The Nineteenth Byte

Bálint

20:00

I'm actually surprised these two functions work

Helka Homba

@EᴀsᴛᴇʀʟʏIʀᴋ Is that just a glorified Worldbuilders? :I

Upgoat

20

Q: How is this different from WorldBuilding?

Proposal: What if(hypotheticals) WorldBuilding deals with these type of questions, how will this site differentiate from WorldBuilding?

Eᴀsᴛᴇʀʟʏ Iʀᴋ

@HelkaHomba not really

10/10 the link got 2 stars but no new followers. >_>

Upgoat

@El'endiaStarman I have some questions about pytek: how are you handling I/O in pytek? (like will you use predefined variables like $@, input()?

@EᴀsᴛᴇʀʟʏIʀᴋ I commited

Eᴀsᴛᴇʀʟʏ Iʀᴋ

YOu didn't?

Helka Homba

20:07

@Leaky Better?

El'endia Starman

@Upgoat Input will be \input() and will use Pythons input() internally. Any more detail you'd like?

Leaky Nun

@HelkaHomba Better.

mınxomaτ

Hi, I forwarded this to the responsible person(s) in the What-If team. That's probably a bit faster, since the "official" mail address usually takes a few months to respond. If this is really urgent ping me in "The Nineteenth Byte" and I can message <some admin> of the What-If team directly on the server. — mınxomaτ 4 mins ago

Upgoat

@El'endiaStarman yes. How will you handle printing? Will the \print() function (or whatever it happens to be), go to a buffer which is managed separately? Or will that just execute python's print()

New Main Posts

1

Q: Computer Generated Cracked Soil

Write a program that takes in an integer from 0 to 65535 (216-1) and generates a unique 500×500 pixel image that looks as similar as possible to these 6 real life images of cracked soil: These are thumbnails, click them to see the full-size 500×500 images. The goal here is to make your co...

trichoplax

20:09

@HelkaHomba Wait, the original Worldbuilding isn't glorified...??

Cᴏɴᴏʀ O'Bʀɪᴇɴ

area51.stackexchange.com/proposals/88145/…

Upgoat

@EᴀsᴛᴇʀʟʏIʀᴋ ^

trichoplax

@EᴀsᴛᴇʀʟʏIʀᴋ I instantly starred it because I want more eyes on it, but I will have to read up extensively before considering committing

El'endia Starman

@Upgoat Pytek's \print will almost certainly be a wrapper for Python's print, in the sense that it will be more powerful and have more options. Specifically, the definition of \print will be in Pytek code.

Bálint

@HelkaHomba You asked about cracked soil 1-1.5 months ago, that was slow...

Upgoat

20:11

@El'endiaStarman what more options will Pytek have?

Eᴀsᴛᴇʀʟʏ Iʀᴋ

@Upgoat oh, didn't load.

@trichoplax okay

Helka Homba

@Bálint Did I?

El'endia Starman

@Upgoat Stuff like easier handling of weird printing layouts. For instance, The Ever Amplifying Zigzag should be much easier in Pytek than in Python.

Bálint

@HelkaHomba You said something about it

Eᴀsᴛᴇʀʟʏ Iʀᴋ

5

Q: What would happen to the internet if google disappeared?

There are a lot of great questions about What would happen if the internet failed semi-permanently and What would happen if electricity stopped working. As we all know, google has a great impact on the internet. I'm not only talking about the search-engine, Google offers a lot of other services ...

near-future internet

trichoplax

20:13

@mınxomaτ You know people from the What-If team?

El'endia Starman

@Bálint I thought that was me.

Upgoat

@El'endiaStarman ah, ok. I'm planing on having some sort of string-builder class to aid in ASCII art. Probably would just be a glorified printf though

mınxomaτ

@trichoplax I'm on the team.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

also:

31

Q: Would charging unicorns survive the impact with a line of infantry?

The unicorns would be used to charge a line of infantry with their horn. They are as big as a horse and the horn is similar to this image: The infantry is a standard medieval army made mostly of soldiers wearing longsword, wooden shield and chainmail. If I use my unicorns to charge the enemy...

reality-check warfare mythical-creatures biology

Bálint

@El'endiaStarman Oh, yeah it was, I'm mixing thing up :S

Eᴀsᴛᴇʀʟʏ Iʀᴋ

20:13

@mınxomaτ O_O

mınxomaτ

I don't have push access to the server though. But What If's are managed by many persons.

trichoplax

@mınxomaτ Wow. Also, there are responsible persons on the What-If team? :P

Eᴀsᴛᴇʀʟʏ Iʀᴋ

^

Upgoat

@mınxomaτ :O cool. pls shaer contents of ~/.ssh/id_rsa, its okay im not hackr, only fer educashinal porpoises

mınxomaτ

I wrote this e.g.: what-if.xkcd.com/1/de

quartata

20:15

@Upgoat 5.0.0 and 3.3.6

Upgoat

@mınxomaτ so you are author?

