« first day (1410 days earlier)      last day (3435 days later) » 

6:00 PM
IDs are attached to a player, and every player has a public name
 
Nevermind I'm totally lost
How does observing where they sit help us determine the IDs of the remaining players?
Are IDs tied to a specific chair?
 
no, they are attached to a player
but if I call out an ID, then I can be fairly positive the person next to me is the one I called the ID on
 
@Rainbolt I think the point is that the first guess is random, but each turn you can figure out another ID
 
The first guess is certainly random
 
You win by seating your entire team?
 
6:02 PM
so then the objective is to figure out all the ids first?
 
there are 4 chairs
you win by having your team occupy those 4 chairs
 
ah
 
the 4 chairs are next to each other
 
I thought there were 11 chairs :-/
 
who is on those chairs initially? two from each team?
 
6:03 PM
@Ra
 
there are 11 chairs, but 4 special chairs, and yes 2 from each team
 
yeah that xD
 
interesting
 
lol for enter not filling out the name
What if the random starting position has a team entirely on the 4 seats?
 
You need at least three letters @cjf
 
6:04 PM
Its not random
its every other
 
I like the conflict of wanting to move a particular person from one chair to another, with the desire to get another turn after this one by emptying a chair next to a team mate
 
so then, you already know where they all are when it starts :P
 
you know where your team is, yes
but you don't know their IDs
 
uh, yes you do
 
Not until you call them for the first time.
 
6:05 PM
you need to know your own id, and you control all your team members
 
OH haha. Cj has a point.
 
so if the teams are 1 and 2, is the initial layout like this?
team:    1 2 1 2 1 _ 2 1 2 1 2
special: S S _ _ _ _ _ _ _ S S
 
each instance of your team cannot communicate
 
I assume you can't call your own number?
 
@NathanMerrill yes, but your own turn is perpetual
 
6:06 PM
you can call your own number, but its not as useful as you think
 
by moving the other team's members to elongate your turn and figure out the other ids,
you figure out all the ids in 4 turns
without giving up your control
 
That is really tough
because you have to have 4 people on your team sitting next to each other
AND they all have to know the ID of the person to move
the layout looks like this:
1 2 1 2 1 2 1 2 1 2 _
S S S S _ _ _ _ _ _ _
 
@NathanMerrill How will you prevent communication while also allowing save state?
 
the chair next to the special chairs is the most powerful
 
oh so the special chairs are next to the initially empty one
 
6:08 PM
I don't allow save state
 
If I cannot save state, how will I remember the IDs that were called?
 
5 programs that continuously run from each team
 
I won't even be able to remember the IDs that I called.
 
so the ids are random?
 
randomly assigned. They'll probably be 0-9
 
6:09 PM
@Rainbolt you're told your id on the first turn you take, you can save state whatever you want
 
Well he said you can't save state, but I get what he means now.
You can't save state outside of the state that is saved while running.
 
the problem with this is that the team who takes the first turn can win without even giving the other team an opportunity
 
Can I write five different programs? Maybe I want to assign a leader and four followers.
 
@cjfaure how?
 
CJ impossible
Rainbolt, yes
 
6:11 PM
if you know all your team's ids, you can get all your teammates on the special chairs without letting the other team have a turn
 
@NathanMerrill nice
 
right?
 
@cjfaure but you don't
 
If I move my own teammate, then the other team gets a turn
 
so the team to take the first turn figures out all the team ids in 5 turns
on the 5th turn, they can know all the ids
 
6:11 PM
because the person who calls the ID is the person to the right of the empty chair
 
the other team would have figured it out the turn after
 
you have to have all of your teammates sitting next to each other
 
@NathanMerrill I think you have reached the point where you can write up a solid spec. All of this discussion might get lost if you don't start writing, which will lead to people asking you the same question a second time.
 
ok
well, I'm about to take a final, but I'll start it up when I get home
bye!
 
Good luck!
Unless it's a theater final, in which case, break a leg.
 
6:12 PM
rubber duck debugging works well on koths anyway :P
 
6:23 PM
I almost have a 25/26 for Pyth but I don't know how to read in newlines :/
 
yeah
Pyth can't
( at least the online version.) and passing "\n" is treated as 2 characters
 
can't you define a function?
 
