« first day (3796 days earlier)      last day (1041 days later) » 

10:00 AM
isn't popcon considered a valid wincon? as un-objective as it is
IIRC we haven't banished popcons like trolling
 
they're just discouraged
heavily
 
@lyxal does vyxal have a base conversion that accepts floats
like convert floating point list from base (integer)
 
CMC: Generate a random sequence of letters (length 5). Example "qwask", 'wertya', 'gffga'
 
all lowercase?
 
yes
 
10:08 AM
i have ØaX$5СḊ but it can probably be shorter (jelly)
yeah 5ØaX¤€
 
@hyper-neutrino Pop-cons are generally considered to have an objective winning criteria if there is clear criteria of what voters should and shouldn't vote for
 
@PyGamer0 Ruby, 22 bytes : [*?a..?z].sample(5)*""
 
@cairdcoinheringaahing But "Do this creatively"? Nope.
 
@hyper-neutrino Øaṗ5X... not that I've seen it produce a result
 
10:16 AM
But pop-cons are generally difficult to close, as they're borderline ok, and there's a lot of precedent of leaving one of two similar pop-cons open and closing the ither
 
The last non-closed pop-con was six months ago, and you wrote it well and gave a clear idea of what to look for.
 
Yeah, and even that was borderline ok on objective winning criteria, as a lot of the more boring submissions got highly upvoted
 
@WheatWizard Can we continue here?
There are more comments getting more damning on the popcon question
If you guys want to read other thoughts on it
And it's now +10/-8
But has 40 answers
@cairdcoinheringaahing I know they've been controversial, but I'm starting to wonder if we're not too obsessed with winning criteria
Does it actually matter that someone wins?
Do any of us participate to win?
Do any of us ever pay the accepted answer the slightest head?
 
i feel it shouldnt matter to be honest
 
Or care which have the most upvotes?
 
10:31 AM
@AviFS Having a winning criteria promotes improvement
 
The only argument I remember reading which I could understand was that if the winning criteria is objective, you have a metric for how good your answer is
 
(Just making my point in a sentence, long message incoming)
 
But do you really need to know? That seems a little... obsessive?
It's great to know how well you've done a lot of the time, but to never allow questions where you don't?
 
most authors dont even select an answer. i feel the "winning" criteria can be inferred from the type of challenge
 
winning criteria aren't about winning
they're about having something to optimize for
 
10:34 AM
And plus, you still never really know how well you've golfed. Until there are other answers in your lang, and the optimal algorithms are found, you can't know. And sometimes one never finds out how 'bad' one's answer was
Or how much better it could have been
 
@AviFS My "more creative" Jelly answer (convicted polynomial builtins) was fun to write, but when do I stop? I could keep adding random builtins that slowly converge towards 0.5, but the fact is that I'm trying to create a solution I think people will vote for. HN's rickroll is a good example of that - the entire code doesn't do anything more than my 1 byter (literally, everything other than the . is ignored), but it's "funnier", so gets more votes
 
@UnrelatedString Isn't it slightly obsessive, for a lack of a better word, to demand we always have something meticulously defined to optimize for?
I feel like that's just confirming the worst stereotypes of rigid programmers & STEM dorks
 
A little, but it's extremely confusing...
What do you try for then?
 
When I golf, I know that a 9 byte score is better than a 10 byte score. I also know that if there's already an 8 byte answer in Jelly, I'm unlikely to post my longer answer because it's worse. I know before I post roughly how good my answer is. With pop-cons, I have literally no idea until people start voting. Hell, I was fully expecting people to downvote my 1 byte Jelly answer because it is rather boring
 
@cairdcoinheringaahing But that's part of what it means to appeal to a human audience
That's what you're optimizing for
 
10:37 AM
if the problem isn't especially difficult there's simply no point in even posing it as a challenge if there's nothing to optimize for
 
Instead, it's at +9, because "Golfing languages short!!" (maybe? I don't even know why it's popular)
 
if you want to solve problems and that's that then you go somewhere for that
 
It's like comedians complaining that their job isn't well-defined
Or contestants on America's Got Talent
And programmers can do it, too
They don't begin start-ups only to say, I can't optimize for peeopplleee
 
Like, this was just a guess at what people would like.
 
