How off center. Shame on you, space x.

I'm watching live on spacex.com/webcast

@AlexA. perhaps if we tried now. Back in the origin of the site, when golfscript was a topic of huge debate, I think the site might have disallowed them if it was possible

@QPaysTaxes space

so like 100 km

@QPaysTaxes first stage to deliver supplies to the ISS

@NathanMerrill Maybe. Though surely over time someone would have proposed they be allowed.

@QPaysTaxes yep, this is the real thing!

@NathanMerrill The problem is there are plenty of languages that are short (APL, J, TI-BASIC, etc.) that are not golfing languages.

not that far up - this is just the first stage

If we ban those then people are just going to find another set of short languages

@QPaysTaxes Well I can park in a car better than that, even after a 1 hour drive.

Until all we have left is assembly and C#

@Mego :28873107 that's the problem. Its impossible to define. I was saying if it was possible to define, we might have done it

@quartata you could always just ban language with the intent of being short

@Downgoat None of those languages were intended to be short.

oh,I thought you were talking about golfscript

No.

I honestly don't know why banning more or less arbitrary classes of languages comes up so often. Just do what you enjoy doing and let others do what they enjoy doing.

anyways, I do enjoy the golfing languages here, and I think they one of the biggest contributions we have as a site

@QPaysTaxes before that please share pictures

@QPaysTaxes Nope but to be fair: does this rocket take me grocery shopping?

@AlexA. I'm not sure myself.

I think people don't really understand them and assume they're basically cheating

but I think we might have lost a lot of "cleverness" in other languages because they simply aren't competitive against golfing languages

Other than that any car does have less C02 emissions.

I think esolangs are (not golfing languages). When I first came to this site golfing languages felt kinda like cheating.

@NathanMerrill I think the cleverness is still there; not everyone uses golfing languages.

@NathanMerrill Look at any one of xnor's beautiful Python golfs.

@DigitalTrauma Wait what that just happened?!

@AlexA. I'm not saying its gone, but that we are missing quite a bit of it

@quartata Oh god, yes. @xnor's Python golfing is a sight to behold.

@QPaysTaxes Which would be irresponsible any way, if you only even take the reaction distance in account.

@NathanMerrill I disagree. If anything I think solutions have become more clever.

Competition drives ingenuity

Looking back at challenges from 2011 a lot of the answers are kinda dumb tbh

@El'endiaStarman falcon rocket landed on a boat

@AlexA. only if they are competitive

@Downgoat I see that!

@El'endiaStarman A boat landed on a falcon in outer space

@QPaysTaxes And do you really want a car that needs refueling every 15 minutes?

@NathanMerrill What I mean is, to make your submission competitive against the tiny languages, you're going to come up with ingenious ways of shortening your code.

If anything, I think trying to compete against golfing languages has inspired more crazy golfing in regular languages

@QPaysTaxes oh :|

@AlexA. but that's once in a blue moon. People have already indicated that its the "comparable" languages that drive their competition

@AlexA. I actually think it's less of that and more that our site in general has attracted a lot of very clever people who just want to golf in whatever language they like

@QPaysTaxes thank you

as well as trying to outgolf themselves

Even if you're competing against yourself

Still applies

absolutely

I think that it was around the time of code trolling dying that challenges started to become harder and that's when we started to attract more of these kinds of users

@quartata Who are "these kinds of users"?

but when there are different "classes" of languages, its very hard to be competitive with languages in a different class, meaning you have less competition

@AlexA. that is not true. There is plenty of competition in the keyboard industry but none of them have been ingenious enough to make a keyboard that works for hooves

Like xnor.

Oh gotcha

I guess on other golfing sites you can't see the other solutions (like on Anarchy golf)

That seems weird to me

I really dislike anagol because of that.

Also as @TonHospel said I think our more flexible IO formats and rules are appealing

My favorite part of the site is seeing all of the different submissions and learning from them.

By the way, I wanted to show off this underrated answer:

since where is there jQuery for node

It exists, sadly.

npm install jquery

D:

should I make a jQuery interface for Cheddar?

Please no

@Downgoat :DD

@epicTCK ...

