Create a Magic 8 Ball

Question

As a child, my friend had a magic 8 ball that we would ask questions to and see what the fate of that question was.

Challenge

Your challenge is to write a program (or function) that when run (or called), outputs (or returns) a random answer from the possible answers below. (Random being: each output should have a nonzero chance of occurring but they do not need to meet any other criteria)

The possible answers from the Magic 8-ball are (case-insensitive):

It is certain
It is decidedly so
Without a doubt
Yes definitely
You may rely on it
As I see it, yes
Most likely
Outlook good
Yep
Signs point to yes
Reply hazy try again
Ask again later
Better not tell you now
Cannot predict now
Concentrate and ask again
Don't count on it
My reply is no
My sources say no
Outlook not so good
Very doubtful

Input

No input.

Output

A random choice from above. Case does not matter.

Rules

Standard loopholes are not allowed.

This is code-golf, so the shortest code in bytes for each language wins!

Answer 1

SOGL V0.12, 166 bytes

,▓a⁰²z○½℮ķčλ─fj[Ycψ-⁸jΔkÆΞu±⁄│(┼∞׀±q- υ~‼U/[DΓ▓νg⁸⅝╝┘¤δα~0-⁄⅝v⁄N⁷⁽╤oο[]āŗ=§№αU5$┌wΨgΘ°σΖ$d¦ƨ4Z∞▒²÷βΗ◄⁴Γ■!≤,A╬╤╬χpLΧ⁸⁽aIΘād⁵█↔‚\¶σΞlh³Ζ╤2rJ╚↓○sēχΘRψΙ±ιΗ@:┌Γ1⁷‘Ƨ! ΘlΨιw

Try it Here!

\o/ every word was in SOGLs dictionary!

Answer 2

><>, 438 bytes

x|o<"Yep"
x|^"Most likely"
x|^"Signs point to yes"
x|^"As I see it, yes"
x|^"Without a doubt"
x|^"Ask again later"
x|^"Don't count on it"
x|^"Cannot predict now"
x|^"Very doubtful"
x|^"My reply is no"
x|^"My sources say no"
x|^"Outlook not so good"
x|^"Reply hazy try again"
x|^"Better not tell you now"
x|^"Concentrate and ask again"
x|^"It's certain"
x|^"Outlook good"
x|^"Yes definitely"
x|^"You may rely on it"
x|^"It is decidedly so"

Try it online!

Not that interesting, but I think it's the first answer where the randomness is not uniform. I put all the negative messages as least likely :)

Some explanation:

The pointer starts going right at the first line. x changes the pointer to a random cardinal direction. If it goes up or down, it just encounters a different x. If it goes right, it bounces off the | and hits the same x. If it goes left, it wraps around and pushes that line's text to the stack. Most lines then hit the same track of ^ which changes the direction to upwards. This loops over the o on the first line, which outputs the stack until empty. The special case is the Yep line, which has the horizontal loop |o< instead.

Answer 3

Python 2, 369 368 bytes

print"It is certain.It is decidedly so.Without a doubt.Yes definitely.You may rely on it.As I see it, yes.Most likely.Outlook good.Yep.Signs point to yes.Reply hazy try again.Ask again later.Better not tell you now.Cannot predict now.Concentrate and ask again.Don't count on it.My reply is no.My sources say no.Outlook not so good.Very doubtful".split('.')[id(0)/7%20]

Python 3, 371 bytes

print("It is certain.It is decidedly so.Without a doubt.Yes definitely.You may rely on it.As I see it, yes.Most likely.Outlook good.Yep.Signs point to yes.Reply hazy try again.Ask again later.Better not tell you now.Cannot predict now.Concentrate and ask again.Don't count on it.My reply is no.My sources say no.Outlook not so good.Very doubtful".split('.')[hash(id)%20])

I was previously using the hash builtin to index (hash(id)%20), which returns a random value per-start of the Python interpreter ever since https://bugs.python.org/issue13703. It's not random for the empty-string though (always 0), so need to use something else, the id builtin!

