The Nineteenth Byte - 2017-03-18 (page 1 of 3)

« first day (2240 days earlier) ← previous day next day → last day (2602 days later) »

12:14 AM

1

Q: Rearrangeable Code Bowling

Code-Bowling Challenge Write a program or function that produces the same output, no matter how you re-arrange the source code. (Output can't be an error) Removing any part of the source code however should cause that specific output to not be produced. Pangram Scoring Each character in yo...

restricted-source code-bowling pangram

12:25 AM

Challenge idea: programs are given a chess board, and they know which spots are filled by which color, but don't actually know the pieces. They can query up to half of the pieces. Given this info, score the board.

user image

@DJMcMayhem pretty cool app for android, "Vim Master"

it's little challenges to test your vim-ming skills

has anybody actually used the Lenguage interpreter?

does it allow UTF-8 characters?

12:59 AM

@Riker ok, I'll make sure to check it out

@Riker to be fair he already is vim master :P

@NathanMerrill It's binary so not sure what you mean by UTF-8 chars

yeah, I figured that out, thanks :)

@DJMcMayhem ya

LegionMammal978

Is it just me, or does the circle appear to be closing in on this image?

@NathanMerrill Are you trying to exploit my challenge :P

1:06 AM

@Riker can you give link I need to see if copyright so I can copy to iOS and get the thing that you use to purchase tin can :P

@LegionMammal978 this is super distracting

but yes

LegionMammal978

@quartata k, fixed

No no I didn't mean it in a bad way at all

I meant the illusion is weird

LegionMammal978

Oh, thought I would be suspended for disrupting the flow of conversation or something :P

@Qwerp-Derp Already exploited a different one :)

1:24 AM

Survey: Estimate the percentage of the code you've written is hosted on github.

@NathanMerrill 90%

I think I'm about 25%

@NathanMerrill 50% maybe

Yey I finished two pages of my Clojure tutorial!

I'm making a Clojure tutorial, not following through one

1:38 AM

@dim So the amount of gliders change the pattern?

I just posted a potentially controversial answer. Any opinions as to whether Deadfish~ is disallowed on PPCG (either due to not being powerful enough – it's not a programming language, but AFAICT that doesn't disallow it – or due to taking the program interactively and producing prompts as it does so)?

@NathanMerrill maybe 20-30%? I don't use Github myself, but I do a lot of open source work, and other people often mirror my work there

actually, re the controversial answer (why can't you reply to yourself?), it'd also be helpful to hear from people who think it's acceptable

@ais523 *you can reply to yourself

It's just harder than normal

@NathanMerrill 10~25

@MatthewRoh See?@ais523

works.

@ais523 Challenge is constant output so it's fine. Dennis posted a Deadfish~ answer once

Yeah, I agree too

ah right, I should have searched for Deadfish answers

it's capable of solving questions surprisingly often

Destructible Lemon

1:52 AM

yo!

although it tends not to do very well

True, Deadfish doesnt have input does it

not having input isn't really the issue

the issue is more not having useful data storage or control flow

Yeah, it only has 1 data storage space which doesnt even have good arithmetics

TIL making stuff in GoL is hard as heck

@MistahFiggins All strings are lists, (If something isn't, there's been a bug somewhere), Lists implicitly print like strings.

1:59 AM

I have a challenge idea

invert image and add '-1' in middle, call it 'Downify'

And make that a challenge

There's a Downgoatify challenge, which is essentially that but also make it Black and Red.

@Downgoat what's the font and size of '-1' in your pic?

wow, I may have stumbled across the worst question on PPCG?: codegolf.stackexchange.com/q/21391/62131

that said, I'm not sure it was correct to close it

although the victory condition is somewhat self-contradictory

Retina solves that.

Closing was good

2:04 AM

oh right, Retina prints 1 if given the empty program, doesn't it?

I'm not sure if this makes the challenge better or even worse

It's still a 'Do X without Y' type challenge

It's actually a pretty bad challange.

