« first day (1311 days earlier)      last day (3525 days later) » 
00:00 - 18:0018:00 - 00:00

12:30 AM
Hello @Sparr
 
@PhiNotPi in your layout, bots can still attack on turn two depending on how their neighbors move
 
Well, it's better than before.
It used to be a 1-move thing.
 
1:31 AM
err
move then attack
if your opponent sits still
in your layout, it's move then attack, if your opponent moves
 
1:43 AM
@Sparr I don't see how it's a one-move thing when the opponent moves.
The opponent would have to rotate and move to get within one-move distance.
But if you still think it's too dense, I can make it sparser.
@.........@........
.......@.........@.
....@.........@....
.@.........@.......
........@.........@
.....@.........@...
..@.........@......
.........@.........
......@.........@..
In my original layout, there's only a small chance that you could attack on turn 2.
 
2:17 AM
The bounty on Traders to the Death will expire in 2 hours and (I think) won't be awarded to anyone.
 
2:30 AM
Unless, of course, someone were to upvote my submission... :)
That won't happen though.
Okay, I know somebody's watching this, because it just got upvoted.
Bye for tonight.
@Sparr Consider a sparser map a definite change.
 
3:17 AM
@PhiNotPi not all bots start out facing the same direction in Code Bots
also, I ran a full pass of the code bots controller with a 1/4-as-dense map. the outcome didn't change much.
less density gives bots one or two extra turns to set up blocking/defense (good), and makes short code for moving around in square paths much less effective (bad?)
 
 
3 hours later…
6:06 AM
0
A: Sandbox for Proposed Challenges

hobbsUnicode Case Mapping code-golf Challenge: Given a string as input, print it in both lowercase and uppercase. The string will be valid UTF-8, and your output should be as well. Characters without lowercase or uppercase mappings should be printed unmodified. Invalid, un-assigned, and private-use...

 
 
1 hour later…
7:34 AM
@Sparr you really could have finished and posted your work-in-progress for the Horror Movie Search Party chalenge :P ... now my bounty rep goes down the drain
(well you've still got 14 hours to do so)
 
 
2 hours later…
9:13 AM
@MartinBüttner it seems like marbelous is slowly starting to spread a bit, have you noticed?
 
@overactor I saw a new user using it, yes.
did you guys ever consolidate the design goal document?
@Rainbolt wat?
 
@Sparr is of the opinion that only the handling of STDOUT needs some work before we can give it some more exposure
I think we ironed most things out
We didn'te explicitely discuss the design goal doc in much detail though
 
I'd still like to write an interpreter in JS, based on which we could then build browser-based tools or even an IDE
but I'm not sure when/if I'll get around to do that
 
Yeah, that would be a logical next step
 
next on my list is writing the controller for the Tetris Tournament
and I'm still putting off two KotHs
 
9:27 AM
seems liek everyone is quite busy
@MartinBüttner Wow, this is a brilliant idea! The Marbelous language is so fun!– rubik 49 mins ago
 
9:42 AM
where was that?
 
28
A: If a program terminates and there is no one to see it, does it halt?

overactorMarbelous 68 66 bytes }0 --@2 @2/\=0MB }0@1\/ &0/\>0!! -- @1 00@0 --/\=0 \\@0&0 Marbelous is an 8 bit language with values only represented by marbles in a Rube Goldberg-like machine, so this wasn't very easy. This approach is roughly equivalent to the following pseudo-code: function recursiv...

Oh, I just reached 1k rep and got my first gold badge today, cool
 
@overactor Congrats!
 
I wanted to say: "I'm playing in the big league now!"
But that wouldn't be entirely accurate.
I'm out of the pee wee leaugue now!
 
10:08 AM
can someone explain to me the pun in "Two-Many Outputs"?
 
10:33 AM
I don't get it either
 
The only thing I can see in it is "Too Many Outputs", but that's probably not the pun.
 
well I thought it would be a pun on that, but then I don't get "Two" or the hyphen.
 
I think it is a pun on Too many
I don't get what's puny about that though
 
0
A: Sandbox for Proposed Challenges

Beta DecaySaved work error code-golf While looking around the internet, I came across the following error message displayed by an Amiga TOS: Your challenge is write a script which throws an error after more than two hours of keyboard entry with no disk entry. As this is code golf, the shortest program ...