On second look, I could use id directly, but it seems to always produce even numbers. IIRC, id(object) in CPython just returns the memory location of object, so this makes sense. Maybe if I used Jython or IronPython, I could skip the divide-by-7. Anyways, hash(id) vs id(0)//7 is equal in Python 3, but can use the / operator for truncating integer division in Python 2, saving a byte.

Answer 4

PowerShell, 354 bytes

"It is certain0It is decidedly so0Without a doubt0Yes definitely0You may rely on it0As I see it, yes0Most likely0Outlook good0Yep0Signs point to yes0Reply hazy try again0Ask again later0Better not tell you now0Cannot predict now0Concentrate and ask again0Don't count on it0My reply is no0My sources say no0Outlook not so good0Very doubtful"-split0|Random

Try it online!

Ho-hum. Takes all the outcomes, concatenated together with 0s, then -splits on 0 to create an array of strings. Passes that array to Get-Random which will randomly select one of them. That's left on the pipeline and output is implicit.

Answer 5

Python 2, 385 bytes

-1 byte thanks to ovs.

from random import*
print choice("It is certain.It is decidedly so.Without a doubt.Yes definitely.You may rely on it.As I see it, yes.Most likely.Outlook good.Yep.Signs point to yes.Reply hazy try again.Ask again later.Better not tell you now.Cannot predict now.Concentrate and ask again.Don't count on it.My reply is no.My sources say no.Outlook not so good.Very doubtful".split('.'))

Try it online!

Answer 6

Applescript, 391

I love how AppleScript's lists have a some item method:

{"It is certain","It is decidedly so","Without a doubt","Yes definitely","You may rely on it","As I see it,yes","Most likely","Outlook good","Yep","Signs point to yes","Reply hazy try again","Ask again later","Better not tell you now","Cannot predict now","Concentrate and ask again","Don't count on it","My reply is no","My sources say no","Outlook not so good","Very doubtful"}'s some item

Answer 7

Bash + GNU utilities, 230

• 15 bytes saved thanks to @Dennis.

sed 1d $0|zcat|shuf -n1
# zopflied 8 ball list

The binary zopfli output is not well represented here; instead you can reconstruct the script from base64 encoded data:

base64 -d << EOF > 8ball.sh
c2VkIDFkICQwfHpjYXR8c2h1ZiAtbjEKH4sIAAAAAAACAz1QSZJCMQjd5xRv1fOlMEGlzIdfgbRF
n75NOayYeYMExFF5BImWe9W4SuPWE27lKnG2GSA0m4coyWvhKCrBPUvaxEaJcStgColCDoEzQ+IH
t/WymQe6XNa+zehmF5zMWknei8tJHbuJBsKw9gfvPXGmv0SMBJ0WNfLLPUOn4FEOHMEDaoHg3rGI
qF1LJV29fXCTGveWaWWNQcEgbXi9Ks30PVBtauBOfkvc4cWhtkq3OSo7nBJqLwELxO2u45dH3u05
zv4=
EOF

Note, as allowed by the question, the compressed data decompresses to all lower case. This makes zopfli compression a bit more efficient and saves 16 bytes.

Answer 8

Charcoal, 203 184 bytes

‽⪪”}∨74Dυ3↖u➙Ｈ�↖vＩ⁻VＲ‹ψ#�Ii»ψPNξ⮌≔;≡8ν}¬Ｈ⁺ºº↖H⁴Ｋ⌕êτ｜⁼➙⟲W»″φ◨⟦(τ(jK“Ｎ\⍘“↷⊙ⅉvＴ>➙§⌊Ｆζ³⁻↔;ＴaÀ✳⁴≔67⍘i4¬⸿-A8⁻f7¡<⁰Zχ}ζ'¡¹→Oaε!ＯυP₂ïμ´MuＰ⁺Ｍ⮌1№-k¹№FvξＤü⊟ζⅉ⁰xW:Dε7ＴvM₂⊞θC⪪Ｒε⁰“Ｄ¡⸿⁰″A⊕λξ↥~Ｏ·PＥ&”¶

Try it online! Link is to verbose version of code. Edit: Saved 19 bytes by lowercasing everything. Explanation:

  ”...”     Compressed string of newline-delimited responses
 ⪪     ¶    Split on newlines
‽           Random element
            Implicitly print

Answer 9

R, 360 bytes

sample(readLines(),1)
It is certain
It is decidedly so
Without a doubt
Yes definitely
You may rely on it
As I see it, yes
Most likely
Outlook good
Yep
Signs point to yes
Reply hazy try again
Ask again later
Better not tell you now
Cannot predict now
Concentrate and ask again
Don't count on it
My reply is no
My sources say no
Outlook not so good
Very doubtful

Try it online!