Helka Homba

@Malty popcon alert: codegolf.stackexchange.com/q/83092/26997

quartata

Relatively recent

@Upgoat Just a German translator I presume

mınxomaτ

@Upgoat I haven't done anything for over two years, though. The publishers hired someone to write new translations. What Ifs are written in Markup. The new guys send in their translations as a freaking Word document.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

@mınxomaτ in english also or not?

mınxomaτ

20:16

@quartata That's not dutch

That's German.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

^

quartata

Sorry, German. Saw .de and had a brain fart

Eᴀsᴛᴇʀʟʏ Iʀᴋ

deutsch though iirc

Upgoat

@HelkaHomba idk if this is too subjective, I'd vtc as too broad but you are #1 challenge writer so I guess you know what you're doing :P +1

whats stopping me from outputting a white 500x500px image

Doorknob

@Upgoat Vote on the post, not the user.

2

Eᴀsᴛᴇʀʟʏ Iʀᴋ

20:18

^

Upgoat

@Doorknob I know :/ I still vtc'd as too broad

trichoplax

@mınxomaτ There was me thinking it was just Randall going through a huge list of emails alone

Eᴀsᴛᴇʀʟʏ Iʀᴋ

@Upgoat a lopophole

Upgoat

@EᴀsᴛᴇʀʟʏIʀᴋ which one?

mınxomaτ

@EᴀsᴛᴇʀʟʏIʀᴋ Nah, just German. Then the publishers took over. Though it probably could use more english authors, since it's frequently on hold because lack of questions and/or Randall being busy.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

20:19

@Upgoat idk, go google

@mınxomaτ oh okay

El'endia Starman

@mınxomaτ Is Randall not the sole/primary author of all of the What-Ifs?

Upgoat

@HelkaHomba seems similar to this imo in the fact the "goodness of an answer is subjective of what the user thinks.

Helka Homba

@Upgoat It's fairly similar to codegolf.stackexchange.com/q/43274/26997 which was received well. But no worries :) vote how you feel

quartata

@Upgoat much less specified though

mınxomaτ

@El'endiaStarman Yeah, but not the technical side. What-If's are overlooked by 4+ people.

Helka Homba

20:21

@Upgoat Well, there is the rule that images must be "visually distinct", though I'd bet you'd mainly get downvotes anyway.

El'endia Starman

@HelkaHomba Should get mainly downvotes.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

^

TimmyD

@HelkaHomba Wasn't there a make a picture look like a wall patch or something? How is this different?

quartata

in Pytek, Apr 16 at 0:50, by quartata

I've been thinking about how Pytek will handle encoding some more.

regarding encoding, click to see whole conversation

mınxomaτ

TimmyD

20:22

@HelkaHomba Yeah ... >.> that one ...

El'endia Starman

@mınxomaτ *gasp* It's you.

quartata

So \input[FIXED_8]() reads in a code point 8 bits at a time

Helka Homba

@TimmyD Yeah, I just linked that. It could be considered a dupe but I think that's a stretch, and people will still have fun with this one

Upgoat

@HelkaHomba so the input is essentially a seed to generate the random image off of?

Helka Homba

@Upgoat yes

Eᴀsᴛᴇʀʟʏ Iʀᴋ

20:24

@HelkaHomba update the accepted answer?

30

Q: Photomosaics or: How Many Programmers Does it Take to Replace a Light Bulb?

I've compiled a mosaic of 2025 headshots from the avatars of the top Stack Overflow users. (Click the image to view it full-size.) Your task is to write an algorithm that will create an accurate photomosaic of another image using the 48×48 pixel avatars from this 45×45 grid of them. Test Imag...

Fatalize

@HelkaHomba What prevents me from storing 65k real images of cracked soil and return those

Eᴀsᴛᴇʀʟʏ Iʀᴋ

the popcon part

Bálint

2

A: Convert UTF-8 string with only 8 bits per character

You can get the ASCII value of a character with .charCodeAt(position). You can split a character into multiple characters using this. First, get the char code for every character, by looping trough the string. Create a temporary empty string, and while the char code is higher than 255 of the cur...