It's literally, "Find a program that does X"

@ATaco Oh yes that too

it's "do X, but if you do it in 0 bytes you get a huge bonus"

however the scoring is [tag;popularity-contest], so it's not equivalent to a code-golf with a linear function applied to the score

btw, How do I use generate TIO permalinks (not on TIO)

2:07 AM

Carefully.

@ais523 codegolf.stackexchange.com/…

@MatthewRoh Look at DataBot's source

> Winner will be decided on *1st April* 2014
April..Fools?

@ASCII-only DataBot doesn't work with the new TIO, does it?

@ASCII-only lowest-voted doesn't necessarily mean worst

although it's one of the lowest-voted anyway

@Dennis Oh true

2:09 AM

hmm, now that that Deadfish~ answer is in the polyglot question, I'm vaguely wondering if there are any languages which print a fixed integer given any program at all

2014 is like that, but it doesn't work any more, and the number's too high to plausibly reach

@Dennis Then what does it work with?

But v1 permalinks still do, so ¯\_(ツ)_/¯.

an abstract permalink system?

hmm, TIO permalinks look a lot like base64, but attempting to decode them like that implies that they aren't

in talk.tryitonline.net, Feb 17 at 20:20, by Dennis

@Dada It's fairly easy if you don't want to include settings. All fields (header, code, footer, input, and any non-negative amount of arguments) as UTF-8 encoded byte strings, joined with separator \xff, compressed with DEFLATE, then converted to Base64 with @ instead of + and no tailing =.

2:12 AM

oh, it's blatantly compressed

Doesn't really make sense for short permalinks, but does wonders for long ones.

I wanted to see the source which makes the permalink but Developer tools disagree with me and lag a lot

zlib-flate thinks the header is wrong

It's DEFLATE, not zlib. DEFLATE doesn't have a header.

Developer tool hates me ;_;

@Dennis could you post the script part as a gist here?

2:15 AM

hmm, zlib-flate is already pretty low-level, but maybe not low-level enough

TIO is on GitHub.

...oh ok

Encoding: github.com/TryItOnline/main-server/blob/master/tio.run/…

@MatthewRoh I dunno but the Downgoatifier might be relevant

But seriously, why do you not save the js externally?

2:18 AM

Why would I?

Why would you not?

Retrieving a single file is faster than retrieving multiple files.

Compression works better too.

JS externally would allow for caching and it generally better practice

Here we go with the caching again...

Caching a single file is still better than caching three separate files, unless you employ nasty hacks such as cache breakers.

copies function and goes to stack snippet interpreter challenge to cheat

:P

2:22 AM

The whole nexus.html (HTML, CSS, JavaScript, SVG) weighs only 12,360 bytes after compression. It's really not a big deal if you have to re-download it because of a change.

And changes to HTML, JavaScript, or CSS are rarely self-contained. Most of the time, if you update one, you also update the others.

It looks like theres a bit more to copy than what it seems

like: pluralization, languages list, etc...

You asked for permalinks. Are you trying to run code?

It's still not that big though

@Dennis nah

It just looks like some functions are linked to the permalink function

@Downgoat o_O wot

@ATaco what can you do to lists? Can you do math to each element individually?

2:31 AM

You can iterate through them with %, you can concatenate them with +, and you can explode them with e.

> explode them with e

-_-

I think, I'm going by memory here.

Do we need emojis in the chat userscript?

Yeah, that works.

like :grinning: and :unamused:

emojis are super hard to type

2:33 AM

Chat userscript?

userscript to make things like :grinning: become the grin emoji

prolly clientside (i.e. not an autocorrect, just a viewer)

but why?

@Riker Why not?

@Dennis Yes, someone made one and its nice

So the rest of us will see a literal :grinning:? grabs trout

4

Nah

2:35 AM

old-fashioned ASCII trigraph emoticons should be easily enough to express a useful range of opinions :-)

It edits the textarea when it sees one

ah

hands dennis trout anyway

(incidentally, I started using :-) as my main emoticon for expressing light-heartedness/happiness when I realised it was actually specified in an RFC)

o-O

link to rfc?

btw taco made it

2:36 AM

yah

@Riker tools.ietf.org/html/rfc1855

search for :-)

> - Use smileys to indicate tone of voice, but use them sparingly.
:-) is an example of a smiley (Look sideways). Don't assume
that the inclusion of a smiley will make the recipient happy
with what you say or wipe out an otherwise insulting comment.