 
Okay, if I understand Geobits's code correctly, his helper function either sums all the prime factors of n (for the first and third input) or it sums all but the largest prime factor and adds 1 (for the second input). then he does: integer square root twice for first input, times 11 for second input, two times integer square root twice for third input. And then he returns the sum of these three numbers.
ah no wait
for the second input it determines the number of prime factors instead of summing all of them
 
11:16 AM
I'm working on Code Bots 3 and I can't decide on how to place the bots on the map at the start of the game.
Some options:
@...@...@...@...@...
....................
..@...@...@...@...@.
....................
@...@...@...@...@...
....................
..@...@...@...@...@.
....................

....@.......@.......
.......@.......@....
..@.......@.......@.
.....@.......@......
@.......@.......@...
...@.......@.......@
......@.......@.....
.@.......@.......@..

@...@...@...@...@...
..@...@...@...@...@.
@...@...@...@...@...
..@...@...@...@...@.
@...@...@...@...@...
..@...@...@...@...@.
@...@...@...@...@...
..@...@...@...@...@.
 
11:33 AM
I like 2 best, followed by 1, then 4 and then 3
 
@MartinBüttner I saw your comment on the Two Many Outputs (because you told me to tell you when I saw it).
3 was how it was in the original
 
@PhiNotPi Thanks! I'll delete that now.
 
I should probably start cracking the new one before somebody else does.
 
go ahead! :) (not that it'll give you a huge score for the tie-breaker though)
 
@overactor Have you seen a new instruction which I might add?
@MartinBüttner I don't plan on being in a tie. I plan on winning.
 
11:40 AM
@PhiNotPi Jump or copy2?
 
both, I guess
 
@PhiNotPi that's the spirit!
 
It's interesting
Does jump make for shorter code?
 
I really like how motivated people are for the counter-challenges on these, although there is no rep incentive at all
 
Jump makes for faster code, mainly.
 
11:41 AM
as for copy2, should it have a parameter as well?
 
I was thinking about adding a parameter to be the amount of delay.
 
How many lines of code were you thinking per bot?
 
24 for right now
If I find a way to compress stuff, that should work out nicely.
 
the code limit in the original was a bit tight
 
I could probably do 32 if needed.
 
11:54 AM
0
A: Sandbox for Proposed Challenges

Von IlyaFind Lines of Symmetry Some characters are symmetrical to others. For instance: d is symmetrical vertically to b W is symmetrical horizontally to M O is symmetrical both ways to O ! is symmetrical horizontally to i (takes some imagination!) ^ is symmetrical horizontally to v (also takes...

 
If I were to add a parameter to Copy2 (or DelayCopy, whatever it will be called), how should I handle conflicts?
For example, a bot copying with a 2 turn delay, then copying with a 1 turn delay next turn, so that both copies occur at once.
 
hmm
whatever was posted first gets copied??
or do them in reverse order of when they were posted
 
I think that the oldest delaycopy will be applied last.
Do you think that any more variables are needed?
 
12:14 PM
0
A: Sandbox for Proposed Challenges

Vi.King of the server - virus wars There is a virtual machine with Linux running various services (SSH server, Apache (with SSL), FTP server, rsync server, etc); There is tool which checks that the server is running (periodically accessing all its services - trying to view the page from Apache, do...

 
@MartinBüttner Wrong?
Much sadness.
 
@PhiNotPi sorry ;)
How did you do it?
 
really?
 
@PhiNotPi I can't think of any
 
I made a list of the possible outputs for Mod[NextPrime[n],240] and then made all the combinations of three numbers.
I got {2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61, \
67, 71, 73, 79, 83, 89, 97, 101, 103, 107, 109, 113, 127, 131, 137, \
139, 149, 151, 157, 163, 167, 173, 179, 181, 191, 193, 197, 199, 211, \
223, 227, 229, 233, 239, 1, 77, 91, 119, 133, 143, 161, 169, 203, \
209, 217, 221, 121, 49, 187}
For the number of possible single values.
 
12:18 PM
@PhiNotPi Your range is correct (quite obviously), but there should be more possible combinations.
 