Not exactly the most elegant solution. R has a neat feature where stdin will redirect to the source file, so you can put (small) datasets into source code, saving bytes for string splitting or worse, constructing the vector itself (all those quotes add up in a hurry). Along with builtins for random sampling, this makes a short-ish answer.

Answer 10

Retina, 333 331 321 bytes

0cert10decided2so¶with34a d3bt¶yes definitely¶y3 ma5re26as i see it, yes¶mos4likely7good¶yep¶signs poin4to yes¶rep2haz5tr5ag18ain later¶better 94tell y3 9w¶can94predic49w¶concentrate and 81don'4c3n46m5rep2is 9¶m5s3rces sa59794so good¶ver5d3btful
9
no
8
ask ag
7
¶3tlook 
6
on it¶
5
y 
4
t 
3
ou
2
ly 
1
ain¶
0
it is 
G?`

Try it online! Edit: Saved 1 byte by compressing doubt and 1 byte by lowercasing everything so I could compress reply. Then saved 10 bytes by using @Leo's Retina Kolmogorov golfer on the lowercased text (which coincidentally is the number of bytes it saved on my 333-byte answer).

Answer 11

Coconut, 380 bytes

Coconut port of totallyhuman's answer

from random import*
choice$("It is certain.It is decidedly so.Without a doubt.Yes definitely.You may rely on it.As I see it, yes.Most likely.Outlook good.Yep.Signs point to yes.Reply hazy try again.Ask again later.Better not tell you now.Cannot predict now.Concentrate and ask again.Don't count on it.My reply is no.My sources say no.Outlook not so good.Very doubtful".split('.'))

Try it online!

Answer 12

Jelly, 201 bytes

-2 bytes thanks to Mr. Xcoder. -1 byte thanks to user202729.

“æ⁽IẊ?⁽ʋṠ¶ÐƝKW¬ḃỴɓ⁾:Eṇ⁵ṾɱD×⁴2ṇỤðċỊ¥ḷƬị÷ṣÐṆⱮ$u²OŀṚƁȮ1⁼ṁ$bp⁾v]Ɠ-/NẓḲnỵdḳḋ½ȥṿ=kv¥ɓl[kR AḞ¶gḣḞiẊŒẊḳçȤ⁻Ɱʋx:ØṖ|zY=ṾḌẓY1Ḃ$50d⁹⁸ŀhʂƤṢM;ḢoƁṾ⁷-uṙu¡Ọ3ṣȮ@⁹ðẹȥXƭ⁸|ƬẋẆḢɠœxḳsĿƘ(0çỌ~A½YIEFU3Ọ=⁷ɗḷBḷİṄhṗgṡƊẏẏḄ#Ṙʋ$ʂȷĠ»ỴX

Try it online!

Damn, SOGL's compression is good.

Answer 13

Ruby, 362 361 bytes

puts"It is certain.It is decidedly so.Without a doubt.Yes definitely.You may rely on it.As I see it, yes.Most likely.Outlook good.Yep.Signs point to yes.Reply hazy try again.Ask again later.Better not tell you now.Cannot predict now.Concentrate and ask again.Don't count on it.My reply is no.My sources say no.Outlook not so good.Very doubtful".split(?.).sample

Try it online!

• 1 byte thanks to @benj2240

Answer 14

05AB1E, 171 bytes

“€•€ˆ‹ì€•€ˆŸíly€Ê„›€…¬³…ܴ΀˜€‰€•€œ I€È€•,…Ü‚¢îÙ®½‚¿ yepŸé…®€„…Ü…ƒ hazy‡Ü†îˆ¹†îŠ´…瀖ˆœ€î€Ó€©notßä€Óä考ˆ¹†î€·n'tš‹€‰€•€¯…ƒ€ˆ€¸€¯Žç…耸®½€–€Ê‚¿‚Ò¬³ful“#•8∞f{ʒβ®•6в£ðýΩ

Try it online!