@ais523 tha'ts actually a really nice guide

I think imma bookmark that to link to people

> Limit line length to fewer than 65 characters and end a line with a carriage return.

Sure.

nvm

@Dennis that's still a valid guideline for email IMO ("carriage return" probably doesn't mean \r)

2:38 AM

/tableflip:

(ノ°Д°）ノ︵ ┻━┻

the guidelines for talk are amusing, though, mostly because it's not a protocol that really exists nowadaysa

@Riker :D is better

:D is different from :-)

:D is "I'm happy" vs :-) meaning "don't take this seriously"

IMO

If anything, :-) and :) are equivalent.

ya

2:41 AM

I agree that D-based and )-based smileys are different

but you don't agree with my definition of them?

@Riker So :-) == Lenneh-faec

(-:D

@MatthewRoh .... no.

D is for outright jokes that don't even resemble serious conversation (either from you or from someone else)

2:41 AM

lenny = "this is suggestive"

..So, Kappa?

not really

idk

Like, uhh, sarcasm?

@MatthewRoh more of just a joke

IMHO, but obviously some people disagree

ok then

2:46 AM

Anyone know how to see chat replay on a past twitch broadcast (e.g. this) on the twitch mobile app (for android)? I know it's possible because if you click a live channel the chat shows up on the right but for past broadcasts a list of video suggestions is there.

not me

btw you watch WAN shows too? good

btw oh hi @Doorknob

→ 11 messages moved to Trashcan

@MatthewRoh hi.

hi

Anyone know a SE site where I could ask ^^^^^^^ ? :)

I thought there was some new Youtube/video based SE but maybe I'm thinking of video.stackexchange.com

idk, weird that many twitch questions are on SO

3:02 AM

@HelkaHomba help.twitch.tv/customer/portal/articles/2337148-chat-replay-faq

this says you can just open the website in desktop mode though

"Q: Will chat replay work on mobile? A: Chat replay functionality is not available on mobile at this time." But it is :I sometimes at least

But thanks :)

Or I'm wrong. Chat works on mobile. Chat replay does not. Ok

@Riker >_< I use both interchangably is that bad

Yes. The smiley police will be showing up at your front door any second now.

My userscript is not running even if I enable it. what's wrong?

Check the @match compared to your URL.

3:14 AM

oh wait

*://*.stackexchange.com/*

@ATaco Is something wrong?

Not that I can see.

@MatthewRoh Then there's an error probably

It doesn't say there's a syntax error

@MatthewRoh Need to escape slashes and periods and match more than one char? https?:\/\/.+\.stackexchange\.com\/?.*

If it's standard regex

Nah

3:22 AM

@HelkaHomba No, it's greasemonkey/tampermonkey, that's correct

@MatthewRoh Link?

I didn't upload to github yet

but heres a source code

wait, hastebinning

hastebin.com/isajisujet.js

@Downgoat the knob is correct

fear the smiley police

I have a maths question

There are two people, each with 100 balls, numbered 1 to 100.

Every turn, they both pick one ball at random. That ball is discarded.

What is the chance that on at least one turn, both people discard the same ball?

Anonymous

On the first turn, it's 1/(100)**2. On the second turn, it's 1/(99)**2. Do you see the pattern?

I tried changing an exact text to a jsfiddle embed

but I failed

3:38 AM

@Mego But you're implying that on the 100th turn, the chance is 1/1**2, which isn't true.

Oh wait

Anonymous

I had it right the first time :/ I failed to read "at least"

Don't worry

Actually no that works

It seems that I did a dumb thing called 'script in p'

Anonymous

Oh wait discarding is a problem

Anonymous

Because there's a chance that A discarded X before B drew X

3:38 AM

Hmmm that's true

With two balls, the chance is 1/2.