And I ended up with this list of combinations:
{3, 5, 6, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, \
37, 39, 41, 43, 45, 47, 49, 51, 53, 55, 57, 59, 61, 63, 65, 67, 69, \
71, 73, 75, 77, 79, 81, 83, 85, 87, 89, 91, 93, 95, 97, 99, 101, 103, \
105, 107, 109, 111, 113, 115, 117, 119, 121, 123, 125, 127, 129, 131, \
133, 135, 137, 139, 141, 143, 145, 147, 149, 151, 153, 155, 157, 159, \
161, 163, 165, 167, 169, 171, 173, 175, 177, 179, 181, 183, 185, 187, \
189, 191, 193, 195, 197, 199, 201, 203, 205, 207, 209, 211, 213, 215, \
 
@PhiNotPi That list is already missing 3. But even for that list, I get more combinations.
 
The list wasn't sorted.
 
yeah, it's missing 3 anyway :D
 
hmmm
 
12:20 PM
I get 589 combinations for that list though
e.g. you don't have 4, but you can get that from {1,1,2}
 
I get 7 for {1,1,2}
 
sorry, that's not what I meant
 
Yeah, f[{1, 1, 2}] = 7
 
from your list, one possible sum is 4 by choosing 1, 1 and 2. but 4 doesn't appear in your list of results
 
Noooo! I found my mistake.
much frsutration
 
12:24 PM
what do you get now?
 
589
 
yes, that's correct for your list of intermediate results (which is still wrong itself ;))
 
Okay whatever.
One major change that was suggested for Code Bots 3 was the addition of "non-executable" lines, or lines that would be skipped unless specifically called for by another line.
For example:
If D #1 #2
\\ Copy #3 *#E
\\ Move
\\ Flag
Next Line
So, one turn will execute If D #1 #2 while the next turn would execute Next Line.
 
That does make things more flexible
you also suggested that one block per line is max, right?
 
That's how it currently is, correct?
 
12:40 PM
no, if the blocks originate from different lines of code, they stack
But i like single blocks better for your specs
what about a modulo operator?
 
I was also considering a modulo operator.
 
it would be useful
 
How would operator precedence work?
Mod first, I assume.
 
yeah
mod is same precedence as multiply/division
which you likely won't have
Also
If I do
*E + E = 0
 
That way, Copy #12+E%3 *#*C would copy a random flag.
 
12:42 PM
what's the maximum value of *E + E?
@PhiNotPi yup
 
I think 46 would be the maximum value.
 
good
how is it in CB1?
 
Mod 24, that would be 22.
Also, I was planning on = and == having different meaning for integers.
In CB1, they do the same thing.
 
also, I would like to suggest two random values instead of one
and maybe < and >
 
= might mean mod 4 and == might mean mod 24
 
12:48 PM
not necessary if you have modulo operator of course
 
yeah
perhaps = could mean mod 24 and == could be without mod?
The mod for the equals would be applied after the addition/subtraction, rather than before.
Two random values could work.
I was thinking about < or >.
 
the two randoms would be useful to make two random choices independent of each other
< > would be great too
 
I wonder if % would allow that.
 
it would sort of
but it would be a hassle I think
what about AND, OR and NOT?
too much?
it would make for shorter code
but not by all that much
would mainly support more complex code
 
1:05 PM
I think AND OR NOT might be too much. Or... maybe I just don't know what would happen.
Maybe a <=> comparison which can go three ways?
I think that would have more benefit, since it fits more branches in less space.
 
@MartinBüttner I've forgotten a bunch of graph traversal stuff that I need to remember for HMSP :(
@PhiNotPi with your instant jumps, non executable lines just save one line of code, and only in one direction and one layout
@overactor you can already make two random choices if you know you have an opponent, by using E and *E
 
@Sparr yes, i did that in one of my bots
but what if you don't have an opponent?
 
then you've got plenty of time :p
 
@Sparr but not plenty of lines
 
@PhiNotPi I think you should concentrate on conceptual changes that remove freezing from the game. As neat as it is, I think that's a big weakness of codebots(1).
 
1:16 PM
@Sparr ah, that's unfortunate :(
 
@Sparr Not sure if I agree, it made for a lot of interesting bots
 
it's too powerful :(
 
and running the game enough removes bad luck
sure, you basically have to freeze
 
I think freezing should be possible in some form, just not a sure-thing like it pretty much is now.
 
but with some tweeks, what you do after you freeze could become more important
 
1:18 PM
making block easier/stronger would reduce the strength of freeze
 
The fast loop makes it easier to block all your lines.
 
I mean, I like being at the top of the leaderboard, but I'd rather there be more diversity
@PhiNotPi you mean the instant jump?
 
Yes.
 
