« first day (3245 days earlier)      last day (1598 days later) » 

12:30 AM
@WheatWizard Hmm, I think it may be possible in Backhand with some sort of convoluted structure
though it might be tough since you can't have two of the same character in a row
 
Oh yeah that is a good point.
 
it would probably be easier when I finish my next update of Backhand, which has the get/put instructions
 
@flawr Curious - what's your current strategy? mine is to form 2s from 1s, 6s from 2s and 3s, and play it like 2048 from there. I'm not having much luck though
Aw, just scored 773, not sure I'll do better for a while
This is addicting
 
12:58 AM
scratch that, I don't think it would be possible in Backhand, my strategy doesn't really hold up
 
1:11 AM
@WheatWizard Just checking, can we read the source code?
 
Ah I didn't think of that.
I suppose in fair spirit yes.
 
@WheatWizard Then I suppose it wouldn't be terribly hard to use a 2D answer from the Geiger Counter question, like the ><> one and check the second line against the first (with some hardcoded checks for the necessary differences and the removal of the newline)
 
 
2 hours later…
3:16 AM
0
A: Sandbox for Proposed Challenges

Wheat WizardLost Compass The idea for this challenge is pretty much complete. I just want to create a few tools to help in building and debugging, before I post this. Lost is a 2-Dimensional programming language where the instruction pointer starts at a random location, moving in a random direction. A...

 
4:01 AM
@Jono2906 You are describing the pipe | operator.
 
@A̲̲ I am still uncleared on a few things
Is it code-golf?
 
@WheatWizard Yes this is.
 
4:18 AM
0
A: Sandbox for Proposed Challenges

A̲̲This is a repost of the underspecified challenge. Task: Remove all duplicates from one list. A list is simply a sequence of connected values that allows the same values to be stored at different positions in this sequence. (If Quintec agrees with me, I will limit the possible values to only con...

 
 
4 hours later…
8:02 AM
@Quintec I try to make 6es, and then power-of-2-multiples of 6es. so 6-12-24-48 (always in the corners or at the sides if possible), but every now and then you get in a situation where you are forced to make a 4 or so, and things go downhill from there
 
 
2 hours later…
10:03 AM
7
Q: Create a diagonal sudoku board

Stewie GriffinA Diagonal Sudoku board is a special case of Latin squares. It has the following properties: The board is a 9-by-9 matrix Each row contains the numbers 1-9 Each column contains the numbers 1-9 Each 3-by-3 sub-region contains the numbers 1-9 Both diagonals contains the numbers 1-9 Challenge: ...

 
CMP: How much rep have you bountied away?
Me: 6,900
 
10:17 AM
I think I once gave out a 100 (was it 200?), nothing else
 
@Bubbler 100
 
Me: 0
 
10:45 AM
I have a question about challenges
If the byte removed causes a syntax error, does that version count?
In other words, is an entry valid if any of the possible permutations syntactically invalid?
 
@Adám is this number displayed somewhere or do you have to compute it yourself?
 
22
Q: Pass on your radiation

Wheat WizardThe task here is pretty simple. You should write a program that takes a string as input and outputs it (that is a cat program). Additionally when the \$n\$th byte of your program is removed (without replacement) the augmented program should take a string as input and remove the \$n\$th characte...

 
ngn
@Adám 4700
that's almost 1/3 of my total earned rep
 
Ah yes, philanthropy should be measured in %, not abs amounts.
 
ngn
11:00 AM
s/philantropy/scientific experiments on humans instead of lab mice/ :)
or s/philantropy/apl advertisement/ in your case
2
 
I don't get any of those two.
 
ngn
you give away rep to popularize apl, right?
 
@ngn ah, yes, but that wasn't my emphasis with the cmp
 
ngn
i do it for the fun of seeing people struggle with a good challenge (i'm very selective about what i bounty) and often i get unexpected results
like those scientists who put mice in mazes with a piece of cheese on the other side of the maze
 
ah, now i get it
 
ngn
11:11 AM
i wonder if failed bounties (timed out without being awarded) are counted in the score
 
Don't they go to highest scoring answer?
 
ngn
@Adám i think so - they are halved and automatically awarded, but what if there aren't any answers
 
ngn
@Adám but you still lose the rep
 
Yes, which means that rep can be destroyed. (Well, it can through downvotes too.)
 
ngn
11:20 AM
it looks like they are counted. for example the +500 bounty i offered for a knotty situation is in my list but it hasn't been awarded to anyone. (later i received a separate +50 bounty for my solution)
 
@ngn Hm, that's like doing a tax deduction on burning money.
 
ngn
:D
the original author didn't have much rep, so actually this should be considered generous
 
 
1 hour later…
12:36 PM
@Adám ah thanks! (nice link:), so for me that's 0:)
I wrote a question on stats.SE, but have gotten no answer so far, maybe someone here can help? stats.stackexchange.com/questions/386975/…
 
 
3 hours later…
3:31 PM
@Jono2906 It depends on the question. radiation hardening is not a standardized tag, removing a byte isn't even the only type of radiation.
 
@Adám 400 + 700 available without deadline
 
@Adám 9,900
 
@cairdcoinheringaahing Oh well, I currently have +29,200 without deadline.
@WheatWizard Wow.
 
@Adám That is a lot :P
 
I believe that I have awarded the most rep in bounties on the site, but don't quote me on that.
 
3:39 PM
@WheatWizard I think Adam's gunning for that record :P
 
@cairdcoinheringaahing Who needs rep above 10k anyway?
 
@Adám my ego :)
you never know, maybe after raising the question reps they will soon start lowering the answer reps
 
4:38 PM
Man, having PS ranges behave when you have a joined interval (e.g. $n..0+1..$n) and need to handle inputs of [1,whatever] sucks
 
5:21 PM
Special case the 1 and call it a day
 
Pretty much
 
At least it's not
(hopefully)
 
It always is ;~;
 
5:40 PM
I wonder, if you had already answered a question, and came up with an another solution with similar/identical approaches in the different language, should it be posted as another answer, a part of the existing answer, or would it be frowned upon to post altogether?
 
@Naruyoko I usually post 2 answers
I've seen some challenges where 1 person has posted 10 answers, all in different languages, all with the same short algorithm
 
Yeah, usually people just add a link to their 1st answer saying "port of my foobar answer"
 
Ok, thank you!
 
@Naruyoko I occasionally post an answer in APL and in J where the only differences are in spelling, everything else being the same.
 
6:04 PM
2
Q: Create a singular matrix of integers

DoorknobGiven an integer n ≥ 3, produce a matrix with the following properties: The matrix has dimensions n × n. The matrix contains each integer from 1 to n² inclusive exactly once. The matrix is singular. There are several equivalent definitions of this term: The determinant is zero. The matrix is ...

 
@NewMainPosts @flawr just blew this challenge wide open :/
 
@cairdcoinheringaahing my poor 10 green internet points :(
 
Shame it got deleted, had a 2 byter in Jelly :(
 
maybe it would still have been a fun simple challenge if it was just "print 1...n^2 in a square shape"
@Doorknob I heard there is this thing called sandbox, have you heard about it?
:P
@WheatWizard Is your challenge actually solveable using currently known methods?
 
6:24 PM
@flawr Which challenge?
 
@flawr lol
yeah, i probably should use it more often
it did not occur to me that the trivial solution would just work
 
6:45 PM
-3
Q: Simple Question! how can i dont let the behind side data in same place?

Royal If my input data too long,how can i write for solve the behind side?

 
I get that it's probably an English-as-a-second-language thing, but I can't understand what NMP is asking.
 
@AdmBorkBork I think the complete lack on context might contribute as well
 
Well, from the screenshot it's a basic book inventory system (a relatively common beginner coding exercise).
But beyond that, I'm really not sure. :-/
 
7:25 PM
@WheatWizard the knotty one:)
 
I'll assume you mean the one I didn't already solve :)
I think it is possible, I have not personally solved it but it seems like the sort of thing that should be solvable.
You could likely put an upper bound on the number of Reidemeister moves required and then just sort of explore the possibilities exhaustively.
 
ah you're right
there is no polynomial time algorithm however (for the unknot problem)
I confused it with that
 
8:12 PM
@flawr Just got a 1805.
 
@Bubbler nice!
so what is your strategy now?
 
Have y'all tried to score low? I got 1 35
 
Miseré, eh?
 
@El'endiaStarman misère? :)
 
Misery?
 
8:14 PM
@flawr [googles] Oh, you're right.
@AdmBorkBork Common term for games where you try to lose.
 
Same. I put the highest tile at (4,4), but I doubt it's the best
 
@El'endiaStarman Interesting. Never heard of such a thing. I've played games like that, though, like whist, but I've not heard the term.
 
@Bubbler do you have any tactics for avoiding the "checkerboard" (getting to a point where no two adjecent squares are equal)?
 
@flawr Nah.
 
8:36 PM
@KritixiLithos (belated response) Awesome! By (read) do you mean user input? I don't have any plans for that in tinylisp, no. There is basic user input in the related language Appleseed, though it's syntactically verbose and poorly documented. :P
Standard practice in tinylisp golfing, therefore, is to submit functions rather than complete programs.
Except for challenges that don't require input, as I see you've already discovered. :)
 
9:14 PM
@JoKing As a bit of a proof of concept `))"!"))"E!"))@))"!W"))"!"))@` Will output the direction it started, if that direction is east or west. Transposing it will create a program that detects N/S.
From here you can stitch them together like:

#X
X#

Where `#` is E/W detectors and `X` is N/S detectors. Some extra work is actually needed to fix some edge cases but this *almost* works.
To clarify # is a giant block of detectors with one on each row, stacked to be high enough that every row of X has one detector on it. Likewise with X but columns instead.
 
0
A: Sandbox for Proposed Challenges

caird coinheringaahingScore a 1 player game of Carcassonne Carcassonne is a tile-based game, where the objective is to construct Roads, Cities and Monasteries, in order to score points. The game works by players taking turns to draw and place tiles to construct a landscape, then claiming roads, cities and monasteries...

 
Oh wait I got ( and ) confused, all ) I used everywhere should be replaced with (.
 
10:15 PM
@WheatWizard Yeah, I was thinking pretty much the same thing, but I was trying to figure out how to make it overlap efficiently.
 
I think I need to post the challenge before I solve it myself.
 
10:29 PM
@WheatWizard The Lost verification could be a lot more efficient, right now it loops over every single position with every direction. Really, you only to execute once for directionals (><v^), and you only need to verify once in a consecutive run of no-ops (I think)
 
Yeah, I am currently working on a new Lost interpreter.
It will have more efficient verification amoung other things.
 
noice
@WheatWizard You know, I've had a couple of thoughts about a variant of the language called Forgotten that would be a radiation-hardened version, i.e. one random character is removed at the start of execution
 
Both random execution and the deletion or only the deletion?
 
just the deletion. Though the pun would be if it was a polyglot than it would be Lost and Forgotten
 
I have been proto-typing a version of Lost where all of the operations are also non-deterministic.
 
10:33 PM
Lost and Confused?
 
That was one of my title ideas.
Are there any special hurdles that need to be solved for a language like forgotten? e.g. for lost we need safety because otherwise any program could have a chance of instantly halting.
 
@WheatWizard I don't think so. Generally you can just use a simple format to have two copies of the code and only execute the non-irradiated one
Not sure how this might extend to multi-line programs though
 
I'm also going to change how input works for the new version of Lost.
The current method was not very well thought out and as a result it is very hard to do non-trivial programs that take input.
For example I don't think you can do a fibonacci program unless you have a sufficiently contrived IO scheme.
 
10:52 PM
Is it just me or the beginning of this song is absurdly similar to a theme song in Hearthstone?
 

« first day (3245 days earlier)      last day (1598 days later) »