« first day (1828 days earlier)      last day (3015 days later) » 

12:01 AM
why popcon
 
@flawr I finished my implementation. As you can see, it works exceptionally well on real-world photos (and videos) (even better than many commercial variants I tested), but it does suffer from the low information content in the test images (apart from the 1bit images, those are rendered perfectly). I'll type up the answer tomorrow, as the explanation is rather complex :)
 
That looks nice, what kind of algorithm did you use?
Oh ok.
 
It's a hybrid algorithm (texture synthesis and a bunch of "intelligent" fill methods) that was invented at Cambridge Microsoft Research (afaik). I improved upon that, both speedwise and in terms of auto-adjustments. I'd say it's about 20% original ideas.
The main thing is that it accepts unlimited masks (masking by user-selected color) in any shape or form.
 
You know I'm kinda surprised there aren't any queue-based golfing languages
@mınxomaτ whoa that's nice
 
12:17 AM
@quartata I wondered that some time ago.
 
hmm
 
One should be made, and called Q.
But anyway, I think there's a difficulty of sorts with queue-based languages.
 
Does anyone know (or know where post on SE) anything about the ARM GPIO?
Is there a general hardware/software SE or should I just go on Programmers?
 
@El'endiaStarman What kind of difficulty?
I just thought about that and came to the conclusion that you'd just have to more or less "reorder" your programs but you could do about the same?
Lets make a heap based one=)
 
I'm not sure. It might just be a paradigm-shift kind of difficulty. Plus, it's harder to think through the program because any operations you want to do on values you just pushed on have to wait until whatever's on the queue already gets used up.
 
@El'endiaStarman It would be certainly impractical, right=)
 
@phase The line above also has tabs.
And line 64.
 
@quartata What do you mean by queue-based?
 
And 71.
And 72 and 75...
Okay, I'll stop looking. :P
 
even more reasons as to why they shouldn't write software
 
12:37 AM
deque based language?
 
@ETHproductions Do you know a stack-based language?
 
plant based language where you have to wait a week for your program to run
3
 
@El'endiaStarman Yes
I mean, sort-of
I understand how stack-based languages work.
 
Okay, stacks are last in, first out and queues are last in, last out.
In both cases, you push stuff on the end. In stacks, you pop off the top, and in queues, you pop off the bottom/front.
 
Ahh, that makes a lot of sense. Then it won't be much different then stack-based.
 
12:40 AM
Well, actually, I do think it'd be significantly different.
Wow. Wrote the complete opposite of what I meant. :P
 
haha
Yeah, I guess so, if operators operate on the same end as is output first
 
@El'endiaStarman I can join the PPCG Minecraft server in a bit, wanna join too?
Wait, is the server down?
NVM.
 
@RikerW ooo I'll get on later
 
WOAH OVER 128 GOLD
Welp, don't leave a chest deep underwater on the bottom of a lake people.
 
I'd love to play, but I can't right now
 
12:57 AM
Some people need to sleep.
 
I was thinking about it because with traditional stacks (no swap, rotate, etc) an automata with two stacks can simulate a TM whereas only one queue is needed to simulate a TM
 
@flawr well, 5 here...
Sorry.
 
2 here
 
then again I might as well do a one counter multiplicative machine :P
Actually is there an esolang like that? I know deadfish is basically a one counter additive machine
 
dunno
 
1:03 AM
a two counter one (which woud be TC iirc) would be cool
 
I was thinking about making a stack-based esolang where there are only 10 operations: increment, decrement, swap top two, pop X and bring Xth item to top, duplicate, pop, input byte, output byte, compare top two items, and a loop.
 
@ETHproductions Brainfuck has like 6 and it's turring complete, maybe you should aim for 4
 
Then you can define an operator on each line leading up to the main program
 
Was trying to talk to myself and got ninjad
 
@phase 8, and it'd really really hard to implement any one of these operators if it was removed
@quartata Yeah, that happens
 
1:10 AM
Anyways I actually started work on such a language with 4 bit instructions but ran into some issues trying to make non trivial programs
 
@CᴏɴᴏʀO'Bʀɪᴇɴ was doing a 6-instruction language, but he and I found about 4 more operators necessary to make it Turing-complete.
 
just implement one operator that emulates busy beaver and bam
 
What were the original 6?
@phase haha
 
implicit input/output would also help
 
Anyways I think I'll try to make a TC deadfish. I'll need an loop and two counters
 
1:13 AM
in PPCG Code Snippet Chat Bot, 23 hours ago, by Cᴏɴᴏʀ O'Bʀɪᴇɴ
Which are: NAND TRUE FALSE JUMP OUT BIT
@phase And this language has implicit input as well
It now also has pop, dupe, and loop conditional
conditional combined with jump == loop
 
@ETHproductions yeah you need a loop or otherwise it cannot emulate partially recursive functions
 
Really though you only need two instructions: instruction and next instruction in set
 
| - NAND
# - push 1
~ - push 0
@ - pop 8 bits and make a byte out of them
^ - pop X, jump to instruction X
* - output
() - execute code inside if top != 0
$ - pop
: - duplicate
@quartata Yeah, that would make a good bit-based esolang
 
I think I might do that instead of a 4 bit stack based language.
 