I think bots that freeze and then do something interesting should become the way to go
 
Block #A
DelayCopy A+1 A
If A #3 #4
Jump 0
 
1:20 PM
that blocks 24 lines in 48 turns instead of 72 turns
 
*Block #A
 
Although you would need an extra line to break out of the loop.
 
you use an If to break out, doesn't slow down the loop
but someone can already duplicate the block/increment code if they want to loop faster...
 
Also, if I add DelayCopy with extended delay, that could be even more useful than blocking.
 
Block #A
Copy A+1 A
Block #A
Copy A+1 A
If A :loop :dontloop
Jump 0
 
1:22 PM
Also, DelayCopying your C value also lets you evade a block
 
that blocks 24 lines in 60 turns in codebots 1
 
The original is actually just as fast.
Jump executes the target line on the same turn.
Oh, codebots 1
 
in CB1, you can block 24 lines in 72 turns easily, 60 or 56 turns with extra code.
in CB3 you can block 24 lines in 48 turns
 
*CB3
sad that the 3D version is CB2 and the one that builds on the original is 3
 
so, blocking is a little faster, but I don't think it's faster enough to make blocking a major strategy
 
1:25 PM
Something like DelayCopy #A #A 23 would allow you to make the a portion of your bot reappear after being overwritten.
 
@Sparr You can also block only values that your C value will actually be set to
 
@PhiNotPi you could just modify CB1 controller to have instant jumps and see if that makes a blocking bot viable
 
Also, with sensing you know when you can safely block
 
@overactor slightly safely
@PhiNotPi another thing that would make code denser is a jumping version of If
 
@Sparr you know you won't get attacked the next round
 
1:27 PM
@overactor you know you won't get attacked between the sense and the block, but you could get attacked between the block and the action after that
 
what about replacing If as a line type by a ternary operator?
 
oh, duh, you can sense as an If and block on the same turn
 
@Sparr yup
or would nested ternaries be too powerful?
 
too powerful and much harder to interpret
 
@Sparr I see your point
 
1:29 PM
@overactor good work with The Common Cold, btw
 
@Sparr It's actually not my favorite bot by me, by a long run.
but thanks
Have you seen adaptive bot?
 
no. neat!
I'm gonna run a round with that bot
 
I don't think it will do great
 
grrr
 
but should be better than metaInsidious for instance
 
1:31 PM
someone posted a bot without a h1-name
so my getbots script broke
 
edited
two of COTO's bots had h2 titles
only accepting h1 is pretty naive though
 
that whole script is pretty naive :p
it just looks for a name in h1 and code in <code>
hell, originally it assumed the first <code>
and, of course, one of the new bots breaks the CodeBots script
 
I might be able to replace Jump with JumpIf for jumping only if a condition is true, otherwise it continues with the next line.
There could be an Age variable, incremented every turn.
Which would basically auto-cycle.
 
how about for If lines
instead of just two line numbers
 
Let's say age is G, then this could work to block in 24 turns:
 
1:39 PM
you can also add an optional Jump before each line
 
don't have time to debug and fix this
if anyone else wants to get get getbots and codebots working with the current entrants, please do
 
do you know which bot breaks codebots?
 
BlockFreezeAttack I think
somehow there's an extraneous " A" at the end of its last Action, if I read my debug output correctly
 
Block G
If G #2 #3
Jump 0
Jump 4
 
an age variable is a neat idea. would allow a lot of repeating behaviors with less code
would also allow for "do X only for the first 24 turns, then switch to Y" bots
 
1:42 PM
age modulo 24?
 
yes, age modulo 24
 
sweet
 
but remember you can multiply variables modulo 24, so you could easily have behaviors 24,12, 8, 6, 4, 3 turns long
 
you can multiply?
 
I think integrating Jump and If might not work. For example, If D *#*C #7 allows you to execute an opponent's line, but you can't Jump to an opponent's line.
 
1:44 PM
A*2 is A+A
A*3 is A+A+A
 
A*A?
or only constants?
 
only constants
 
still useful
 
I haven't added multiplication.
 
well, it's just shorthand
 
1:45 PM
one of my bots used E+E+E+E+E+E+E+E+7 to get a random number 7,15,23 which is where it kept its flags
 
That's creative.
 
nice
 
I thought so :)
 
I'd still like a modulo operator though
even if it's only with constant modulo
 
Modulo is officially in the spec for CB3.
 
