental clashes

Real-life example: "glass" (the material) and "hourglass", as clues for "sand". You kind of have to rule that they clash, but you also feel they shouldn't, because they're completely different meanings of "glass" and it's just an unfortunate coincidence that they clash.

"Different meanings of 'glass'"? They certainly have different links to "sand", but presumably an hourglass is so called because it is made of glass? ...[Yes, that is the etymology.] -Nozaquirce

: Maybe "meanings" wasn't quite the right word. But it's just an unfortunate coincidence that a material made from sand and an object that contains sand have related names. They're not both using the same link between "glass" and "sand". --Rachael

There's also homographs: "wind" (as in breeze) and "wind" (as in spiral) could both be clued for "tornado", and, again, they would have to clash but it feels like they shouldn't.

Coincidental clashes involving acronyms are more uncertain. I guess "bash" (the verb to hit) and "bash" (the Unix shell) clash with each other, for the same reason that "wind" and "wind" do. But the latter would clash with "shell" and the former wouldn't! Presumably you can't reject a clue just because it has an alternative acronym meaning that the clue giver wasn't aware of?

Numbers

Alex pointed out that "3.1415" and "3.141592" would clash. This raises the question of how close two numbers have to be to clash. If one player misremembered the expansion of pi and wrote "3.1419" that would clash, in the same way that spelling mistakes clash with the correct spelling. If the player got mixed up between pi and e and clued 2.71, I guess that wouldn't clash?
I once clued "1000 MB" for "gig" (we've house-ruled that numbers with units are allowed; YMMV). Someone else could have clued "1024 MB". They would clash, but they differ by more than 3.14 and 2.71 do. I don't think even a percentage difference works; I think "2cm" and "3cm" should clash for "inch".

Alex points out that cluing "2cm" and "3cm" for "measurement" probably shouldn't clash, at least not unless you're saying that units clash with each other like in the 40W/60W example below. So that would be a case of two clues clashing or not clashing depending on what they're a clue for!

To make the point about absolute or percentage differences clearer: suppose we didn't allow units. Then 1,000,000,000 and 1,073,741,824 (if you could remember or calculate it) would both be valid clues for "gig".

: I'm actually thinking that those perhaps should be considered different. It's not like the pi example, which was two different approximations to the same irrational number; those are two distinct, exact, definitions of the GB. -Nozaquirce

Also, would "40W" and "60W" for "lightbulb" clash? They're not a direct clash on the number, unlike 1000MB/1024 MB and 2cm/3cm. They're cluing via the same concept, but that's OK: it's like cluing a type of cheese for "cheese" and hoping you don't pick the same one as someone else. Or is a unit of measurement like part of a compound word, so "40W" and "60W" clash in the same way that "firewood" and "fireman" clash?

Another similar example: We recently had the word "grey" and Clive clued "#7F7F7F". What if someone had clued a different greyish hex code? I feel like it should clash, but it's hard to justify, especially if it didn't contain any 7s or Fs. Perhaps the # counts as a unit, like the W in the lightbulb example; but that doesn't solve the problem, as RGB hex codes without the # could still work as recognisable clues for some players.

A possible criterion I came up with is that two numeric clues clash if they are part of a continuum of values that would convey the same idea, or something close enough to a continuum. Going over the examples above, this criterion would remove the different-roundings-of-pi and the greys, while allowing the 10^9/2^30 and the 40W/60W; and as for the 2cm/3cm, it goes a little against the intuition expressed above, removing them when the word is "measurement", but having an unclear result when it's "inch" -- we might have to add "interpret numbers as rounded when appropriate", in which case those two would stand for the ranges [1.5, 2.5) and [2.5, 3.5) and be removed, but I'm not sure if that makes it too broad. -Nozaquirce

Another relevant case that happened was with the word being "number"; several of the clues were different real numbers, which nobody thought clashed, but would clash by the above criterion. Maybe the correct refinement is to allow it if the continuum extends to infinity? (This would also allow the "measurement" example.) This would let that type of clue have an extremely small risk of clashing, but that may be OK because it is rarely applicable. -Nozaquirce

Recently, the word was "width", and clues "1.8m" and "10m" were deemed to clash. This seems to me to be similar to the "measurement" case above; is there some relevant difference between these cases that explains the different judgement, or is this just a difference in people's views? -Nozaquirce

Prefixes and suffixes

Again, there's a sliding scale between morphemes that clearly clash and ones that don't. Clearly the "ing"s in "walking" and "running" don't clash, but presumably the "green"s in "greenery" and "greenhouse" do. Where's the line between a meaningful morpheme and a purely functional one? When the word is "dinosaur" I try to avoid cluing any dinosaurs that end in -saurus, because I think they'll clash with each other and with the target word, but not everyone seems to agree. (If you do follow that restriction, that doesn't leave many well-known dinosaurs. Presumably T-rex clashes, because the T stands for tyrannosaurus. If I avoid all the -saurus ones and clue "pteranodon", does that clash with "iguanodon"? I didn't know the answer, so I looked it up just now. The -don in "iguanodon" means tooth, but "pteranodon" comes from pter- and anodon, meaning toothless, so it's kind of an indirect clash, derived from a different Greek word that's clearly related to the -don in "iguanodon" but not actually the same word.)

Words that sound related but aren't

English is full of similar-sounding words that are actually etymologically unrelated, like "horde" and "hoard", or "adult" and "adultery". I would rule that those don't clash, but it might be hard convincing someone who thought they did.

Words that are related but distantly

If two words go back to the same proto-Indo-European root but have no nearer common ancestor, they apparently don't clash. But how far back must the common root be for two cognates to not clash?

Words that don't sound related but are

You can't clue a word that's a simple grammatical variant from the same root as the target word, so you can't clue "lovely" for "love". But English has lots of noun-adjective pairs that aren't spelled as similarly as that, like emperor/imperial and sun/solar. I have seen someone clue "imperial" for "emperor" and get quite annoyed when challenged. I think emperor/imperial is a pretty clear-cut clash, but sun/solar and moon/lunar are less so, because they're examples of that weird thing where English has ended up with one half of the pair from a Germanic root and the other from a Latin root, and it's fairly arbitrary that it's ended up that way. (See also go/went, where the present and past derive from different words, but they're now two forms of the same word and I think we'd say they clash.) But I think sun/solar still has to be a clash: "solar" is to "sun" as "lovely" is to "love", and it's not like there's an alternative adjective for "sun" that's a closer match: "solar" is the adjective for "sun" (unlike "terrestrial" for "earth", where there is the option of "earthly" and indeed "earthy", so I'd be more inclined to say that "terrestrial is just a synonym for "earthly" and therefore allowed as a clue for "earth" - but then it seems inconsistent if "terrestrial" is allowed for "earth" but "solar" isn't allowed for "sun".)

Summary

It's not a simple case either of "intent always matters" or of "intent is always overruled by the facts".
With the glass/hourglass or wind/wind examples, I think the facts have to overrule, even though different meanings of the word were intended.
But with the "bash" and probably "pteranodon" examples, I think intent has to win out: you can't penalise someone for not knowing some computing acronym that's spelled the same as their clue and whose expansion clashes, and I think you probably shouldn't penalise someone for not knowing the relations between distinct Greek words that dinosaur names are derived from.

It's a mark of how great JustOne is that I still love it and don't write it off as unplayable because of all the rules ambiguities.

Update

The rules have been [updated for 2023]; the revision explicitly allows hyphenated words. --PT

CategoryGames CategoryLanguage
OP=Rachael