Can't figure out how to test it to make sure it works :/
I'll try CJam
 
@Sp3000 are you using transliterate in Pyth ?
(in that 25/26 version)
 
Er... nope. I did essentially what's in my Python one, just with something else to index
Or at least that's what I tried to do
(just read your CJam solution, not having inbuilts makes it hard...)
 
6:34 PM
well, if transliterate worked, I would have had 22/23 byte solution
not having inbuilts ?
@FryAmTheEggman played around with the new transliterate X in Pyth ?
 
@Sp3000 There are some bugs in Pyth right now
Yeah its broken
 
:(
 
Inbuilt split
Er... is there any way to assign a multiline string to a variable?
 
@Sp3000 / is the split
 
I've been meaning to tell isaac he just wrote lambda: lambda: basically :P
 
6:36 PM
I mean Python splits by lengths of whitespace, but you had to built up the string :/
 
For the multiline thing, you had to use \n in the input before, but right now the online eval is busted
 
:D
well done isaacg :)
 
I think Isaac was trying to prevent evil people from breaking his comp :P
 
its not his comp
 
If you download it using \n should work
well the MIT one that he is hosting it on :P
The online compiler is different (i.e. doesn't let you use $)
 
6:38 PM
@Sp3000 So python can directly split on multiple continuous occurrences of a group of characters ?
"abd \n 123".split(" \n") == ["abc", "123"] ?
 
Yeah
Gets rid of leading/trailing whitespace too
 
@FryAmTheEggman it does. (at least it did)
 
Er, you don't need an argument in the split
It's just .split()
 
split on what ?
 
Oh, we are allowed to use split?
 
6:39 PM
If split is called with no arguments it splits on runs of whitespace
 
@FryAmTheEggman yes
 
nice!
 
@FryAmTheEggman Yeah, hence me trying to use Pyth
 
Also it doesn't in the newest version, he added the "safety measures" recently :P
 
@FryAmTheEggman he changed that last minute
(martin)
 
6:40 PM
@Sp3000 Yeah just write it appropriately and I'll test it with the downloaded one
Or something
 
Well for reference the idea was @[lf}T(bdC9)z cz))-l(0 0)1
Basically (0 0) length 2 array -> (00) length 1 array
Or something
 
You can use a space to prevent automatic printing
 
@MartinBüttner is the input string guaranteed to be consisted of printable ascii ? (except for tab)
 
@Optimizer yes printable ascii + newlines + tabs
 
newlines are printable though :P
 
6:43 PM
I don't think so
I think printable ascii is strictly 0x20 to 0x7E
newlines are control characters
ASCII (/ˈæski/ ASS-kee), abbreviated from American Standard Code for Information Interchange, is a character-encoding scheme. Originally based on the English alphabet, it encodes 128 specified characters into 7-bit binary integers as shown by the ASCII chart on the right. The characters encoded are numbers 0 to 9, lowercase letters a to z, uppercase letters A to Z, basic punctuation symbols, control codes that originated with Teletype machines, and a space. For example, lowercase j would become binary 1101010 and decimal 106. ASCII codes represent text in computers, communications equipment, and...
 
I meant just "printable"
 
@Sp3000 ` lf}T(bdC9)z cz`
 
lol, love the pronunciation guide...
 
remove the first space for count, second space for split...
 
wow that's short.
You should post it yourself.
 
6:45 PM
Oh, you can do that? (I really still don't get how Pyth works)
 
Yeah 90% of what I had yesterday was just writing split
Pyth will automatically print any evaluated statement, but a leading space suppresses that behaviour
 
i always said ascii as "ahsee"
 
@FryAmTheEggman writing split on my own was just 10 bytes extra
 
"asky" is so weird D:
 
Yeah its way easier is stack based languages :P
 
6:47 PM
Well I meant I can't tell whether a function does the right thing because I'm never sure how to test it :P
 
that last minute built in allowing ruined everything XD
Curse you @MartinBüttner . That was never in the sandbox.
:P
 
yeah I forgot to edit that out
 
To be fair, it was in the Sandbox... for 5 minutes :P
 
I already decided that two days ago
 
as far as I saw, sandbox always had no built ins to split
 
6:48 PM
2 things are abundant in sandboxes
rough questions, and your neighbor's cat litter
 
and as far as I can see, it was to prevent Python only.
 