1:51 PM
@Doorknob Should I work this into this?
18 upvotes and no downvote sounds like enough community consensus :D
 
@MartinBüttner Sure, everybody seems to agree. Just add another section.
 
Should the age variable be writable?
 
@Doorknob Why not separate answers? I think that's how they're doing it on meta.SE
 
Meaning that I could store something into it and it would be auto-incremented after every turn.
 
@MartinBüttner Doesn't meta.SE use one canonical answer? That seems to be the SE-wide policy for almost all faqs
 
1:55 PM
interesting, let me find out where I got the idea they're doing it differently
ah lol, probably because that's the case in all the faq's I've favourited there: meta.stackexchange.com/questions/59445/…, meta.stackexchange.com/questions/120576/…, meta.stackexchange.com/questions/67397/…
okay the second one is a bad example
and for the other two it's probably just necessary due to the size of the answers
interesting
 
@PhiNotPi I say yes
 
And, I think I'll change Jump [expr] to JumpIf [cond] [expr] to allow shorter loops.
 
@Doorknob done. please review my changes to the FAQ post.
 
I still like
If cond [Jump] line [Jump] line
better
 
@MartinBüttner Perfect.
 
2:10 PM
how exactly should that be formatted?
 
@Doorknob awesome. What about chilemagic's suggestion? I think a section like "How to review?" would be quite nice.
 
@PhiNotPi just like the current If, but with an optional "jump" before the two line numbers
JumpIf means fewer changes to an interpreter, and is almost as powerful, though
 
if A=B jump #7 jump #8
if A=B jump#7 jump#8
if A=B &#7 &#8
 
@MartinBüttner Sounds good; if you want to write it, go ahead. (This is all just stuff copy/pasted from the previous boilerplate, and I've actually been planning to rewrite some of it and add to it more later)
 
JumpIf would be easier, since If can actually be used to execute lines from opponents, but you can't exactly Jump to that location.
 
2:12 PM
okay, then the only thing I have left on my agenda is to convince everyone to delete posted challenges :D
 
@PhiNotPi I never thought of using If to run opponent lines :p
 
(the corresponding meta answer got another upvote btw, but 6 vs -4 is still no consensus)
 
Oh, I also wanted to make the sandbox revisions / comments preservation thingy, so that people who want to keep answers for educational value can be happy too. puts on to-do list
 
@Sparr I have but didn't find a usecase for it
 
What do you mean by "sandbox revisions / comments preservation thingy"?
I really don't think anyone uses that for educational value. If anything people will follow the currently developed proposals, but I highly doubt that anyone goes through old ones and reads old revisions and comments for that purpose.
 
2:17 PM
That was a terrible description. :P I think I mentioned this on meta once: if there was an automatic process that copies the revision history and comments of finished sandbox posts and hosts them elsewhere, that stuff could still be preserved.
 
Ahhhh, I see.
 
If I remove the ability for If statements to execute opponent's lines, then they can be given integer arguments rather than lines.
 
Yes, that would be good (and then you can monitor the traffic of that and tell me how many people use it :P)
 
At the very least, if posted challenges do get deleted, I would go back and undelete all the deleted comments because they wouldn't be in the way anymore.
 
By standard SE practices, comments which have been worked into the proposal should always be deleted anyway as they're not meant to stick around with important information.
 
2:19 PM
Taste Of Your Own Medicine Bot
If D *#*C #2
Jump #0
Copy D+1+E%3 D
 
If the posts are deleted, though, it wouldn't really matter. Keeping the comments seems like a reasonable compromise.
 
Well the thing is that even deleted posts are still clutter when sorting by active.
 
sadly that doesn't work of course
 
Also, I was thinking Jump would take a line number (not a line) as its argument.
 
ah, makes sense
to avoid enemy lines
 
2:29 PM
A JumpIf would also take a line number, although that is somewhat inconsistent with the If taking lines.
 
jumpif could do the jump and the if separately, conceptually
 
Do you think I should go ahead and replace Jump with JumpIf?
 
Why not both?
having to write JumpIf 0=0 2 the entire time seems annoying
 
Okay.
 
ah, i know what Sparr meant now
Jump [If cond] lineNum
 
2:41 PM
@Doorknob I've added a "How to review" section, but that one definitely needs some work I think.
 
Having an If statement which can be appended to any command?
 
What about the proposal template? I think that would better fit in a second answer.
 