With three balls, I'm pretty sure the chance is 2/3.

I might have calculated that wrong though

Anonymous

There are 100! orderings for each player, so (100!)**2 pairs of orderings. The goal is to calculate how many of those pairs have equal values in paired positions.

The 3 ball case should be 1/3 + (2/3)(1/4) = 1/2? (although that doesn't seem quite right)

Anonymous

(100!)**2 is a bit large to compute

As what I know trying to embed jsfiddle in my profile messes up the site a lot

@Qwerp-Derp Derangements is the term you're looking for

the chance of an overlap is very nearly 1-1/e

Anonymous

3:42 AM

Oh yeah good catch

Anonymous

So it's the number of orderings that aren't derangements

Anonymous

Rather, the probability of an ordering not being a derangement

Wait what if we call the first ball that the first person removes 1, the second ball 2, and so on?

I'm not sure if that works

that does work

the game is the same if we assume the first person draws all their balls first, then the second person draws all their balls

so, after the first person draws, label them in order they drew them 1 through 100

the labels don't affect anything the second person does

Anonymous

The probability for n = 100 comes out to be approximately 0.36787944117144233

Learn How To Be Transparent

3:47 AM

Can anyone access meta now?

I am getting redirection loop error.

works fine for me

Anonymous

The probability for any n is sum(k in [0,n]){(-1)**k/k!}

So around 63%?

Anonymous

Yep

Anonymous

63% for no derangements, that is

3:50 AM

Hmm

I have atleast some progress on embedding jsfiddle in my bio

Come to my bio and run in console:

document.body.innerHTML= document.body.innerHTML.replace("<p>[f]","<iframe ");document.body.innerHTML= document.body.innerHTML.replace("[ef]</p>","></iframe>");

The thing is, the part under the bio is missing then

@Mego Now that I know the possibility, I might consider making a bet with someone about this :P

Anonymous

@MatthewRoh I'll pass on running arbitrary code you put in your profile, thanks

@Mego Okay, then. I'll post a picture.

Common sense would tell me that the chance is much lower than 63%, but probability tramples all over common sense

user image

^ this is my current progress

however, the thing under it is cut out

3:57 AM

@Qwerp-Derp my reaction was "I'm not sure if that's supposed to seem high or low, but it probably has something to do with e"

if you're drawing the balls with replacement, then it's just a simple compounding equation, which tend to converge to e-related numbers

and it's not obvious that drawing without replacement makes any difference; at least, I wouldn't expect it to make a large difference

Anonymous

@ais523 Yep, the probability of an ordering being a derangement is approximately 1/e.

Hmm, it seems to cut everything below the fiddle

Anonymous

It's cool how e, pi, i, primes, and factorials are all related

I am trying to change [f](data here)[ef] to <iframe (data here)></iframe> with an userscript, how can I?

Uhh wait, did I discover a bug in chat?

Or is it just the userscript

Does the '<iframe' part show like a tag to you too?

If then, it's a bug

Lemme see if its reproducible easily

ಠ_ಠ

It is

4:13 AM

@MatthewRoh what do you mean by that?

@Mendeleev Look: <this_looks_like_a_tag_but_it_really_isnt>

user image

user image

Huh.

Weird, is it just me?

might be my auto highlighter userscript

user image

4:16 AM

@MatthewRoh wat

I don't even know why this happens

Meh, it's just the prettify.

I think I got exposed to real savagery when I saw Drop the Mic, 3savage5me

Would you prefer to have a pong in GoL question

YAS

Now look in my bio

Quick

I embedded a fiddle in my bio

Now I just.. put it in a userscript..

4:47 AM

Does this look like a good userscript?