TimmyD

@HelkaHomba I'm struggling with how objective the "it must fit in with these six" is. That seems very subjective, and was the reason e.g, Mona Lisa got bombed.

quartata

I honestly think the only language that handles encoding to my satisfaction is Perl (sorry)

I really, really hate the way Python does it. (one of the few things I dislike about Python)

Bálint

20:25

The OP said, that my function is slower than the currently accepted one

quartata

Reading binary data from STDIN is literally impossible

mınxomaτ

I hate the way Python does everything.

Bálint

Is it?

quartata

@Bálint Strings aren't based on code points but rather based on the characters. So basically everything has to have an encoding

Normally you could bypass that bullshit when reading from a file by opening it with rb but you don't have that option with STDIN

And of course the ASCII "codec" doesn't support anything >127 so you have to use CP437 or something to read it

(Latin-1 doesn't work for all values)

Helka Homba

@Fatalize Good faith too. People here are less conniving than one might fear.

@EᴀsᴛᴇʀʟʏIʀᴋ Nah. I can't constantly be updating the winners of old pop-cons. The accepted answer was the highest voted when it was accepted.

Fatalize

20:30

The best approach I can think of uses machine learning but I don't want to waste my time building a dataset of cracked soil images

trichoplax

@Fatalize Just model particle suspensions and evaporation :P

Fatalize

@trichoplax Building a model that yields realistic images is probably even more time consuming :p

trichoplax

@Fatalize I know - the one thing that would probably take longer than building your real world dataset :)

Bálint

Are Ã and ¶ ASCII characters?

Cᴏɴᴏʀ O'Bʀɪᴇɴ

no

Lynn

20:33

Nope

Cᴏɴᴏʀ O'Bʀɪᴇɴ

extended ascii, yes

Adnan

Or CP-1252

Helka Homba

@Fatalize Base something off of this math.marywood.edu/~tfkent/research/fractal_soil

quartata

Extended ASCII is a misleading term

Bálint

extended

quartata

20:33

It can refer to any number of 8-bit encodings

Cᴏɴᴏʀ O'Bʀɪᴇɴ

shh

trichoplax

@Bálint ¶ is sometimes used to represent an unprintable ASCII character, but isn't itself one

Fatalize

@HelkaHomba The hard part in making it realistic isn't the "cracks" part

it's the texture + lighting + shadows + small details part

trichoplax

@quartata Was about to make a joke about the technical upper limit then realised there really isn't one...

Cᴏɴᴏʀ O'Bʀɪᴇɴ

Couldn't I just make a code-golf challenge a pop-con? It's just changing the winning criterion.

Helka Homba

20:35

@Fatalize Well sure, I'm not really expecting perfect photorealism.

quartata

@trichoplax graham's number-bit encodings

trichoplax

I see it as the contestants competing against each other rather than against the photos, with the photos used to judge between them

quartata

fuck Unicode we're gonna have all the characters

trichoplax

@quartata We found a problem where Graham's number is too small

Helka Homba

@CᴏɴᴏʀO'Bʀɪᴇɴ Technically yes. But I wouldn't do it. It opens up a bag of worms on this site

Cᴏɴᴏʀ O'Bʀɪᴇɴ

20:37

@HelkaHomba Yes, true, just hypothetical.

quartata

@trichoplax haha

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@trichoplax we did? o_O

trichoplax

@CᴏɴᴏʀO'Bʀɪᴇɴ There are more than Graham's number of possible encodings for extended ASCII

Cᴏɴᴏʀ O'Bʀɪᴇɴ

Oh. right

trichoplax

(only because it's infinite - not any interesting large number...)

miles

20:38

@Dennis oh you can see it? I read chat later and realized maybe Zgarb was supposed to answer so I didn't want to steal the glory, but then the question was edited and lots of other stuff

Eᴀsᴛᴇʀʟʏ Iʀᴋ

@HelkaHomba okay

Upgoat

can anyone tell if my solution works: jsfiddle.net/evnfh4cw, my computer turns into a star when I tri to run it

2

Dennis

@miles Yes, users with 2k+ rep or diamonds can see deleted posts. Commenting on deleted posts is mod-onñy though. Answers are first come, first serve, so I don't think there's a reason for deletion.

quartata

@Upgoat Holy fuck, your computer turns into a star? You've made a fusion reactor with nothing but a little JavaScript?