@AviFS Honestly, a lot of people have problems with those shows, because they're so subjective
"This person is more talented, but this other person has a sadder backstory, so they win"
 
10:39 AM
True, that was a lame example
But what about stand-up comedians? Magicians? Start-ups & business people?
Artists? Musicians?
Authors?
Museum curators?
I'm not saying just because they have to, we should too. But it's certainly not unprecedented
 
I agree that sometimes you have to go with what you think people will like, but life is chaotic enough already in this case.
 
I agree too, in terms of keeping the norm as is. But I find odd the resistance to 'cherries on top' every now and then
Especially when it's so easy to sort by tag
In principle, we could even elect a separate moderator to be in charge of #popularity-contests
And more generally, the non-objective ones
My sense is most CGCC higher ups don't want anything to do with it
 
i don't think there's anyone arguing against the creation of code meant to be entertaining
it's just that it doesn't necessarily fit within the framework of a competitive programming site
 
So perhaps as a slightly different wing
 
We did dozens of these back in the day.
 
10:43 AM
or at least in the form of a popcon
 
They are boring. I don't see why we need more.
 
Everyone finds different things boring, though
 
You can still answer the old ones and it's not like this one provides anything at all new.
 
40 different answers
Those people found it worthwhile
 
Again. That's really not a good thing.
 
10:44 AM
None of which are very interesting.
 
Answers are not upvotes.
 
@WheatWizard But I'm saying clearly lots of people don't find it boring
 
people figure it's worth a shot because they happened to think of something to throw in the mix
 
@UnrelatedString I still don't see what competitive has to do with it
My sense is that most of us are not here to compete
And code-golf is our thing, there will always be plenty of places to compete
 
@AviFS Is this code golf?
 
10:46 AM
WDYM?
Which?
 
Sorry comment unrelated to image
The question isn't code-golf, it's just "Do this creatively"
 
I saw the comment you screenshotted already, though
I know, but I'm saying we have lots of code-golf, and it's doubtful it'll ever stop being that way
Certainly a few popcons won't get in the way
 
We already have lots of them.
I'm not sure even what you are proposing. Pop-cons are not banned, they are allowed.
 
This one is just very generic.
 
So, you can have creative answers to code golf challenges that aren't the shortest approach, and we explicitly allow longer approaches that make an effort to optimise their score. You can also have pop-cons that make an effort to be as objective as they can (see my last ). This recent one doesn't do that, but it also doesn't inspire many actually creative answers because the task is so simple
 
10:50 AM
i think pop-con does better when the creativity is pointed towards the output rather than the code itself
 
@WheatWizard Oh wait... good point, haha
I guess... just to hate them less? lol
We still have the consensus on objective winning criterion
I guess that's what I'm pushing back against
 
good popcons are good
it's just that there's a high bar for good popcons
 
But what of the obj win crit?
 
I have no problem with interesting pop-cons that try to make some effort to fit with the site rules of an OWC. I do have an issue with low-effort, uninteresting pop-cons that just say "Do X but creatively"
 
That makes a lot of sense, but again, why OWC?
I think that's really what my comments were getting at
 
10:53 AM
Why have any rules? Why don't we just have free-form do whatever?
 
30
A: Why do we have objective winning criteria?

Martin EnderDennis's answer touches upon this, but I wanted to write a separate answer to make it more explicit: Objective scoring is essential to drive competition, and competition inspires creativity. Here is how a challenge works that doesn't have objective scoring: you come up with a solution that is v...

 
I've read it several times but only that one argument has stuck with me. And I explained why I thought it was silly
But I'll go read it again
 
I like the idea that with an OWC, I know at all times how good my answer is. Otherwise, I have to guess what other people like and have to try to go towards that
 
@lyxal Can we have a flag for python-style lists? ([a,b,c])
 
There are plenty of places on the web for "Do cool coding stuff". I think having a well defined scope and limits helps us to maintain quality and build a community. Some people enjoy pop-cons even the bad ones. But some people enjoy knitting and that's not a reason to start hosting knitting challenges here. You can seek that out elsewhere. This is about the scope and aim of our community.
 