Explanation

“ ... “ pushes a string of all the required words.
Some words are taken directly from the 05ab1e dictionary.
Some are written out in plain ascii (like haze).
Some are combined dictionary and ascii (like do+n't).

Then the processing code is:

#                 # split string on spaces to a list of words
 •8∞f{ʒβ®•        # push the number 2293515117138698
          6в      # convert to a list of base-6 numbers 
                  # ([3,4,3,2,5,5,2,2,1,4,4,3,5,3,4,4,4,4,4,2])
            £     # group the list into sublists of these sizes
             ðý   # join on spaces
               Ω  # pick one at random

Answer 15

T-SQL, 393 bytes

SELECT TOP 1*FROM STRING_SPLIT('It is certain-It is decidedly so-Without a doubt-Yes definitely-You may rely on it-As I see it, yes-Most likely-Outlook good-Yep-Signs point to yes-Reply hazy try again-Ask again later-Better not tell you now-Cannot predict now-Concentrate and ask again-Don''t count on it-My reply is no-My sources say no-Outlook not so good-Very doubtful','-')ORDER BY NEWID()

The function STRING_SPLIT is only available in SQL 2016 and later.

Best I could get for prior versions using VALUES('It is certain'),('It is decidedly so'),... was 464 characters.

Formatted, just so you can see the working part:

SELECT TOP 1 *
FROM STRING_SPLIT('It is certain-It is decidedly so-...', '-')
ORDER BY NEWID()

NEWID() generates a new, pseudo-random GUID, so is a way to do a pseudo-random sort.

Answer 16

Python 3, 386 bytes

from random import*
lambda:choice("It is certain;It is decidedly so;Without a doubt;Yes definitely;You may rely on it;As I see it, yes;Most likely;Outlook good;Yep;Signs point to yes;Reply hazy try again;Ask again later;Better not tell you now;Cannot predict now;Concentrate and ask again;Don't count on it;My reply is no;My sources say no;Outlook not so good;Very doubtful".split(';'))

Answer 17

Java 8 , 433, 392, 380, 379 bytes

 a->"It is certain~It is decidedly so~Without a doubt~Yes definitely~You may rely on it~As I see it, yes~Most likely~Outlook good~Yep~Signs point to yes~Reply hazy try again~Ask again later~Better not tell you now~Cannot predict now~Concentrate and ask again~Don't count on it~My reply is no~My sources say no~Outlook not so good~Very doubtful".split("~")[(int)(Math.random()*20)]

Try it online!

• 41 bytes thanks to AdmBorkBork!
• 10 bytes thanks to Kevin!
• 1 byte thanks to Oliver!

Answer 18

Javascript, 372 bytes

-10 bytes thanks to Shaggy

_=>"It is certain.It is decidedly so.Without a doubt.Yes definitely.You may rely on it.As I see it, yes.Most likely.Outlook good.Yep.Signs point to yes.Reply hazy try again.Ask again later.Better not tell you now.Cannot predict now.Concentrate and ask again.Don't count on it.My reply is no.My sources say no.Outlook not so good.Very doubtful".split`.`[Math.random()*20|0]

Try it online!

Answer 19

05AB1E, 208 217 bytes

"don'".•W˜FζÃT¥„ò.1₁∍Y<`Ì°5jýúž+ìmHSéÁ¬–xȃø‚ž}_Øviòª§l["]0â^)„2æδ∍G1∊EÌLÝ'îôΛβ;ƒĀαÏw L°gðÈγ³€wE‘f饤šαrˆQŠë¢-º8Æ~ÁŠ∍δBx®(β™Žü6»ƶÙÐ~†«\%ÍŒΘ-´sÈƵJŸ₃H7Ó˜:Å∍₂èÑï∞—Râú'óвb…ÓUXʒǝ₄ÝrÒ₄¨÷ä褓oθWÎλî~bj(Ri
Þиe‘ãj]•", yes"J'x¡Ω

Try it online!