// ==UserScript==
// @name Embed Fiddle in bio
// @namespace http://tampermonkey.net/
// @version 0.1
// @description try to take over the world!
// @author You
// @match *://*stackexchange.com/*
// @match *://*stackoverflow.com/*
// @grant none
// ==/UserScript==
document.getElementsByClassName("bio")[0].innerHTML= document.getElementsByClassName("bio")[0].innerHTML.replace("<p>[f]","<iframe ");
document.getElementsByClassName("bio")[0].innerHTML= document.getElementsByClassName("bio")[0].innerHTML.replace("[ef]</p>","></iframe>");

@MatthewRoh lots of XSS porential

Well, right, but typing in the javascript console worked, why doesn't it work in userscript?

5:13 AM

Whew

gist.github.com/hjroh0315/8a4c9bcd248f1f741b4dc68a9fac37f4 here it is

apply it and go to my profile

Does it look good?

XSS for days :)

XD

Now, a diceroll.

tbh I think this is an innovation

But I think it would be way easier if there were stack snippets in profiles

5:44 AM

Python, I am disappointed. a, dict[a] = dict[a], b is NOT equivalent to tmp = dict[a]; dict[a] = b; tmp = dict[a].

Now I have to use the temporary variable. My day is ruined!

5:57 AM

@MatthewRoh yes, put more gliders in the loop to add more "pixels". You can extend the loop, too, but for both these modifications, you have to take care the gliders still arrive at the right time at the reflectors. Basically, new gliders need to be multiple of 15 cells away from the current ones and loop needs to be extended by the same amount. Because there needs to be 60 generation steps between gliders, and a glide moves one cell every 4 steps.

you can put gliders on the other side of the loop, too, but only at the right places. To check that, just advance a multiple of 60 gen steps, see where the gliders are at this point, and put them at generation 0 at the exact same place.

New Sandboxed Posts

6:37 AM

0

A: Sandbox for Proposed Challenges

Pratt certificates Your task: write code that generates a Pratt certificate for a prime number, and write code that verifies an existing Pratt certificate. What's a Pratt certificate? A Pratt certificate for a prime number p is a proof, of a particular type, that p is indeed prime. Historicall...

Anonymous

7:07 AM

@feersum Probably because it's trying to unpack the dict. Try a, dict[a] = (dict[a], b)

Anonymous

Actually I have no idea what you're trying to do

@Mego swap dict[a] and b without using a temp var i think

Anonymous

That's not what either of those snippets do

I'm trying to do the thing in the second code snippet.

Oh, I wrote the wrong thing in the second one.

It should be tmp = dict[a]; dict[a] = b; a = tmp.

Anonymous

I guess the tuple-for-compounded-assignment trick doesn't work with setitem

Anonymous

7:18 AM

It's because tmp = dict[a]; tmp = b isn't the same thing as dict[a] = b

But it is possible to mutate dicts with compound assignments.

d[1], d[2] = d[2], d[1] will swap the things assigned to keys 1 and 2 for example.

Wait doesn't dict[a], a = b, dict[a] work?

Maybe, I didn't try it with the first and second assignments flipped...

Because what you did made it assign to a first, but if you do it before a is assigned to it should work fine

Anonymous

The only reason I can think of for a, dict[a] = dict[a], b not working is that the a = dict[a] part gets evaluated before the dict[a] = b part

Anonymous

7:28 AM

Which appears to be the case

Then why does d[1], d[2] = d[2], d[1] work

Anonymous

Because of the order of the stores

Anonymous

My understanding is: With the a,d[a]=d[a],b version, a gets referenced after it is updated, so the update is done immediately. With d[a],a=b,d[a], a doesn't get referenced after it is updated, so the update is delayed. Same with d[1],d[2]=d[2],d[1] (but simpler because constants instead of variables).

Anonymous

This should prove enlightening

Anonymous

The first version essentially is compiled as d[a]=b=a, while the second one is compiled as tmp=d[a];d[a]=b;a=tmp.

Anonymous

7:38 AM

There's a reason for it, but I don't know what that reason is

7:49 AM

Guys, my approach to embed iframe in my bio turned out as a success!

(It's not efficient, but whatever)

8:21 AM

0

Q: Abusing the sandbox

Maybe "abusing" is the wrong word, but... like a lot of new users do, I come here and look around, if a new challenge is out, answer it as quickly as possible, only to find out that sometimes a challenge gets very high quality answers in a short time. We all know how it works: that challenge has...