[command] [If cond]
Although that would be awkward if the command was another If statement.
 
I'd only do it for Jump commands
otherwise those are basically ternaries
 
Jump [linNum] [cond], with [cond] being optional
 
2:51 PM
needs feedback/work:
0
A: How does the Sandbox work? How do I use it?

Martin BüttnerTemplate for Challenge Proposals Every proposal should contain several fixed pieces of information, including a title, the tags you intend to use and body of the challenge as you intend to post it on main. For a more uniform look of the sandbox (and easier orientation of reviewers), it is recomm...

 
adding a status to the title also seems like common praxis
#[Challenge Title] [[Status]]*
As in:
 
huh, is it? I've never done/seen that.
 
#Draw butts [Incomplete]
or
 
okay, there are three such challenges on the first page
 
#Output Graham's number [input wanted]
6 in the summary
not saying it should be added, but it could
 
3:47 PM
0
A: Sandbox for Proposed Challenges

Beta DecayRoulette king-of-the-hill Roulette is one of the simplest casino games to understand, so I invite you all to pit your bots against each other to see who gets the most money. Coding You will be supplied with a singe argument via the command line. This argument will be the amount of money you h...

 
4:11 PM
@Rainbolt did you make any progress on the Tetris library? I might start work on the controller today.
 
Regarding a Roulette KOTH, I don't see how that could work.
 
I'm considering adding an option to the controller which lets you write a protocol of the game to a file (without timing information), such that someone (potentially myself) could write a visualiser which reads in the protocol and renders a replay of the game. People could then use that to include a GIF of the first x rounds of their bot playing in their answers (for whatever x gives interesting result without blowing up the file too much).
@PhiNotPi Dennis cracked my Mathematica answer, so I can tell you know what you missed: negative inputs ;)
Adding negative primes to the list gives 3 additional possible values module 240.
 
Negative primes.
Hmmm.
 
(I believe those are 238, 237 and 235 but I didn't check)
 
Yeah, those aren't real things.
 
4:21 PM
Yeah, but Mathematica's NextPrime returns them for negative inputs ;)
That was also why my first answer was broken, because I had used Prime there, which doesn't work for non-positive inputs.
 
4:38 PM
Does seem like something that could be worked into a challenge?
 
Rule 34 of PPCG: if it exists, it can be worked into a challenge.
@PhiNotPi one idea: pose two problems which both need to be solved. then use either a restriction on or a combined scoring of code size and levenshtein distance between the two solutions.
that is a double code golf in which a few "typos" would change your program from one solution to the other.
it's important to not make it levenshtein distance alone or answers will just be if(0)program1;else program2; and if(1)program1;else program2;. you somehow need to incentivise to reuse as many characters as possible.
afk for a bit
 
5:07 PM
Perhaps it can be the size of one program plus the levenshtein distance to the other.
I think size of shortest program + distance to longer.
 
sounds good to me
the only problem I see is that I still wouldn't want to use as a tag, because a) it looks like a typo and is likely to be fixed by eager users and "f" and "l" aren't even that visually distinct, so people might just overlook it all together.
maybe something like would be more appropriate, and then you can call the first such challenge something like "Code Gofl: Accidentally golf the wrong problem" (or something more descriptive about the two problems you choose)
 
or to be a bit more subtle
 
@cjfaure very memorable
 
If it were longest program + distance to shorter, then people would increase the length of their shorter program to decrease the distance.
 
5:22 PM
I am definitely using a misspelling of Code Golf in the title.
Choosing the two challenges will be hard.
 
yes, I can imagine
if they are too similar it's too boring, it they are too distinct, the levenshtein distance is likely to tend to the code length (although I'm sure some people might surprise you). and they should also be solvable in a similar number of characters
 
I honestly don't know what challenges to do.
It would be best if the two challenges are close typos of each other, as well.
 
you can probably look for old plain code golfs and reuse two from different fields (string vs numbers) which had similar answer sizes. the twist due to the levenshtein distance should make the challenge sufficiently different not to be a duplicate.
@PhiNotPi heh, good idea
 
5:39 PM
The only "original" code golf idea I've had in a while is to find Golomb Rulers. Golomb Ruler <-> Golem Ruler?
 
what's a golem ruler?
 
I don't know.
 
00:00 - 18:0018:00 - 00:00

« first day (1311 days earlier)      last day (3525 days later) »