Pretty basic solution. The possible answers are concatenated with the character x (since it's not present in the answers) and then compressed (inside the •), 'x¡Ω split on x and pop a random choice.

Thanks to @Emigna for pointing out that the alphabet compression doesn't like ' or , much. Fixed by surrouding the compressed string with don' and , yes.

Answer 20

Perl, 366

print((split",","It is certain,It is decidedly so,Without a doubt,Yes definitely,You may rely on it,As I see it,yes,Most likely,Outlook good,Yep,Signs point to yes,Reply hazy try again,Ask again later,Better not tell you now,Cannot predict now,Concentrate and ask again,Don't count on it,My reply is no,My sources say no,Outlook not so good,Very doubtful")[rand 19])

Answer 21

PHP, 412 385 337 384 bytes

<?php $a=explode(1,"It is certain1It is decidedly so1Without a doubt1Yes definitely1You may rely on it1As I see it, yes1Most likely1Outlook good1Yep1Signs point to yes1Reply hazy try again1Ask again later1Better not tell you now1Cannot predict now1Concentrate and ask again1Don't count on it1My reply is no1My sources say no1Outlook not so good1Very doubtful");echo$a[array_rand($a)];

Try it online!

Fairly straight forward solution. Split the string by a delimiter(in this case 1) and choose a random element from the array.

Answer 22

Befunge

1221 870 bytes (perimeter of the entire field is 33x36 30*29 charachters) Thanks to Jo King for helping me to remove the trailing nulls and urging me to change the randomizer.

"<"99+9+1+v
v         <
 >>>>>>>>>>55++v
 0123456789
>??????????<
 0123456789
 >>>>>>>>>>    v
               >88++p       v
v"It is certain"           
v"It is decidedly so"
v"Without a doubt"
v"Yes definitely"
v"You may rely on it"
v"As I see it, yes"
v"Most likely"
v"Outlook good"
v"Yep"
v"Signs point to yes"
v"Reply hazy try again"
v"Ask again later"
v"Better not tell you now"
v"Cannot predict now"
v"Concentrate and ask again"
v"Don't count on it"
v"My reply is no"
v"My sources say no"
v"Outlook not so good"
v"Very doubtful"
>:#,_@

The top line puts the '<' character and the x-position (28) where it should go on the stack. Then we enter the sort of random number generator. This could be improved, but this is what I could deliver on short notice... The "random" number is offset to get the actual "random" line to read.

After the random number is generated, we put the '<' character at that line and push the letters on the stack and on the bottom line output them again.

Note; if you use the interpreter I linked to in this posts title you have to reclick the "Show" after each run, because the addition of the '<' character remains after execution.

Answer 23

Red, 367 bytes

prin pick split{It is certain.It is decidedly so.Without a doubt.Yes definitely.You may rely on it.As I see it, yes.Most likely.Outlook good.Yep.Signs point to yes.Reply hazy try again.Ask again later.Better not tell you now.Cannot predict now.Concentrate and ask again.Don't count on it.My reply is no.My sources say no.Outlook not so good.Very doubtful}"."random 20

Try it online!

It doesn't seem really random in TIO (although it works just fine in the Red Console), that's why I added a random/seed to the header.

Answer 24

Excel, 399 Bytes

=CHOOSE(1+20*RAND(),"It is certain","It is decidedly so","Without a doubt","Yes definitely","You may rely on it","As I see it, yes","Most likely","Outlook good","Yep","Signs point to yes","Reply hazy try again","Ask again later","Better not tell you now","Cannot predict now","Concentrate and ask again","Don't count on it","My reply is no","My sources say no","Outlook not so good","Very doubtful")

Since CHOOSE(X.Y,<>) is the same as CHOOSE(X,<>), no need for an INT

Not much golfing you can do here though...

Answer 25

Aceto, 345 + 1 = 346 bytes (+1 for -l flag)

"It is certain.It is decidedly so.Without a doubt.Yes definitely.You may rely on it.As I see it, yes.Most likely.Outlook good.Yep.Signs point to yes.Reply hazy try again.Ask again later.Better not tell you now.Cannot predict now.Concentrate and ask again.Don't count on it.My reply is no.My sources say no.Outlook not so good.Very doubtful"'.:Yp

Try it online!