@QPaysTaxes don't worry, it's a joke; I would never do such a horrible thing

I was joking.... jquery is ugly

np

@Downgoat I was worried for a second

I can never tell when you people are joking when it comes down to JavaScript

no Safari

I know, Chrome developers weren't lazy at all

all of you KoTH people need to wait so I can get my KoTH runner live

If the cylinder wraps, arent the police and robber HQs right next to each other

gotcha

I just really want to picture this as a tall skinny cylinder lol

@AlexA. Thanks for the kind words.

@quartata The thing is, I don't think I would have joined the site had it been what it looks like now.

@xnor Thanks for the badass golfing.

@xnor What do you mean?

@AlexA. If I hadn't known about the site before and just seen it now, say from HNQ, I'd think it was mostly a site for people to make golfing languages and code in them.

I'd see solutions in Python but think they're just losing badly.

What was different about the site when you joined and what prompted you to join/stay?

@AlexA. Much higher density of golfing langs, more FGITW.

@AlexA. More questions that have heavily upvoted answers before I could possibly write a golf in Python.

Yeah but you say "left and right" in your spec which makes them different

@AlexA. At least, before I could write a good golf.

Anyways, it seems well specified, my only concern is that it might be a quasi-dupe of something somewhere

> goof golf

sorry i had to

haha

Just look at the existing life challenges before you post to make sure (I feel like I've seen something similar, but I'm not sure)

@QPaysTaxes I feel like I/others will have more feedback once you get an example up.

@xnor That was different a couple years ago?

That and/or just a visual example.

Yeah I don't think you need a visual example, it's simple enough as it is

Also TIL xnor and I joined around the same time

taiping 2 hard

@AlexA. As a matter of degree, I think. FGITW has always been an issue, as has people upvoting straightforward golf lang solutions for being short. But it would usually be with one golfscript sol rather than a bunch of different golfing langs, and with slower voting. I think it's not fundamentally different, just more immediately visible to new users.

@AlexA. huh, I didn't realize you were around as long

I wasn't very active for the first year or so

Part of the problem is of course we currently have quite a couple of golfing languages that are "tied" in golfiness. When it was just straight up Pyth > CJam > everything else, you'd usually see a Pyth solution and occasionally a CJam solution and not much else.

But it feels kinda like Jelly 05AB1E and MATL are all kinda competing for Pyth's throne, so usually you see a solution in each

@xnor I wonder if having a sort by random option for answers (in addition to sort by votes and active) would help at all with the FGITW effect. As for voting, the problem is that people just upvote whatever. Even if we had some kind of "guide to voting" I don't think that would change much.

@AlexA. yes, I agree that a guide to voting wouldn't help much

This might sound weird but what if you could only upvote a fixed amount of answers on a challenge?

Like maybe 3 or 4.

That might make the problem worse.

Yeah, I don't like that.

@quartata There is no way in the current SE software to enforce such a thing.

in an ideal world, people would only vote for the best, but people might just upvote the first

Hrmph.

I upvote all of the answers I think are well thought out or well golfed.

i used to do that, but i'm more sparing now

i wish there were more gradations for me to express liking an answer

short of giving a bounty

sometimes there's two good solutions, but one is better

Maybe every answer needs a way for people to say stuff like 2/10 very bore, 7/10 very golf, etc.

@xnor Maybe a sort of "voter's checkmark" that you could only give to one answer on a question

That actually might be useful outside of here, actually.

Oh by the way @ChrisJester-Young should I go for Chez Scheme or Racket?

@quartata that's a great idea!

i mean, not that anything will get implemented, but for brainstorming

@quartata Surely Chris will recommend Racket :P

@AlexA. Well...

@quartata I'm not really sure what you mean. Can you elaborate?

@AlexA. Think of it like a small +15 bounty you can give to only one answer to a question.

except it doesn't decrement your rep

@quartata For learning Scheme, I'd definitely go with Racket. It's more batteries-included and its package repository has many packages.

@ChrisJester-Young OK

To me, Chez's main value proposition is that its pricey version has an optimising compiler, which may be more performant than Racket (I have not tried, as I don't have the paid version of Chez).