OWC helps us say "This is what we are about". That way we can build a community around common interests rather than trying to accommodate every possible desire.
 
It also means that I sometimes have to choose between an answer I find interesting, and an answer I think others will find interesting. For example, I don't find HN's rickroll particularly interesting. The only clever bit about it is a language quirk, and rickrolling is overplayed IMO. There aren't really any answers to that I find especially interesting, because they (mostly) have the same approach of dividing 1 by 2 (or 2 by 4, 3 by 6 etc)
 
@PyGamer0 normal python
@hyper-neutrino do you have an example in another esolang?
 
Even mine is just 1/2 with a lot of trigonometric obscurity.
 
Also, chat is a perfectly fine place for some interesting but non-OWC challenges. If you want to see some creative approaches to trivial tasks, just ask a CMC here and say that "Most creative wins". People like to goof off here
You might get a couple of "VTC as non OWC" replies but even those are jokes :P
 
11:03 AM
I really have to go now, buy I have to admit that main answer by Martin Ender is way better than I remembered
I do still think though that, just like the Sandbox, OWC should be highly encouraged, not required
 
Can you give feedback pls?
0
A: Sandbox for Proposed Challenges

mathCount my Pluses Your task is, to count how many pluses I have. What Pluses? The no plus: 0 Points - The naïve Plus: 1 Point + The double Plus: 2 Points + +++ + The mega double plus: 3 Points + +++ + + + + +++++++++ + + + + +++ + Pluses of higher order tha...

 
@Ausername E already does that: Try it Online!
But if you mean for output then sure, I'll add it to the todo list
 
@lyxal I meant for output, although E is fine as well.
It's just VY_str that prints lists, right?
 
String formatting a list converts it to normal form
@Ausername not really
Generators have their own printing function
 
So both of those need to be modified for this?
 
11:14 AM
Yes
 
Hi everyone
 
Well diddly darn howdy :p
 
???
 
I like adding random modifiers to my sentences sometimes
Hence the :p
 
o_O
 
11:20 AM
Thats a sign of GPT-3
You have used GPT-3 to generate those random words isn't it?
 
What do we want? Aeroplane noises! When do we want 'em? Neeeeow!
 
@Wasif no
@Ausername I thought that said apioform then
 
What is an apioform anyway?
 
what's GPT-3
 
@StackMeter artificial intelligence for generation of random meaningful text
 
11:28 AM
it's a generic predictive transformer, IIRC?
 
It generates the most likely continuation for a given piece of text. It turns out if you let it loose, it looks surprisingly like intelligence.
Just like has already happened before in the form of human brains
 
oh ok
 
There's GPT-2, which is free, but pretty bad at generating text, because there really isn't a ton of data here so, the real test seems to be looking at whether we can figure out which text is more interesting than others and how useful it might be for the audience. I tried that, and then I started looking at all of Facebook's social networks to figure out which was the most interesting
Yeah GPT-2 ********************* (2) "I want for sure I know how the last few minutes were." "Aye, what happened?" "What?" "I felt like I had something wrong."
It's very sus
 
Whoa 40 answers in half day (half challenge)
This community is full of friggin ninjas
 
11:35 AM
GPT3 has already been put into the position of a dungeon master. It's quite creative and mostly long term consistent, but also needs supervision because it can spontaneously go mad ... or too nsfw
 
'night.
 
@JohnDvorak where
@Ausername what timezone are you
it's 12:£6
 
AI dungeon
It's surprisingly good, but at the same time, not
> Ever wanted your D&D campaign derailed by, of all the possible causes, the dungeon master? Well then I've got something for you!
 
@Ausername o/
 
@Wasif it's 12:36, but o/ anyway
 
11:44 AM
12:36 in NZ? AM or PM?
 
in GMT
 
the mysterious third 12 hour period
only accessible by stepping through the magic portal into Narnia at the stroke of midnight
 
12:01 PM
Just at 299 answers...
1 more left to reach 300
 
12:15 PM
CMC?: Make a program in 51AC8. :p
 
i'm going to assume the empty program is valid and probably some variation of the identity function
so that, 0 bytes
 
I thought it was called Allyx or something.
 