@Optimizer no it wasn't to prevent anything in particular
in fact, I would have thought that most built-in splits would actually return empty strings at least at the beginning
 
I've never seen a built-in split return any empty strings :P
 
@FryAmTheEggman have you used other languages than python? :P
 
Well... Maybeee..... (I actually have used C# and Java)
 
6:53 PM
in Java, split returns empty strings at the beginning
(but omits them at the end by default)
 
Oh wow. Didn't know that :P
 
.NET returns all empty strings
 
WHAAAAAAAAAAAAT
 
ruby doesn't
 
Oh shoot you're right, i remember having to use linq to get rid of them now...
I guess I was lying
:X
 
6:56 PM
I think PHP returns them too
I mean, it usually makes more sense
e.g. if you have CSV data with empty fields
 
Speaking of builtins, I just found another one in Python :D
 
isspace returns true for new line and tabs too ?
 
Yup
(just tried)
 
....
 
Python's also got isdigit(), isalpha(), isupper(), islower() and isalnum()
 
7:00 PM
was that renamed in 3 to iswhitespace?
 
hm, it probably also returns true for \r and \v, but you're lucky with my assumption about input characters :D
 
Er... no I don't think so
Heh, hence why I was happy when you said the printable ASCII thing :P
"Whitespace characters are those characters defined in the Unicode character database as “Other” or “Separator” and those with bidirectional property being one of “WS”, “B”, or “S”."
 
I vote for disabling builtins again
 
Oops, I was looking in builtins XD
 
7:04 PM
[This post](http://www.giantbomb.com/forums/general-discussion-30/winter-steam-sale-when-does-it-start-564697/) is hilarious.
Q: Winter Steam Sale, When does it start?
A: The moment you let your guard down.
3
 
Ugh why no worky linky
 
Brackets backwards
[]()
 
That's how I had it...
Oh wow.
Look at the transcript
 
... ._.'
 
7:06 PM
Whatev. If it works in the transcript then I won't bother fixing it.
 
@Rainbolt multiline -> no markdown
@PeterTaylor can you close things as duplicates of duplicates? o.O
 
I think he means how he can instantly close code golf dupes
:P
 
yeah I know he does
 
Oh, I get it now
 
but I didn't think that was a valid dupe target in the first place
 
7:13 PM
If a dupe was a valid target, then you could close two questions as dupes of each other.
 
well maybe you can as long as you don't create cycles
(although I doubt that the system is that flexible)
 
We could create a graph so large that the system crumbles.
 
@MartinBüttner are we allowed to require input formatting (for whitespace challenge)?
 
wow, someone just downvoted the CJam entry, like golfing language has any benefit in that challenge anymore
 
@FryAmTheEggman what do you mean by that?
 
7:18 PM
people are amazing, upvoting invalid C answer -_-
 
Because of how python takes input, the " and ' characters are unfriendly, can I require """ before and after input?
So like """hello "
world" can't"""
 
does that matter? isn't that just syntactic sugar for "hello \"\nworld\" can't"?
 
Oh yeah, is that ok too then? (i.e. require a python formatted version of your string?)
 
that is a normal string, not python formatted string
 
I mean sure why not? your input a sequence of bytes... it doesn't matter how that string was originally built/entered
ha, finally caught up with Calvin's Hobbies badges
 
7:22 PM
ok, just didn't exactly match your one example :P
also, congrats :D
 
@FryAmTheEggman which one, why not?
 
Well, you couldn't input `abc def
gh ij k` as-is. I dunno, some challenges have prevented me from doing that before :P
 
I think it's completely irrelevant how you build that string... for all I care you could do ["abc def","gh ij k"].join("\n") and pass that to the function. the representation of the string in the code that calls your answer has nothing to do with your submission.
 
Woops I had a string literal
 
not in the 117 one though?
 
7:31 PM
Oh true
... I'll put that one back on
 
7:44 PM
@FryAmTheEggman oh wait, you want to expect your STDIN to be evalable?
that's a whole different story :P
 
Yeah that's what I meant :P
I'm good english
 
if you take input from STDIN it should be purely the string itself.
I thought you were talking about function arguments
 
That's why I said input :P
 
sorry for the misunderstanding
 
np, I have a function one written, but it's much longer
 
7:49 PM
What's wrong in evaling input ? So in CJam, I can't do q~ ?
 