I'm not trying to be mean to either side, BTW. I personally know (and have met in person) Racket team members, Chez team members, and Guile team members. :-)

uniquify = lambda l:(lambda s:[s.add(e) or e for e in l if e not in s])(set())

@ChrisJester-Young even though that's Python I'm sure the lisper in you must appreciate that :P

@ChrisJester-Young Didn't come across like that at all. You're just expressing your personal opinion

@quartata to give another point of view, for learning scheme I can suggest lispy: norvig.com/lispy.html

In imagining hypothetical new user xnor coming across the site now, I think an issue is that nothing really communicates that golfing here is (supposedly) a per-language competition.

there you learn to implement a scheme interpreter :)

@orlp (define (uniquify l) (for/fold ([s (set)]) ([e (in-list l)]) (set-add s e)))

@ChrisJester-Young don't think that's equivalent? where is the condition that makes the resulting list unique?

@orlp It's a set, which will not add the element if it's already in.

@ChrisJester-Young ah but does a set in scheme have order?

So my function actually returns a set, not a list.

because that's the crucial part about this

Oh, you want it ordered. Heh.

@xnor What if we developed a leaderboard that ranked languages in their own classes? I.e., Jelly vs Osable vs MATL, Pyth vs CJam, Python vs Ruby vs ..., etc?

@ChrisJester-Young yes, compare these two:

>>> uniquify([3, 1, 5, 1, 8, 2, 5, 9, 2, 3])
[3, 1, 5, 8, 2, 9]
>>> set([3, 1, 5, 1, 8, 2, 5, 9, 2, 3])
{1, 2, 3, 5, 8, 9}

@El'endiaStarman This would be really nice

there would be no point to uniquify if not for order-preservation :)

@El'endiaStarman actually, I forgot the leaderboard

@xnor Right now the leaderboard just compares every language to every other language though

It does have a "best solution in x language" thing though

yes, I think the right column does help

I haven't seen the leaderboards used much lately

They're automatic in the design userscript.

yeah, likely the user's first view doesn't have a leaderboard

We can take Alex's attempt and modify it to compare language answers only when they appear in the same question, to try and reduce the bias of "I'm only gonna answer questions I can do."

might be part of the reason why but I think they should still be included

@ChrisJester-Young if you pay close attention to uniquify you'll notice that the set is constructed to filter out duplicates, and is then discarded

@xnor Perhaps the right and left columns should be swapped to emphasize the "per-language" aspect.

@orlp Yep. Okay, lemme write a quick Racket version.

@AlexA. oh, nice idea

helpful cat hunts those pesky leaves

In other news I'm not sure why I'm watching this over and over again but I can't stop send help please:

(define (uniquify l)
  (define-values (r _)
    (for/fold ([r '()]
               [s (set)])
              ([e (in-list l)])
      (if (set-member? s e)
          (values r s)
          (values (cons e r) (set-add s e)))))
  (reverse r))

Wait, I can improve on that.

(define (uniquify l)
  (define-values (r _)
    (for/fold ([r '()]
               [s (set)])
              ([e (in-list l)]
               #:unless (set-member? s e))
      (values (cons e r) (set-add s e))))
  (reverse r))

There. Much more betterer.

my version did contain one cool trick

uniquify = lambda l: (lambda s:[s.add(e) or e for e in l if e not in s])(set())
uniquify = lambda l: (lambda s:...)(set())

(lambda s:...)(set()) is basically setting a private variable s to set()

@orlp Yes, that's called a let in Scheme.

@ChrisJester-Young yep, but in Python it's not used often at all :P

@orlp Such a shame. let is the best thing ever.

@ChrisJester-Young we have statements

@quartata wait upvoting decrements rep??? what??

s = set()

@orlp Statements are not let, since it does not constrain scope.

@epicTCK bounties decrement rep

@epicTCK No but bounties do

I ninja'd you all :P

wait, who is the one ninja-ing?

@NathanMerrill I knew bounties do but it seemed more similar to an upvote... wiht the +15 and all

@NathanMerrill The quicker one.

ok good

:)

@ChrisJester-Young scope is implicitly contained by def

or class

+15 to the acceptee and +2 to the accepter

@QPaysTaxes but 15 is closer to 10 than the average bountie...

or a bigger upvote...

whatever

what we really need to do is have people rank answers. Then we can run all of the answers through STV