@Adám i dont want an anagram of @lyxal
@UnrelatedString nice
 
@PyGamer0 repo link pls
I lost it
 
12:20 PM
0
A: "Hello, World!"

Pro OdermoniconRAMDISP, 18 bytes. [;[Hello, World!]] explanation: [; - print the following values, stringified, and joined. [Hello, World!] - an array containing ['H', 'e', 'l', ...] ]

 
12:32 PM
@PyGamer0 Charcoal, 4 bytes: ⭆⁵‽β
 
@PyGamer0 APLX: ⎕a[?5⍴26]
 
what's APLX
 
ngn
@PyGamer0 5?"a"
 
@ngn How does that work, and why do I always get the same result?
 
ngn
12:40 PM
@Adám it treats "a" as "abc..z". the prng is seeded with a fixed seed for reproducible results.
 
Is this special treatment of "a" unique to ??
 
ngn
@Neil ^ exotic character? :)
 
ngn
@Adám yes
 
@ngn Yikes, that's something of a special-case.
What does ? normally do on a vector?
 
ngn
12:43 PM
@Adám um.. yes. what's wrong with that?
@Adám n?vec is vec@n?#vec
 
So the reasoning is that if vec has just one element, then ?vec would be vec and therefore you can repurpose it?
 
ngn
@Adám ?vec is not the same as n?vec
 
No, I omitted the left argument for brevity.
Since n?vec is n#vec for when 1=#vec
Just seems awfully inconsistent.
 
ngn
@Adám what would be a consistent way to design it, in your opinion?
 
Let it be n#vec
Oh, forget everything I just said. Sorry.
 
ngn
12:49 PM
ok, forgetting.. :)
 
i think the idea is that if n#vec does the required thing then why bother having n?vec do the same thing
(anyway, saving that trick for later)
 
I got mixed up with BQN, where "a" is a vector. "a" is a scalar in K.
@Razetime But it does. n?scal is different.
 
hm
oh lol
 
@ngn Wat?
 
ngn
@Adám only "a" is special-cased
@Adám "A" probably ends up treated as a number
 
12:51 PM
@ngn No, "A" is even more special cased.
@ngn But how can it give a hexadecimal result?
 
ngn
oh, i see.. cyrillic :)
k is not unicode-aware, so a character beyond U+00ff is actually a string (list of characters)
@Adám 0x.. is a string
@Adám so, it ends up picking 5 times at random from the bytes that constitute cyrillic A
 
Ugh.
 
@ngn yet another funny trick that may work in golf
 
@ngn :-)
Why didn't you extend ! too?
 
ngn
@Adám because k6 didn't
 
1:01 PM
And it did extend ? like this?
 
ngn
@Adám yes
 
I wouldn't have expected that of AW.
 
ngn
@Adám why not? generating sample pseudorandom data for testing is a common requirement for big databases
@Adám as a programmer accustomed to 0indexing, i think ⍳'я' shouldn't include я. as a bulgarian, i think it shouldn't include ы and э :)
 
@ngn And generating A–Z labels isn't common?
 
@Adám ⍳'⍵' doesn't give Greek alphabet :(
 
1:07 PM
@Razetime is not a Greek letter.
 
ngn
@Adám not among arthur's followship :)
 
@ngn Of course not. a–z!
 
ngn
down with capitalist letters!
@Adám no hebrew?
 
No. Can't remember why I didn't add that.
 
CMQ: Favourite postfix operator/quick/adverb?
 
1:12 PM
@ngn “⭆” U+2B46 Rightwards Quadruple Arrow Unicode Character
 
@lyxal ⍨
 
.......... so my now tied 2nd most upvoted answer is just copy-pasting rickroll lyrics into jelly and having UB ignore literally the whole program except a single dot
._.
 
ngn
@Adám ⍳2j3 would be nice
 
@hyper-neutrino Directly taken from APL/J.
 
1:14 PM
unsurprising :p
i mean it's just apply polynomial
 
@ngn Wasn't so clear what it should do. works though.
 