1:22 AM
cat is (*~^)*
 
no matter how verbose it'll be it will still be golfy :p
@ETHproductions cool
 
He had H,W! as well, let me find it
Oh duh, it's there when you load the page
Now, how would we do TM?
I think there's a bug that makes it impossible
 
(*)* right?
 
No, 48 or 49 will always be on top of the stack, since input is in ASCII bytes
Now we need to figure out how to subtract 48 using only NAND....
@quartata NAND on bytes is 255-(x&y), right?
@CᴏɴᴏʀO'Bʀɪᴇɴ Bug fix: jsfiddle.net/kgtn0nd4/5
 
depends if it is signed or unsigned
 
1:33 AM
unsigned
 
Sorry I'm not here to talk. I'm busy today ^_^
Thanks @ETHproductions :D
 
then yeah that is right
 
I think this is TM:
##~~~~~#@|########@|(:########@|##~~####@|*):########@|##~~####@|*
You'll need the updated version I linked a couple minutes ago though
Yes, subtraction takes two byte-pushes and NAND operations :/ Oh well, this lang wasn't made to be a piece of cake!
 
@phase you gonna come on?
If not I am leaving maybe, IDK
Anybdy else for PPCG Minecraft Server?
 
1:48 AM
0
A: Loopholes that are forbidden by default

DennisAbusing native number types to trivialize a problem It is common practice to restrict challenges to cases where input, output and/or intermediate values of the algorithm of choice fit into the language's native number type. At least for input and output, this is generally assumed even if not sta...

 
Can I search deleted posts? I know of another example of ^ that, but I can't remember what question it's on and the regular search doesn't pull it up.
 
@ETHproductions you still have 6 instuctions you can implement and still fit in a 4 bit encoding. Maybe add and subtract should be added
@ETHproductions Mods can I think but not nornal users
 
@quartata For Conor's language or my hypothetical stack-based language?
@quartata darn
 
the nand one
 
OK
Oh, I remember the question now
 
1:55 AM
I specifically had to make catalan numbers be "works up to n=19" to avoid boolfuck junk
 
Question I can't find: count the number of 1's in a binary representation of a number
That had a boolf*** answer
 
Oh I know that one
I think it is called count the number of 1s in your languages integer type
 
1
Q: compute number of 1 bits in a binary representation of a number

Ali.SThe goal is to write a program that returns number of 1 bits in a given number. examples 5 -> 2 1254 -> 6 56465 -> 8 goal the winner is the code which runs in minimum time limit, you can assume that normal int value in your language can handle any number (no matter how large) and ...

 
no not that one
 
No, it was more descriptive
 
1:59 AM
5
A: Print the amount of ones in a binary number without using bitwise operators

Thomas KwaBrainbool, 2 ,. The most reasonable interpretation, in my opinion (and what most of the answers use) of "highest number your language is able to handle" is "largest number your language natively supports". Brainbool is a brainfuck derivative that uses bits rather than bytes, and takes input an...

 
this one had a jelly answer try lookin that up
 
@Dennis Thank you. that's what I was looking for.
 
Search term popcount did the trick.
 
ah there we go
 
popcount?
 
2:01 AM
Population count, i.e., number of 1s.
 
Ah
You should add that answer to the loophole post.
 
also dear lord the number of snippet answer on that question
i think would run out of flags trying to flag them all
@Dennis the bf one and the lisp one are both invalid
The scheme one is not golfed
 
Oh, I just remembered, I got on here an hour and a half ago to actually do something specific
:P
If you don't mind, I'm going to leave chat now
 
@ETHproductions I was (desperately) trying to. (bloody DSL) It is done.
 
Great, thanks!
 
2:08 AM
@quartata Those answers are from 2012 and 2011, so they predate our defaults for I/O.
Assuming that's what you meant.
 
Bye. (to ETH)
 
Does it not apply retroactively?
 
IDThink so.
 
also the shell one breaks the rules of the challenge (no base conversion)
 
Applying that kind of rules retroactively would require revisiting thousands of answers every time community consensus changes about something. That's not at all practically (and not entirely fair).
 
2:11 AM
Okay, but the shell one should probably go.
 
fair enough
 
After that this happened.
 
And it went. To be fair, I wouldn't expect a restriction on base conversion in a challenge called Print the amount of ones in a binary number without using bitwise operators.
 
I was kinda wondering why there were any snippet answers at all... clearly these were from the ancient times :P
@RikerW can you not see deleted answers?
 
2:14 AM
Nope.
I have like 450 rep.
Not 1k/2k.
 
oh I thought you had more
 
Nope. :(
Not "Shell".
 