Better patent that

Upgoat

@quartata yes, javascript is best language, i get free fushion reator

Cᴏɴᴏʀ O'Bʀɪᴇɴ

20:43

@Upgoat nothing is happening

Adnan

@Upgoat This crashes my graphics accelerator

mınxomaτ

This is so satisfying: codepen.io/soulwire/pen/bKens

3

Adnan

wtf

Upgoat

@CᴏɴᴏʀO'Bʀɪᴇɴ it supposed to go over every pixel in the 500px by 500px canvas and so some caluclations and place the corerct color dot. so 250,000 * javascripts slowness = boom

so just wait

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@Upgoat it stopped running stuff.

Upgoat

20:45

...

@CᴏɴᴏʀO'Bʀɪᴇɴ do you know if it's possible to declare a property of a class that doens't get inherited

Cᴏɴᴏʀ O'Bʀɪᴇɴ

?

Upgoat

ಠ_ಠ I looked up "make non-inherited property" and I got this: trulia.com/voices/Home_Selling/…

Bálint

@Upgoat Google and it's wonders

Upgoat

@CᴏɴᴏʀO'Bʀɪᴇɴ like I want a static property on class A. so I can do A.foo but if I do class B extends A I don't want it to get foo

El'endia Starman

@mınxomaτ Way too fun to play with.

Bálint

20:47

BTW, how do you spell it properly? it's or its

El'endia Starman

*click click click click...*

Cᴏɴᴏʀ O'Bʀɪᴇɴ

its in that scenario

Upgoat

@Bálint it's = it is; its = it possessive. e.g. It's cat would not make sense because you wouldn't say it is cat

mınxomaτ

@El'endiaStarman Imagine a touchscreen :D

El'endia Starman

*poke poke poke poke...*

Bálint

20:49

@mınxomaτ It doesn't really work on a touchscreen, as it goes darker for a sec

Helka Homba

So many pop-con haters :/ Don't you close-vote guys want to see the answers? This site is about entertainment after all, not meticulous enforcement of rules that are impossible to make 100% absolute.

mınxomaτ

Imagine a screen that can alter how its surface feels depending on the content. That would be so cool.

Bálint

@mınxomaτ You either need neural connection or a very expensive screen for it

mınxomaτ

Probably possible with nanotech magic.

Bálint

Depends on your needs

@mınxomaτ Yeah, that goes in the very expensive category

Upgoat

20:50

@HelkaHomba I don't hate pop-cons, after all most of my rep came from pop cons. But that pop-con seems to have little spec for a valid answer, really any 500x500 image with a unique output is valid if i understand correctly

Rainbolt

@HelkaHomba It is rules that are impossible to make absolute that I spend the most time stressing over every day of my life

flawr

@HelkaHomba I don't think you have to change much, just declare an objective validity criterion, and perhaps plug it into the sandbox.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@Upgoat idk

Helka Homba

@flawr For a "objective validity criterion" would something like: (1) the image must have the fractal-like cracks, (2) the image must be gray-tinged like the samples, (3) the image must have specs and dots similar to the samples - be good enough? IMO this kind of stuff is a clear goal from saying that images should be as realistic as poissible.

flawr

@HelkaHomba "fractal-like-cracks" and "similar" is not very objective.

Upgoat

20:52

@HelkaHomba can you specify those? Blank image is white fractal cracks on white background

> gray-tinged

so where RGB channels are equal? then your explanes would be invalid

> the image must have specs and dots similar to the samples

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@flawr Are you looking for a pop-con, or a code-golf? He described objective criteria, not a formula.

Upgoat

idk what that really means

flawr

@CᴏɴᴏʀO'Bʀɪᴇɴ Please do not let us start all over again, we've had tons of discussiosn.

Fatalize

To be honest I prefer a pop-con with relative freedom and a decently interesting goal like the one here, instead of a meme pop-con like the upgoat or downgoat one

mınxomaτ

@Bálint MIT devs material that changes colour and texture

Cᴏɴᴏʀ O'Bʀɪᴇɴ

20:54

@flawr and I guess discussion is bad, okay

flawr

@HelkaHomba You could e.g. say that everything in the new images must be sampled from those given tiles (even if it is just pixel wise). This would allready suffice as objective validity criterion. You do not have to pack everything in your challenge into it, and as I said you do not have to change much. Consider it as some paperwork you have to do such that your bosses are pleased=)

mınxomaτ

> “We harnessed a physical phenomenon that we discovered in 2011, that applying voltage can dynamically change surface textures of elastomers,” Zhao says.

Bálint

@mınxomaτ .

Maltysen

@HelkaHomba closed before I got here...