@hyper-neutrino I
just I
dang it wrong link
 
but that's only b10
 
ngn
1:16 PM
@Adám 2x3 matrix from 0j0 to 1j2
 
@hyper-neutrino β or τ are for any base
but for some reason, they seem to be a bit broken
 
CMQ: How useful are trig functions in esolangs.
CMQ: What would be good chars (unicode / ascii) for sin, cos, tan
 
@PyGamer0 They're not used too often, but it'd be a good idea to add them at least as digraphs
 
@PyGamer0 I made them digraphs (two byte elements)
 
1:24 PM
ok like ts for trigonometry sin
or ms
 
@PyGamer0 relevant link
@PyGamer0 ∆s for sine
∆t for tangent
Etc
 
ngn
@lyxal that would make more sense if it was a right-angle triangle
 
@Adám uh, what?
 
ngn
@Razetime apl omega is a different codepoint in unicode from greek omega
 
@hyper-neutrino my highest is a joke answer on the xkcd koth
@PyGamer0 if it is a drawing lang they are indispensable
@DLosc is there a decrement operator in pip which doesn't modify the original
@ngn that moment when apl doesn't support apl omega
 
ngn
1:33 PM
@Razetime that's the least of apl's troubles..
 
oh no what are the troubles
 
ngn
@Razetime i don't wanna go throught them yet again - that would upset those vested in apl
 
@cairdcoinheringaahing maybe this helps?
 
CMM - Keep the current know how to vote ad, or format it into a site-thematic ad as per the second format?
this is an example of what it might look like (for crypto.SE)
 
ngn
@hyper-neutrino keep
 
1:53 PM
oh lol
@ngn where else do you chat?
 
ngn
@Razetime you mean which rooms? click on my avatar
 
no, like other than SE chat
 
ngn
what happens outside of SE chat stays outside of SE chat :)
 
should i rename myself to LitDaMeth
 
@ngn aight
 
2:03 PM
@hyper-neutrino Definitely keep.
 
@JNat If the title in the ad is different to the title of the question, is the ad title used, and is that ok?
 
@Razetime my highest is the rickroll
no seriously
 
2:38 PM
@hyper-neutrino Post both, see how the site thematic one looks, remove it if we dislike it
 
2:57 PM
oh, good idea
then I guess cmm, how should we word that?
 
That's assuming the ad title can be different to the question title (see my question to JNat above)
 
right
 
Maybe something like "Short code isn't everything. Find out how to vote"?
 
@cairdcoinheringaahing If there's a good reason to not use the question title, should be fine, yes
 
I like that wording
sidebar or banner? i'm thinking this would make sense as a banner ad but idrk
 
3:03 PM
@JNat Ah nice (I'm specifically thinking about our Welcome page, which I've already posted, which might be better as a "Want to get involved in code golf?" or similar if we show it on other sites)
 
@JNat Does that mean we can use a different title for ads that target other SE sites?
 
@Adám or for ads here too, yes.
 
alright, posted the site-themed equivalent ad here - hope i did it right :p
 
@JNat Allowed?
 
@Adám I'm still going over all the pings that accumulated over the weekend — I'll leave a note on anything that's not allowed in a timely fashion, don't worry about it. I'll be aiming to not leave said note the minute before the submissions close too.
 
3:12 PM
@JNat I know, Monday mornings… (I'm here for work too.)
 
4:03 PM
@JNat Is "all except self" an acceptable subset of sites for ads?
 
@cairdcoinheringaahing it is
in the sense that it is a subset of sites ;P
 
are you thinking of advertising CGCC to the network via a specific meta post?
I like that idea cuz it's better than pulling attention through HNQ exclusively
 
Actually, shouldn't the title be different depending on the target. I like the new formulation for all other sites, but somehting like "Welcome to the site – read this first" might be better for self-ads. — Adám 5 mins ago
If we have "2" welcome ads where one has a title for the network and one has a title for self
 
that makes sense
 
Also @JNat for the question ads, if we'd like them to be both banner and sidebar ads, should be post 2 answers, or can we say "right sidebar + banner" in one answer?
@JNat Ah good, don't have to list 170+ sites to advertise on :P
 
4:10 PM
@cairdcoinheringaahing one answer per ad, please
so, 2 answers for your example
 