No, what I meant was you input a string of the characters "h" "\" "n" "o" to mean:
h
o
 
@Optimizer not if the input format doesn't allow it. you wouldn't put "abc\ndef" on STDIN, read and eval it. you'd just put the string you want on STDIN and read it.
i.e. if you read from STDIN your input should be a string and not the string representation of a string
@Sp3000 hexagon on hyper hexagon
 
:/ somehow I wish Python had more unary operators than + - ~
 
it had
 
:o wow you're blitzing through this game fast
 
8:01 PM
err Python
 
(and at 68 seconds in fact my longest time yet)
 
I'm sure if you redid Hexagon you could hit 120 :P
 
you were right though Hyper Hexagon was much easier than Hexagonest, and I was lucky not to get that one messy pattern :D
okay, this is a bit ridiculous now
I got hexagon on hyper hexagoner on the first try o.O
 
What.
Too fast!
 
died straight at 60:49 though :D
 
8:04 PM
Ahaha :P
GL with Hyper Hexagonest
 
lol
Difficulty: "Hardestestest"
oh shit
Stair 2...
 
GL, I say, GL
 
I couldn't even do Stair 1 (that's the pattern I luckily didn't get in Hyper Hexagon)
23 seconds... that's enough for now :D
 
I keep thinking "surely I'm done golfing now" but then my brain goes "nope".
 
8:13 PM
@FryAmTheEggman Can we have newlines and tab in that string now ?
oh wait, that was issacg
 
@MartinBüttner ++"abc def"b"gh ij k" a valid input string ?
 
I should get a GitHub account just to work on pyth :P
 
@Optimizer I'm pretty sure that's just building the string and then calling the function with a single string object. so yes.
@Sp3000 btw no one said your functions must be named
 
Oh... anonymous functions are fine?
 
8:16 PM
yes, they always are, unless the question requires a named function
 
Well seeing how it's not going to win anyway I just put a note in :P
It seems awkward seeing an anonymous function followed by 2 lines of junk
 
oh right
that makes me wonder what our policies are on submissions that define a function and also include code outside the function
 
Eh, actually I'll leave the post as is
 
Uh, actually, could you not have the second newline? Even though it will throw, you still defined the function :P
 
Hmm well I thought the idea for an anonymous function is you can insert it into code
 
8:23 PM
hm yes, that would make sense
 
If I didn't mind throwing exceptions, I could just have done <-1or x.split()
I don't like the idea of doing something then throwing an exception, then saying technically the job was done. Just a personal thing :P
 
yeah I don't much like it, but it would be sorta neat if it were technically allowed
@isaacg Hey, there are a few bugs in Pyth that I've noticed with X and the 'safe' eval
 
8:36 PM
I think either (1) the entire code should be an expression that evaluates to a function or (2) the code executes successfully, and after running it a named function is defined
 
8:52 PM
Gah. I've had to change my password like five times in the last few months because I call someone over for code review after I go to the bathroom or something and when I unlock my machine I type my password into notepad or something (because it wasn't actually locked to begin with)
 
LOL
 
If my screen didn't take five seconds to go from idle to active, I would see it sooner.
 
That'll teach you to lock it more reliably :P
So will coworkers that like to mess with stuff, incidentally.
 
I mean I could see it before I type enough of it to make it obvious
My passwords are ridiculously long, so a few characters isn't so bad
 
Bah. If a few characters isn't so bad, why have long passwords to begin with? ;)
 
8:55 PM
@Geobits common practice at the company I used to work for was to send an email to everyone that "you" are gonna bring cake the next day
 
@Geobits That question makes no sense. If my password were a few characters long, then a few characters would be bad...
 
@Rainbolt I meant, if you don't mind the unknown part of your password going from X chars to X-F chars long, you might as well just use an X-F char password to begin with.
That probably still could have been said more clearly...
 
But then you could apply the same logic to X-F.
 
No, because I do care if people see part of my password.
 
Eventually you would reach the case where your password is blank, and you don't mind people knowing that it is blank, which is obviously false, so your original statement was false.
Well I don't care if people see part of mine.
 
8:59 PM
Right. I'm saying that means your password is too long :P
 
My password is too long because other people seeing part of it is ok?
That's a non sequiter.
 
Maybe not "too" long, but unnecessarily long.
 

« first day (1410 days earlier)      last day (3435 days later) »