:P

Bálint

4 mins ago, by Bálint

@mınxomaτ Yeah, that goes in the very expensive category

El'endia Starman

20:56

@Bálint I know this is a thing to do on Imgur, but does it work here?

mınxomaτ

Military-founded. Money is negligible :P. It's the "because we can" category.

flawr

@CᴏɴᴏʀO'Bʀɪᴇɴ Not at all, but we've just had that discussion and we've come to a consensus, so I'd like to see that as settled for a while now.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

I wasn't here for it :/ do you have a link?

Rainbolt

I was just about to ask for that. I looked and didn't find a good match.

Bálint

@El'endiaStarman I added the dot, to make it an actual reply, not just a number

TimmyD

20:57

I've voted to close this challenge as Too Broad because it lacks objective validity criteria. — TimmyD 21 mins ago

flawr

@CᴏɴᴏʀO'Bʀɪᴇɴ Sure, just a moment.

TimmyD

Three of 'em in there.

Poke

@Fatalize As a lurker on this site for a few years I thought the downgoat challenge was fantastic before I knew it was a meme. It is one of my favorites along with codegolf.stackexchange.com/questions/33172/…

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@TimmyD I thought that was only one link, you trickster :P

Rainbolt

I found a post by Martin "How should we close questions without an objective validity criterion?" but no post like "Should we close [...]"

TimmyD

20:57

:D

Fatalize

@Poke This challenge is trivial with any non-godawful state-of-the-art classifier

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@Fatalize You're trivial is evidently different from mine. I don't think I could program that for the life of me.

Poke

@Fatalize But as @HelkaHomba pointed out, the wording of the challenge is pretty much the same as the other one linked in the comments

Fatalize

@CᴏɴᴏʀO'Bʀɪᴇɴ You don't even have too, there are billions of machine learning libraries

Helka Homba

@Upgoat It it of course unclear, but one can't definitively describe in words what makes one image "similar" to another. But I'm confident anyone viewing the challenge could easily determine what answers do look the most similar to the 6 samples. A random 500x500 image is valid perhaps, but so what? That kind of post will just be downvoted, and few will attempt such a silly answer.

mbomb007

21:00

What happened to @squeemishossifrage ?

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@Fatalize then it's trivial trivially. I'm sure libraries can make most tasks on this site trivial.

mınxomaτ

@HelkaHomba Well, PSNR is a good measurement. But there are better algorithms, like butteraugli.

trichoplax

"Must have an objective validity criterion" is not objectively defined. "Must be 500x500 pixels" is an objective validity criterion, but whether it's enough specification is subjective. I think this is a problematic way of deciding if something is on topic

Upgoat

@HelkaHomba popularity contests are to give some creative freedom to answers, not to serve as a substitute for a solid spec. I think perhaps the challenge would of benefited from some sandboxing

flawr

@CᴏɴᴏʀO'Bʀɪᴇɴ Just a collection of posts: one two three

Poke

21:01

@HelkaHomba i actually didn't realize you created the codegolf.stackexchange.com/questions/33172/… challenge until i looked at it just now.

flawr

four five

Helka Homba

@flawr Sampled from the given tiles? Not sure what you mean.

@Upgoat Do you honestly think you won't be able to tell the good answers apart from the bad ones?

flawr

@HelkaHomba E.g. only the colour pallete of the given images can be used, but none of the tiles may be duplicated as output.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

we should compile the pop con tips

@Upgoat I updated that NTFJ answer with an explanation

mınxomaτ

Fortunately, the charging one has been solved now that we've all standardized on mini-USB. Or is it micro-USB? Shit.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

21:03

@mınxomaτ funny, but this isn't standards

Helka Homba

@flawr But that's an unnecessary (and bothersome) restriction. A color not represented but very close could very well be in a similar image.

mınxomaτ

@CᴏɴᴏʀO'Bʀɪᴇɴ If it's a reference for challenge creation that people should follow, it's a standard.

flawr

@HelkaHomba What if you just reduce it to grayscale images instead?

Cᴏɴᴏʀ O'Bʀɪᴇɴ

*competing standards

Upgoat

@HelkaHomba sure I can but again, there should still be a solid spec. We are all about finding loopholes, don't expect none will be found without a solid spec

Helka Homba

21:05

@flawr I'd be open to that if there is lots of support for it, though by now it's perhaps a bit too late.

@Upgoat What loophole could someone exploit that would allow them to win the challenge?

Like I said, I highly doubt a "joke" random image answer could possibly win.