Ok, just thought I'd check :)
Ugh, rate limiting. Got to wait a minute to post another one
 
The post says "Each answer must relate to a single ad submission," @Adám, so an individual submission for a sidebar and banner ad would be required. And the 2nd template guidance above should clear your question about where the ad should be displayed, no? — JNat ♦ Jun 17 at 17:12
^^ had already answered that too, @cairdcoinheringaahing ;)
 
@JNat Yeah, I wasn't sure if the flexibility of the question ads (being more human-designed/processed) would allow for more flexibility on that :)
 
separating them also allows for folks to vote on ads individually
 
Good point, didn't think of that
 
4:14 PM
I may wanna upvote a sidebar ad, but not a banner ad
or vice versa
 
would it make sense to selectively advertise to tech SEs or do we just want to have one for all - {self}
 
@hyper-neutrino I assume that's a question to the room, but: up to you!
 
(yeah, I was mostly asking caird / the room :P either should be technically supported/allowed, just wondering what'd make more sense)
 
@hyper-neutrino I've found that interest here isn't exclusively tech-centric. At worst, we don't attract any newbies from (e.g.) Mi Yodeya, at best we do
 
that's true. I guess advertising to more sites wouldn't exactly decrease our click count :P
 
4:16 PM
So I guess it can't hurt to show a couple of "Why not try?" ads everywhere
 
BTW @JNat do you get click stats on community ads (like how many times clicked and from which sites/locations) and if so do we get any access to those numbers?
@cairdcoinheringaahing yeah, agreed
 
@hyper-neutrino the MSE post lists the metrics I'll be able to provide at the end of the trial
 
alright, cool :D
 
4:40 PM
Heh, it looked like caird kicked another user out (they left just as caird joined)
 
@cairdcoinheringaahing Hey, I came here via HNQ on Mi Yodeya.
 
@Adám My point exactly :P Not all golfers come from tech related SE sites
 
@Adám But you were involved in APL (well) before that
 
@user Yeah, but we care about how people find this site and it's not always through SO
 
Oh, I see
 
4:43 PM
@user Yes, but never in formal code golfing.
 
"formal code golfing"?
 
It's when you code golf wearing a suit or ballgown
 
@user I had tried getting code as short as possible, but never in a organised competition.
 
@Razetime Not at the moment--you'll have to subtract 1 (or add v, if you haven't modified it). I'd certainly consider adding a "minus 1" operator if you think it would be useful. It would have to be a two-letter operator, so in general it wouldn't save bytes over just subtracting, but there might be some cases where it would help with parsing.
 
D is prescribed to modify but -- is not
i suppose changing -- would mess with backward compatibility
adding v sounds fine, no problem
 
4:53 PM
Ah, I see the answer you were working on.
2 can be #t.
(Or #k for that matter)
 
yes indeed
I keep forgetting that golflangs can get length of non-strings
#v lol
 
Ha, true.
 
5:08 PM
Alternately, it doesn't save bytes, but you can just increment again after decrementing: [oUo"miss a few"DhUh]Jk
 
i wonder if it's possible to make a symbolic pip
 
How do you mean?
 
> owo "miss a few" duh jk
 
@DLosc similar to this
 
@Razetime Ah, as in "is the language still usable when programs are restricted to \W*?"
 
5:12 PM
sort of?
more like any ascii which is nonalphanumeric
 
Oh, not \W because _ is allowed.
 
yeppo
 
Hm. I think not, just because all the variables in Pip are made of letters, and it's not tacit so it needs variables.
Well...
 
i suppose g can be changed to a symbol
 
Not all the variables, actually, there are the regex match variables like $$, $( etc
Hmm...
 
5:15 PM
hohoho
anyway good night, I had better sleep now
 
It would still be hard to do much, because all the functional programming operators and all the looping constructs use letters. Some things might be possible if input was stored in one of those variables automatically.
'Night!
 
I will say one thing about the recent pop-con: since it hit HNQ, we've had an increase in FP and LQ reviews :P
 
yeah, the past few KotHs have all done that too :p
 

« first day (3796 days earlier)      last day (1041 days later) »