Not overly interesting, but I can't think of anything shorter in this language, no compressed strings or anything.

"...."   push strings separated by periods
      '.  literal period
        :  split on period
         Y  shuffle stack
          p  print top

Answer 26

Excel-VBA, 362 341 339 Bytes

v=[1:1]:?v(1,Rnd*19)

Where A1:T1 contain the different options. Reads entire first row of sheet into array v and indexes a random point along the first 19 values.

Surprised to find that indexing an array doesn't require integer values

Answer 27

C - 426 bytes

char a[][99]={"It is certain","It is decidedly so","Without a doubt","Yes definitely","You may rely on it","As I see it, yes","Most likely","Outlook good","Yep","Signs point to yes","Reply hazy try again","Ask again later","Better not tell you now","Cannot predict now","Concentrate and ask again","Don't count on it","My reply is no","My sources say no","Outlook not so good","Very doubtful"};int main(){int n;puts(a[n%20]);}

Uses an uninitialized variable mod 20 to index into an array of strings containing all possible outputs. Compilers complain that stdio.h isn't included, but it works OK. Probably because it just so happens to have the standard library linked in anyways. Lucky me.

Answer 28

Go, 530 Bytes

package main;import"fmt";func main(){for k:=range map[string]struct{}{"It is certain":{},"It is decidedly so":{},"Without a doubt":{},"Yes definitely":{},"You may rely on it":{},"As I see it, yes":{},"Most likely":{},"Outlook good":{},"Yep":{},"Signs point to yes":{},"Reply hazy try again":{},"Ask again later":{},"Better not tell you now":{},"Cannot predict now":{},"Concentrate and ask again":{},"Don't count on it":{},"My reply is no":{},"My sources say no":{},"Outlook not so good":{},"Very doubtful":{}}{fmt.Print(k);break}}

Please note that, on the Go Playground, because of how seeding works, it always gives the same result. When running on a regular computer, everything works as it should.
I think it is possible to save a bit more but my knowledge in Go stops there :)

Formatted and testable version

Answer 29

VBA, 358 bytes

An anonymous VBE immediate window function that takes no input and outputs to STDOUT.

?Split("It is certain1It is decidedly so1Without a doubt1Yes definitely1You may rely on it1As I see it, yes1Most likely1Outlook good1Yep1Signs point to yes1Reply hazy try again1Ask again later1Better not tell you now1Cannot predict now1Concentrate and ask again1Don't count on it1My reply is no1My sources say no1Outlook not so good1Very doubtful",1)(19*Rnd)

Answer 30

Excel Google Sheets, 319+21 = 342 340 Bytes

=INDEX(A:A,1+20*RAND(

Where cells A1:A19 contain the different output options. Reads from the array using a random index. Saves a whole load of bytes by avoiding a delimiter in the list

Saved 2 bytes thanks to Taylor Scott

Answer 31

Java 8, 379 Bytes

b->"It is certain-It is decidedly so-Without a doubt-Yes definitely-You may rely on it-As I see it, yes-Most likely-Outlook good-Yep-Signs point to yes-Reply hazy try again-Ask again later-Better not tell you now-Cannot predict now-Concentrate and ask again-Don't count on it-My reply is no-My sources say no-Outlook not so good-Very doubtful".split("-")[(int)(Math.random()*20)]

Try it online

Answer 32

C (gcc), 408 bytes

f(s){char*p="It is certain\0It is decidedly so\0Without a doubt\0Yes definitely\0You may rely on it\0As I see it, yes\0Most likely\0Outlook good\0Yep\0Signs point to yes\0Reply hazy try again\0Ask again later\0Better not tell you now\0Cannot predict now\0Concentrate and ask again\0Don't count on it\0My reply is no\0My sources say no\0Outlook not so good\0Very doubtful";for(s=time(0)%20;s;)*p++?0:s--;s=p;}

Try it online!

Users are hard to predict, making the timing of their decision to call the function pseudo-random enough for our purposes.