Upgoat

@HelkaHomba we'll see :/

@HelkaHomba I just think it needs a spec on what's "cracked soil". And how one answer is better than another

quartata

@Upgoat You can't possibly spec that

Upgoat

btw, what's stopping me from hardcoding a couple images of soil from google and just modifying a couple pixels

quartata

Nothing, nor should there be

Maltysen

@Upgoat you would just get downvoted to shit

Adnan

21:08

^

Upgoat

@Maltysen My profile picture has already been downgoated, but my name hasn't

Helka Homba

@Upgoat I can put in a "how to vote" section if that helps. Basically - vote for the answers that make similar images - it's already implicit.

@Upgoat From spec: "...your 65536 (2^16) possible output images should be visually distinct from each other and the sample images, especially with regard to the arrangement of the cracks. You must truly generate your images, don't just rotate and translate a selection from a preexisting photo."

Upgoat

@HelkaHomba how are you distinguishing between "photorealistic" and a "preexisting photo"

Maltysen

@HelkaHomba that seems like a validity criterion to me, even if it isn't completely objective, IMHO its should be more than enough.

Upgoat

I could compress a real photo of soil and write a program to generate a program to generate that image

flawr

21:11

@HelkaHomba Well then just add some of the really objective criterions you already have there and give them the title objective validity criterion, you really do not need to change much.

Upgoat

There is no object-winning criteria. pop-con is not replacement for object-winning critera as we've say to new users

Maltysen

@Upgoat yes it is

the winning-criterion is most votes

Upgoat

@Maltysen "Make a program that does something", winning critera: most votes

that's not a good challenge (yes it is exaggerated example)

Helka Homba

@Maltysen thank you :)

Maltysen

@Upgoat well, yes, that's why we have validity-criterion

(and too-broad close-votes)

Upgoat

21:14

@Maltysen if pop-con is excuse for subjective winning critera then why was this question closed, for example:

7

Q: Make the pattern from composition notebooks

This challenge was xnor's idea, taken from Digital Trauma's challenge donation thread. Your challenge is to write a program that, as xnor put it, "procedurally generate[s] the splotchy black-and-white marbled pattern on the covers of composition notebooks": Your program can either display i...

popularity-contest graphical-output

flawr

@Upgoat Please note that an "objective validity criterion" is something else than an "objective winning criterion".

mınxomaτ

@Upgoat That's not what he's saying.

Helka Homba

@Upgoat "You must truly generate your images, don't just rotate and translate a selection from a preexisting photo."

El'endia Starman

@HelkaHomba: Other resolutions are available now on your latest Minecraft video.

Upgoat

4 mins ago, by Upgoat

I could compress a real photo of soil and write a program to generate a program to generate that image

Helka Homba

21:16

@El'endiaStarman :D Thanks for telling me

Rainbolt

Someone should edit the pop con tag: "Challenges must have somewhat objective validity criteria."

Cᴏɴᴏʀ O'Bʀɪᴇɴ

ppcg history is being made

Upgoat

@mınxomaτ ?

Cᴏɴᴏʀ O'Bʀɪᴇɴ

maybe

flawr

@Rainbolt somewhat is somewhat contradictory to objective

Rainbolt

21:17

@flawr Yes that's the joke...

flawr

also, we had a lot of discussions and just recently changed the rules=)

Upgoat

@HelkaHomba ¯\_(ツ)_/¯ it's using pop-con instead of an objectiveness. I think it's off topic, but you're free to disagree with me

Helka Homba

@Upgoat You'd hopefully be downvoted for that. The number of required outputs (2^16) should make that kind of solution completely unwieldy anyway

flawr

@HelkaHomba Can you just please add some of your objective rules to a new paragraph titled objective validity criterion? I think that would make a lot of people happier.

Upgoat

@HelkaHomba again, I can always just slightly modify the image, either by slightly modifying some pixels, or by moving some blocks of pixels

^^

flawr

21:20

@Upgoat You're free to do so, I just don't think you're going to get many votes

Rainbolt

This discussion is so interesting. "I could do this." "You'd be downvoted for this." "I could do that." "You'd be downvoted for that."

Upgoat

@flawr Why do we have any rules at all? You cheat or abuse loopholes, it's okay, but you might get downvoted

Rainbolt

One of you has to care enough to make a meta post about it

Bald Bantha

@flawr I'm kentuckian and I (don't actually) find this offensive

quartata

@Upgoat These aren't "loopholes" in the sense that they're not standard loopholes

Also downvotes actually matter here

feersum

21:23

If you get downvoted in a popcon, you get downvoted in real life.