8:42 AM

@WheatWizard Super late to the party, but I was surprised to find out that my rarest badge is the bronze code-challenge tag badge (16 total awarded). Might have something to do with the fact that Pip seems to do pretty well in source-layout challenges.

9:07 AM

-1

Q: number of integer between 10 to 20 and 50 to 80

i need a java program that will asked the user of the number of array elements. the program should be able to check the total number of integer there between 10 to 20 and the total number of integer between 50 to 80

code-golf code-challenge array-manipulation

Destructible Lemon

9:28 AM

For school, I am writing an essay, to turn into a video, on how Boston dynamic's Handle signals the end times for humanity

2 hours later…

11:00 AM

@DestructibleLemon ...what?

btw since when were you a lemon

and, will you ever turn into a lime

11:17 AM

Hello

11:56 AM

Oh hi

@Dennis What is the use of arena-server of TIO? Does it have any use currently?

Learn How To Be Transparent

@MatthewRoh transparent avatar?

WHAT?

@LearnHowToBeTransparent Gravatar

@LearnHowToBeTransparent Grav..darn, ninja'd

Learn How To Be Transparent

@MatthewRoh r u at work?

@LearnHowToBeTransparent ...no, I'm too young for a career >_>

Seriously did everyone forget I'm 13

Learn How To Be Transparent

12:02 PM

are you free now?

Yes

@MatthewRoh arena.tryitonline.net

oh

Learn How To Be Transparent

@MatthewRoh Can you use it khanh2003.github.io/Hoc462 on mobile and tell me if you can move forward?

I can

(probs because I'm using a bluetooth kb)

Learn How To Be Transparent

12:04 PM

@MatthewRoh what about dragging on screen?

Somewhat, but it's awkward because the page scrolls too

Learn How To Be Transparent

I need e.preventDefault() then.

@MatthewRoh Thanks for helping!

I will come back next hour.

@MatthewRoh :P jk the arena is the sandbox (well, at least internally)

@ASCII-only Then will it be used for future cases (like the competition I suggested)? (sorry for editing a ping)

@MatthewRoh what do you mean

@LearnHowToBeTransparent ok i'm on win10 tablet, imo the control scheme is a bit weird, i'd prefer an onscreen joystick control/maybe reversd vertical controls u.e. dragging down moves forward

12:11 PM

@ASCII-only Could the arena site be occupied for some reason in the future, like the annual competition I suggested in meta

@ASCII-only WHAT?

Hi all

Hi

how things?

Pro tip: delete your DNS records after destroying your droplet.

@MatthewRoh The arena servers are where user-supplied code is run. There are currently two of them: one used by Nexus/v2 and one for testing. They're swapped whenever I have to update and/or restart something in order to not interrupt the service.

Load-balancing will come soon-ish, and I plan to rent a handful of load-balanced arena servers.

12:23 PM

@Dennis May I post code to generate permalink in the Stack snippet interpreters challenge and mark it as non-competing?

(TIO permalinks)

btw, can I change my name to a single blankspace? If then, I will truly be transparent

@MatthewRoh That "challenge" asks for interpreters, not permalink generators.

Oh wait, right. OK then, ~~I'll post the interpreter part~~

@MatthewRoh Thankfully, no.

I know, I know. It was a joke.

(But it's always interesting to see the source if not copying without permission)

Can I actually set my name to a single blankspace? (or another character which functions like a blankspace, like an NBSP)

All usernames must start with a letter or a digit.

12:34 PM

Oh, okay then.

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Nearly-match a string Our near miss algorithm is too slow. The boss is blaming the Levenshtein distance builtin for being overkill for our purpose as we're only interested in a distance of 1. Please code something leaner. Input: Two different strings, in any reasonable format. At least 36 diffe...

« first day (2240 days earlier) ← previous day next day → last day (2602 days later) »

Transcript for

Mar '1718

The Nineteenth Byte

The Nineteenth Byte: General discussion for codegolf.stackexc...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile