« first day (2752 days earlier)      last day (2081 days later) » 

12:43 AM
@NewMainPosts That's new, normally it's people thinking this place is SO.
 
0
Q: Any Sprite Animators

ProfessorSamuelOaki am really really sorry if this isn't allowed here.. I don't really know where to find a site to ask this question but, i am so so sorry if this breaks any rules but.. Are there any sprite animators? Other than 'Pivot Animator', are there any other animators that support images, sprites and ...

 
0
A: Sandbox for Proposed Challenges

Chas BrownalphabeTree Challenge When looking at a list of words, sometimes it seems like a lot of redundancy is present. Consider the following list of words: balderdash ballet balloonfish balloonist ballot brooding broom All of the words start with b, and the first 5 start with bal. If we just look a...

 
1:00 AM
Wait ... the NMP got starred?!
 
 
4 hours later…
Anonymous
5:18 AM
user image
6
 
5:48 AM
FYI: sandbox viewer is fixed for new header
2
 
lol
6:33 AM
Hi
I made a Google search bar on my laptop, idk why but I like it
Oh, and the mic icon isn't functional, it's just a part of the bar
 
lol
6:53 AM
:D
 
 
4 hours later…
10:42 AM
@ASCII-only i have question about charcaol
 
10:52 AM
NodeMCU is so awesome. Like Arduino, but better and with Lua (or Python).
 
@Downgoat Tex doesn't work in the Sandbox Viewer
 
11:41 AM
0
A: Sandbox for Proposed Challenges

NeilMake a reversible formula For the purposes of this question: a basic formula \$ y = f(x) \$ takes one of the following forms: $$ x \\ g(x) + c \\ c + g(x) \\ g(x) - c \\ c - g(x) \\ g(x) c \\ c g(x) \\ \frac {g(x)} c \\ \frac c {g(x)} $$ where \$ g(x) \$ is a basic formula; a reversible f...

 
 
2 hours later…
1:48 PM
Oh I hate this illusion
A demo of lightness perception https://t.co/BSVpgcuIw1
 
 
4 hours later…
5:40 PM
2
Q: Should competitors be able to run games for KoTHs?

Redwolf ProgramsIn this question, every "game" takes about 5 seconds. Ten thousand runs, maybe eight thousand, are planned. Running these on a single computer would take forever, so I suggested letting competitors and volunteers run some of the games. However, this result in cheating. My question is: should this...

 
6:24 PM
TFW you spend several hours trying to solve a problem and it goes away by itself
 
@Pavel Yep
 
I actually gave up and was already placing a Fedora installer onto a flash drive
But then the problem went away just as I was about to reinstall the entire OS on my computer
My mouse cursor was randomly teleporting all over my screen every few seconds and I couldn't figure out why
 
7:20 PM
Language Design Challenge: Make a language with this concept at its core: Range
8
 
@EriktheOutgolfer What do you mean?
 
Terms of the LDC: After August 20, 2018 in your timezone, the earliest possible if DST adjustments are to be made, the Challenge will be closed. Any language is allowed to participate, as long as it is your own. However, for the sake of the spirit of the Challenge, the Language must follow the Challenge's description (Make a language with this concept at its core: Range) reasonably closely.
@Zacharý make a language based on ranges :P
 
@EriktheOutgolfer as a distinct concept from arrays, right?
 
not all arrays are ranges
not all ranges are arrays
 
...
 
7:30 PM
I think you're probably not operating on the same definition of range that Zach or I are
(you're the D person right)
 
Rogue goat may have helped dozens of farm animals escape https://nyp.st/2OR36VP
 
@Downgoat i pinged you with one of these stories ages ago and you never told me what the deal was
 
@quartata what when
 
Aug 3 at 16:20, by quartata
 
@quartata Yeah.
 
7:38 PM
@quartata definetly not part 2 of goat world domination plan
 
@Downgoat What was part 1?
 
Downgoat was part 1
 
@quartata hm, what do you think? "range" is only a word
 
@EriktheOutgolfer Do you mean a subset of arrays where elements are consecutive? Or an entirely different concept?
 
if it can be described as "range" (well, reasonably, as mentioned), it's valid, the spec doesn't say anything more ;-)
 
7:45 PM
@Pavel it happened in my city too 😛
 
@EriktheOutgolfer Maybe something where data types are ranges, and functions are mappings from one or more ranges.
Let's make up some syntax,
 
@Adám there's the range datatype in Python 3 (xrange in Python 2)
that's why I said above that not all ranges are arrays
 
@EriktheOutgolfer Right, but this language has nothing called an Int.
 
@Adám Python? how?
of course there's int
 
@EriktheOutgolfer No, I mean the new imaginary language to solve the challenge.
 
7:49 PM
So ... sorta like R, how scalars are just 1-element vectors? (blech)
 
b[s]e denotes the range from b to e with step-size s. And b(s)e is the exclusive range.
 
@Adám looks like you've already started thinking :P
 
@Zacharý No no, a vector isn't a range. The closes you'd get to a scalar 42 would be 42[0]42.
 
(also, nothing says it must be "new")
 
@EriktheOutgolfer AFAIK, no language has the range as its fundamental concept.
 
7:51 PM
@Adám idea: what if there are also b(s]e and b[s)e?
 
@EriktheOutgolfer I thought that was obvious.
 
(note: ( is not very usual)
 
You can make a "table" as a range between two ranges: a[b]c[s]x[y]z
 
@Adám how about extending the definition to higher dimensions as well?
 
@EriktheOutgolfer I thought that was obvious.
 
7:55 PM
"I thought that was obvious" count: 2
 
@Adám sorry, I guess I can't really guess everything :P
 
Anonymous
@EriktheOutgolfer I guess I'll actually work on ibmf :P
 
@Mego International Body Music Festival?
 
@Mego Has that been a thing before this challenge, or is "ibmf" some acronym I'm unaware of
 
@Adám so, a "range", a "table", a "stack", a "storage room", a "warehouse", a "complex" and an "industrial city" are the first seven dimensions? ;)
 
Anonymous
7:57 PM
Infinite Byte Matrix Filter - the tarpit language idea I came up with a few weeks ago and never really did anything with
 
We can have "strings" as well: "A"[1]"Z" is the uppercase alphabet.
 
@Adám 'A[1]'Z seems to be better, no characters/strings involved; there is only sugar.
 
Anonymous
You perform operations on an infinite array of range(256) lists to transform input to output
 
@Zacharý Sure, that'd work.
 
@EriktheOutgolfer I presume that you were referring to Mego's language?
 
7:59 PM
lol no
@Zacharý eh, 'A[1]'Z looks like it's going towards the golfing direction, so, whatever @Adám thinks looks better :P
 
@EriktheOutgolfer I'm not sure I like your names: range, field, barn, silo. After that, it is probably easier to use numbers for the dimensions.
 
This answer is very underrated imo so just posting here:
9
A: Upgoat or Downgoat?

qwrOpenCV with Hough Transform, 100% My original idea was to detect the vertical lines of the goat's legs and determine its vertical position relative to the body and horizon. As it turns out, in all the images, the ground is extremely noisy, making lots of Canny edge detection output and corresp...

 
@Adám but barns can't contain fields, can they? ;-)
> barn: a field that contains smaller fields
 
@EriktheOutgolfer Hm, ok, how about 1D being range, 2D being field, 3D being farm?
 
What ever happened to table?
Just call them all ranges, 2D range, 3D range, etc...
 
8:04 PM
@Zacharý Sure, they are all ranges, but it is easier to say matrix than rank-2 array. So too, ranges of ranges can informally be called fields.
Btw, we don't need indexing, just a intersection function.
Question is whether intersection goes on value or index (i.e. n×step). I'm leaning towards the latter.
 
How would operations like + work?
 
@Zacharý Obviously only works on ranges with the same number of steps, and then it maps.
E.g. 10[10]100 ∩ 4[0]4 gives 40[0]40 Edit: no, that can't work.
 
How would 1[1]3 * 1[1]3 be displayed? It isn't really a range...
 
@Mego I can see some hope in there, as long as the operations make sense, and I can view the whole output of at least one program ;)
 
@Zacharý I'd say lazy evaluation until value is displayed, and at that point it is converted to a list.
 
8:19 PM
I think I have an idea for output (different notation though, {a;b;c} represents a{b}c where {} are brackets/parens): [1;1;3] * [1;1;3] => [1;[2;3;5];9], having it cycle through the steps.
Leading to very crazy output possibilities, good if this is going to be esoteric.
 
@Zacharý I don't get it. The inner bracket is a list?
 
@Zacharý if you're referring to Adám's language, I think you mean 1[1]3 * 1[1]3
 
@EriktheOutgolfer Same language, just a different notation.
 
Or, in Adám's notation, 1[1]3 * 1[1]3 => 1[2[3]5]9, having the range's step cycle through the elements if the "rank" of the step is higher than the "rank" of the endpoints.
 
@Adám oh, I thought Zacharý was referring to an output notation and confused the code with the output a little
 
8:25 PM
That way there's no such thing as lists.
 
I still don't get it. 1[1]3 * 1[1]3 should give 1[3[2]5]9.
 
WHOOPS.
pretend this never happened
It should be 1[3[2]5]9, I screwed up >_<
 
[1, 2, 3] * [1, 2, 3][1, 4, 9]1[3[2]5]9, 1[3[2)7]9, etc.
(inclusive ranges are better for output notation)
 
OK, this may actually work. For every sequence can be constructed based on shorter lists.
 
@Adám concatenation?
 
8:28 PM
Every "list" could be constructed with ranges.
@EriktheOutgolfer Then it couldn't be represented as one range.
 
@EriktheOutgolfer No, deeper specifications of steps.
 
@Adám yeah, the key is in the steps
 
How would [[1,2],[3,6]] be represented? I'm concerned with ambiguity, and since we already use ()... what parentheses pair should it be?
Maybe spaces?
 
The list 3,1,4,1,5 would be represented as 3[-2[5[-7[11[-24]-13]-9]7]4]5 (!)
@Zacharý A matrix?
 
Shorthand... for jsonish representation of a matrixy-range (no clue what it would be called).
 
8:33 PM
lol see the debate above
 
OK, so [1,2] is 1[1]2 and [3,6] is 3[3]6
 
how do we go from [1,2] to [3,6] though
 
Notice how I mentioned "rank" before
 
@EriktheOutgolfer Easy, the step size just needs to vary from 2 to 4 i.e. 2[2]4
 
Would 1[1]2 [2[2]4] 3[3]6 work?
 
8:35 PM
@Zacharý Yes, but you don't need those spaces.
 
@Adám Then it might parse as (using <> for grouping): <<1[1]2>[2[2]4]3>[3]6
 
1[1]2[2[2]4]3[3]6 is unambiguous. Higher rank is denoted by more parens on the outer level, while dynamic step sizes are denoted by nested parens.
 
Wait ... how would 1,1,1,1 be represented?
 
@Adám but 1[1]2[2[2]4]3[3]6 will result in [[1,2],[3,4]], no?
 
@Zacharý How is that a meaningful range?
@EriktheOutgolfer Yes.
 
8:38 PM
@Adám I assume we're going in the esoteric direction?
 
lol I'm pretty sure a language based on ranges is alone esoteric enough
@Adám then how do we go about [[1,2],[3,6]]?
 
@EriktheOutgolfer Wait, I misread.
@EriktheOutgolfer No, the last range will be 3[3]6.
 
With our current direction, we can't represent anything with consecutive identical elements.
 
@Zacharý 1[0]1. They are functionally equivalent :-)
@Zacharý Of course not! Such is the nature of ranges.
Our languages influence how we think. If you use a language that only has lists (and lists of lists etc.) then you cannot fathom a 0-row 3 column matrix.
 
@Adám Poor souls whose languages didn't allow lists of lists nor multidimensional arrays, having to do every thing the mod way.
 
8:43 PM
In Ranger (preliminary name…), the concept of two identical consecutive elements simply doesn't exist.
 
Is Home on the Range taken?
 
@Zacharý And yet, such languages exist.
 
Yep
 
The standard implementation will of course be called StRanger
 
LOL
 
BWO
8:46 PM
@LDC: So a range basically enumerates stuff. What if stuff becomes a function?
 
1,2,2,1 can be represented: 1[1[-1[2]1]1]1
@BWO Like APL: functions are entirely different things.
 
BWO
Ideas for stuff: some Goedel-numbering of programs, Ackermann's 3-argument function, list of cycle through [+,-,*,/]
 
No stinking lists in Ranger!
 
Maybe we could only have infinite ranges, making it so we have a dyadic range constructor: start[step]
 
BWO
@Adám Well what's the difference really? It's really just a question of implementation, right?
@Zacharý I don't understand
 
8:49 PM
@Zacharý No no, just allow infinities. Natural numbers are 1[1]∞
 
@Adám , -∞, floats
 
@Adám Oh ... you're one of those people.
Instruction numbering might work...
 
@EriktheOutgolfer [] should denote a step size of 1÷∞ or the appropriately ranked range.
Floats are -∞[]∞
 
strangely enough, [] can be represented in an infinite number of ways: a[c]b, where sgn(c) ≠ sgn(a - b)
 
The LDC needs this to be implemented, no?
 
8:52 PM
yes, otherwise it's just a concept of a language and not the language itself ;-)
 
So, who's going to implement this horror?
 
@EriktheOutgolfer Or just 0(0)0
 
@Zacharý I believe IEEE-754 floats are actually countable
either that, or a specific precision limit
also, it's the L that's implemented, not the LDC :P
 
If we only have infinite ranges (as literals) can be used to get finite ranges: 1,2,3 => 1[1]∩3[-1]
 
@Zacharý Ugh, I really dislike the asymmetry.
 
8:58 PM
@EriktheOutgolfer I'm referring to the parsing as well: it might be possible to not have to deal with floats if we use sympy.
 
0
A: Sandbox for Proposed Challenges

Rushabh MehtaA Knotty situation Today's challenge will involve some interesting knot theory, and involve you calculating one of the most important objects in all of knot theory. But before that... Background Although there are more technical definitions, for this challenge, it is enough to think of a knot ...

 
@Adám [1|1]∩[3|-1]
 
@Zacharý That's ambiguous, no?
 
@Adám How?! [start|step] everything is enclosed in [], so ambiguity there shouldn't arise.
 
you might want to make a variant
 
9:01 PM
Much less ambiguous then having to determine rank on-the-fly for parsing.
@EriktheOutgolfer What do you mean?
 
a variant of Ranger...it's your decision after all :P
 
RangerRangerRangerRanger...
 
@Zacharý Rank is simply ⌈number_of_outermost_parens÷2
 
@Adám But for simply parsing?
StRanger and RangerRangerRangerRanger...
 
@Zacharý I don't design languages with the parser in mind. Let the implementer suffer.
 
9:04 PM
lol
 
@Adám What implementer? This isn't Dyalog :p.
 
also, parsing a[b]c for whatever b should be easy if you use recursive regexes ;)
 
For b, yeah, but what about for a and c? How would a[b]c[d]e be unambiguously parsed? Could be (a[b]c)[d]e or a[b](c[d]e)
 
@Zacharý a[b]c[d]e isn't a meaningful range.
 
a[b]c[d]e is invalid...
 
9:08 PM
There must be an odd number of parens on any given level.
 
Doesn't matter: that's why "colorless green ideas sleep furiously" is a thing.
(or colourless)
 
ngn
@Adám k5 had ranges, most likely k6 has them too
 
@ngn And k7?
 
Digit ::= 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9
PosInt ::= Digit [PosInt]
Number ::= [-] (∞ | PosInt)
Range ::= Number | Range \[ Range \] Range
the parser will most probably accept things like 3[7]5[9]11 too; it's not the parser's job to verify that
@Adám that isn't necessarily valid either; I'm not really sure about 1[2]3[4]5[6]7[8]9[10]11
the number of left brackets ((/[) must be a power of 2 minus one
 
@EriktheOutgolfer Ah, yes, of course.
@Adám Doesn't recognise linear relationships.
 
9:27 PM
ListToRange (now handles linear sequences)
 
9:38 PM
Anybody up for implementing the inverse, RangeToList?
 
@Adám might be a good challenge...
 
@EriktheOutgolfer Both ways may, actually.
 
@Adám ListToRange might be supported with the condition that "the output should contain the least possible amount of numbers"
 
@EriktheOutgolfer Of course.
 
where "numbers" means actual numbers, not digits (some people might get confused by this)
 
9:44 PM
@EriktheOutgolfer So far, range is only good at one thing: Concisely expressing a list of numbers where the n-th derivative is linear.
 
@Adám or characters
("A"[1]"Z" is the alphabet)
 
@EriktheOutgolfer And "!"[1[1]13]"|" is !"$'+06=ENXco|
 
@Adám I think that's true for all finite lists of numbers (1D only :-( )
 
10:23 PM
@LeakyNun How can that be solved?
 
11:15 PM
@Adám So, polynomials... (for finite lists, everything is linear: just like you can fit a polynomial to any amount of points)
 
nah, not really
[1,1,1,1] is also a polynomial ;-)
 
@EriktheOutgolfer y=1 ... sorta.
 
Well, f(x)=1 cannot distinguish between {1,1,1,1} and {1,1,1}, so that makes sense.
 
Maybe polynomial interpolation will help here ... is there anything on fitting a set of points exactly to a polynomial in more than one dimension?
 
11:52 PM
@EriktheOutgolfer ^ is my language so far. I can't make a wiki for now because it's too late but I'll probably be able to explain it tomorrow. The code is pretty readable though, for those who know Python
Also I forgot to add some important function but those can pretty much be extended at any time
Updated interpreter. A minor indexing convention issue. I forgot to mention, the language is called Rrange!
 

« first day (2752 days earlier)      last day (2081 days later) »