6

Upgoat

@quartata If we start to allow any challenge to be on-topic under pop-con our site quality would greatly decrease, there's a reason code trolling was killed off

flawr

@Upgoat that's why we have rules

@BaldBantha =P

Maltysen

10 mins ago, by Maltysen

(and too-broad close-votes)

Upgoat

2 mins ago, by Upgoat

@flawr Why do we have any rules at all? You cheat or abuse loopholes, it's okay, but you might get downvoted

we don't need rules when we have downvotes

flawr

@Upgoat The rules we are talking about are about the challenges, not the submissions

quartata

21:25

How about we make it a code-challenge where the score is the difference of all possible images to the test set calculated by this (lowest score wins) and the absolute difference of the actual RGB values is above a certain threshold

flawr

@BaldBantha So how many you's are there per gun?

quartata

Since apparently no one can agree whether or not pop-cons are OK

mınxomaτ

I don't get the fuzz. A "post sh*t - get hit" policy is completely fine for me. You can't replace thinking with rules.

Bald Bantha

@flawr 0 im 14

flawr

@BaldBantha Is there an age restriction?

Upgoat

21:26

@quartata no one said anything about pop-cons being evil. There is just no critera for valid answer from bad (invalid answer, but not invalid because there's no spec). What's to say my answers better from someone elses besides votes?

@flawr i dont think so. would infridge on right to bear arms

quartata

@Upgoat Your argument is literally "because it's a pop-con it's wrong"

Of course this is what you are saying

Upgoat

@quartata how?

quartata

9 mins ago, by Upgoat

@HelkaHomba ¯\_(ツ)_/¯ it's using pop-con instead of an objectiveness. I think it's off topic, but you're free to disagree with me

Maltysen

@Upgoat because the definition of a pop-con is most votes win

Bald Bantha

@flawr IDK my parents probably wouldn't let me have one so I don't really want one RN... I will when I'm an adult

quartata

21:27

13 mins ago, by Upgoat

@Maltysen if pop-con is excuse for subjective winning critera then why was this question closed, for example:

You get the idea

Upgoat

@quartata I'm saying pop-con shouldn't be an excuse for lacking an objective-winning critera

Maltysen

@BaldBantha one, or multiple? :P

Upgoat

that's our one rule for challenges

flawr

@Upgoat Yes there are no rules for answers, why don't you open a meta post about it?

quartata

Anyways, I think making it a code challenge will make it less fun but less controversial

@Upgoat he has both an objective winning criteria and an objective validity criteria

Bald Bantha

21:28

FUN FACT: Since when did KY have a mass shooting with all these guns around? guns don't kill people

quartata

The objective validity criteria is just better measured by a human

Upgoat

these are the two rules that all challenges should have:

Bald Bantha

@Maltysen One then I'll like it so much i'll buy out the store :P

quartata

@BaldBantha guns don't kill people people kill guns

Upgoat

Bald Bantha

21:28

@quartata Nigahiga?

Helka Homba

I've added a few points to the rules list. But I'm done debating for now

Upgoat

it completely lacks #1 imo

quartata

@BaldBantha Bloom County

Rainbolt

@Upgoat Most votes is an objective winning criterion, is it not?

quartata

Whoever this "Nigahiga" is stole it from the 1980s

Upgoat

21:29

@Rainbolt Yeah, that's #2

quartata

@Upgoat It has both

Bald Bantha

How To Run For President!

►

Quill

Morning

Bald Bantha

evening

Rainbolt

@Upgoat Right, the rest of the room is talking about #1 while you keep bringing up #2 (intentionally or not I don't know)

Bald Bantha

21:30

afternoon

Upgoat

@Quill hello

quartata

1 min ago, by quartata

The objective validity criteria is just better measured by a human

It could be measured by a computer:

Quill

Hey @Upgoat, I'm thinking of doing a unit tests blog post... Thoughts?

quartata

5 mins ago, by quartata

How about we make it a code-challenge where the score is the difference of all possible images to the test set calculated by this (lowest score wins) and the absolute difference of the actual RGB values is above a certain threshold

flawr

@HelkaHomba Can't you just please add the title OVC before some of your rules?

2

Upgoat

21:31

@Quill oh cool. Are you going to go over like design of unit tests or implementation/code?

@flawr congrats!

El'endia Starman

@flawr Ding!

orlp

hi chat

what's going on

flawr

OVER 9000

@orlp STUFF IS GOING ON

Bald Bantha

4 Non Blondes - What's Up

►

Upgoat

@Rainbolt Only thing it does to attempt to fulfill #1 is specify "make your computer generated soil as photorealistic as possible" along with a large set of rules attempting to ban all the loopholes discussed hear due to lacking a spec. 10/10 very specific and objective

Maltysen

21:32

> You may base your algorithm on information gathered from the 6 sample images, but your 65536 (216) possible output images should be visually distinct from each other and the sample images, especially with regard to the arrangement of the cracks. You must truly generate your images, don't just rotate and translate a selection from a preexisting photo.
You should not otherwise hardcode your outputs. A generic algorithm should be used and numbers larger than 65535 should theoretically produce valid outputs. (I've restricted it merely to accommodate small-maximum integer types.)

Upgoat

@orlp debate about whether pop-con is substitute for any objectivity

orlp

@Upgoat it is objective: gather the most upvotes on a particular stack exchange

Rainbolt

@Upgoat I didn't say anything about whether or not the challenge meets our rules or not. All I said is that you keep talking about a totally different topic.

flawr

@BaldBantha -1, they are not blond

orlp

it's not 100% under your control though

Upgoat

21:34

@Rainbolt My topic is whether or not pop-con is substitute for objectivity.

orlp

"the winner of this challenge is the first person that posts the string that hashes to a number with remainder 0 mod 10^bignumber"

Bald Bantha

@flawr 4 NON blondes... ??

orlp

that's a 100% objective challenge with a purely random winner

Bálint

1

Q: Make images on last row the same size as others in flexbox gallery

I'm attempting to build a very simple CSS flex image gallery that shrinks the width of the images depending on the size of the container: HTML: <div class="grid-container"> <div class="grid-item"> <img src="small-thumb-dpi.jpg" /> <div>this is the item description</div> <...

html css css3 flexbox image-gallery

flawr

@BaldBantha I just say, they're not blond.

Bálint

21:35

Check the snippet on this question

@Upgoat ^

Bald Bantha

@flawr sho u only like dee blond ladies?

Upgoat

@Bálint ?

Rainbolt

@Upgoat Nevermind. The chat is moving too fast for me to demonstrate clearly how off topic you were a moment ago.

flawr

@BaldBantha No, but I like downvoting for stupid reasons=)

Bálint

@Upgoat Check the snippet

Bald Bantha

21:36

@flawr sighs

Bálint

Wait, those aren't goats, sorry :S

Upgoat

@Bálint :O :D +1

@Bálint close enoguh

still fellow bovids

flawr

@BaldBantha But that's how SE works, isn't it???

Quill

@Upgoat design a little, mainly about how they work

Upgoat

@Quill ping me when you publish, I need to do unit tests for cheddar too

Rainbolt

21:39

Ping me as well

Quill

Okay :) sure

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@Upgoat did you see the NTFJ answer?

Rainbolt

Or just say something funny alongside your announcement during business hours and I'll see it on the transcript

Upgoat

@CᴏɴᴏʀO'Bʀɪᴇɴ oh no, link?

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@Upgoat this thing, I added explanation

Bald Bantha

21:41

Upgoat

@CᴏɴᴏʀO'Bʀɪᴇɴ oh cool, will read in a bit. (already +1d though)

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@Upgoat k :)

Quill

@Rainbolt me? funny? ha, good one

Rainbolt

@Quill Step 1 of unit tests: get your developers to start writing them. Our definition of done for a user story has a specific unit test coverage percentage. If we close all of the tasks and our code coverage has dipped below 50%, we add a task to write more unit tests. We can't close the story until it meets our definition of done.

Poke

@Rainbolt Code coverage metrics should be taken with a grain of salt, imo

Rainbolt

21:45

That's why we hold it at 50%

Poke

even then it's going to vary based on your project domain

Quill

I check in 100% coverage tests for each controller I add during my feature PRs

Rainbolt

@Poke Are you suggesting that the percentage be removed from our definition of done, or that we adjust it, or are you just sayin'?

Poke

100% coverage doesn't even mean you test all your usecases :\

@Rainbolt more just sayin' than anything else. It's not a terrible metric and if it works for your team then far be it from me to tell you to change your process

Rainbolt

Anyway... it's unit test coverage, not use case coverage.

It's not like anyone is using it as an excuse to not test 50% of the user stories

Quill

21:57

@Poke I go for 100% coverage and 100% branch logic

SEJPM

evenin' everyone, may I ask what is a good time period to let your Q stay in the sandbox (w/o any activity on it) before posting it to main? Does 24h w/o any activity look like enough?

Poke

@Quill that sounds good as long as you're sure you're writing good/useful tests. I can get 100% coverage and have nothing really tested. In practice I've seen that striving for high coverage can result in poorly written tests and/or poorly written code

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@HelkaHomba I'm going to try to make a starting submission for your soil challenge

Transcript for