@quartata Back then, you could have just downloaded the proper program from pastebin or post print "the amount of ones in a binary number without using bitwise operators".
 
Also, the shell answer should have been bash?
 
it's 2k for deleted answers on beta sites iirc
@Dennis ah those were the days :p
 
2:16 AM
*/s
 
Back when Dennis was still a fractal thingy... :P
 
Wat?
 
@Doorknob @Dennis joined in December 2013.
 
Dennis is a fractal?
 
@Dennis ... wow, really? o.o
 
2:18 AM
Dennis is a fractal? I though he was a paraguayan.
:P
 
@Doorknob Yup. Code trolling lured me in.
 
WHAT FRACTAL THINGY?
 
identicon
 
Nope.
 
No, the other thing.
 
2:20 AM
Pics pls?
 
^ my old avatar
 
oh right that thing
 
Cool.
 
I only changed that a few months ago, around the mod elections. I figured if my username is that common, I should have at least an easily recognizable avatar.
 
iirc you changed that around TPLQ
Hadn't made my account then but was stalking PPCG heavily
 
2:23 AM
^
same
 
even stalked chat
 
that I did not do
 
I am a little disturbed by how much I stalked PPCG actually
maybe should see a psychologist
 
I'm a sertified armchair psychologist.
SAP for short.
 
@quartata I was appointed on August 31, and I changed my profile pic on September 3. The PLQ started on August 17, so we're both right.
 
2:30 AM
Pop quiz: when did I change my avatar?
 
Never?
 
Correct!
 
\o@
What did I win?
 
1 chat star
 
Still better than exposure.
 
2:35 AM
Wait wasn't \o@ the very first Jelly answer?
 
It was. Sadly, the syntax has changed.
 
I cannot remember what it was to
 
20
A: Cover up zeroes in a list

DennisJelly, non-competing 3 bytes This answer is non-competing, since it uses features that postdate the challenge. o@\ Try it online! How it works o Take the logical OR of its arguments. @ Reverse the argument order of the link to the left. \ Do a cumulative reduce, using the lin...

 
What do I win for recognizing it?
I'd like a puppy.
 
No problem. Send me your mail address, SSN and CC number (for ID verification). I'll ship it right know.
^ not a scam
^^ real puppies
 
2:39 AM
oh boy!
Real puppies are my favorite kind of puppy
5
My bank account is an old tube sock that fell behind the dryer
Can you verify my ID with that?
 
I could, but I don't want to.
 
(not a stolen joke from Dilbert)
 
@El' @Cᴏɴ I've updated the Crayon docs again, mainly about regexes this time: github.com/ETHproductions/Crayon/blob/master/docs/Regexes.md
 
I may change the meta-character from @
 
2:43 AM
@ETH That what I call commitment. Even your pings are golfed.
 
to ' probably
@Dennis Thanks, haha
 
don't use @ that will give you a severe disadvantage in nethack related challenges
 
Does @El' work?
yep
-1 not golfed enough :P
 
@...
 
2:44 AM
@Doorknob Fixed
 
Did you know, that if you make a list of every possible chess game, taking into account every possible move for each player at every turn, that there will be more possible chess games than there are dollars in my bank account?
 
Er yes
 
@PhiNotPi Maybe more chess games than all the dollars in the universe? :P
 
Pretty sure it's more than the number of atoms in the observable universe
 
Did you know, that if you make a list of every tic tac toe game, taking into account every possible move for each player at every turn, that there will be more possible tic tac toe games than there are dollars in my bank account?
 
2:46 AM
The better question is are there more chess games than dollars in Donald Trump's bank account
 
9! possible tic-tac-toe games (not counting early endings)
 
@Sp3000 haha
 
== 362880
 
Early endings count. Trust me.
 
Google says the exact number is 255,168.
 
2:46 AM
Did you know, that if you make a list of every possible checkers game, taking into account every possible move for each player at every turn, that you'll get a pretty big number I guess idk
 
^^ What Phi said
If you take all drawn games, you'd still have more than my bank account
 
Breaking news: Sp3000 has 255,167 dollars in the bank (possibly less)
 
Actually, my bank account is roughly the number of tic tac toe games that end on the fifth move :) (each player moving counting as one move)
(pretty broke)
 
Did you know that if you count all of Dennis's rep taking into account upvotes and downvotes that you get more than a whole lotta dough
 
2:49 AM
@quartata I would totally exchange all my network rep for $1/rep point if I could. :P
 
Who wouldn't?
I'd quit my day-time job and become a professional golfer.
 
What about 1¢/point?
 
That would give me... about $1,000.
 
not so worth it
 
With the rep cap, that's only 2 dollars per site per day. Not even Jon Skeet could make a living that way.
 
2:52 AM
But Jon Skeet could stop posting now and still get $2 per day for the rest of his life
(plus bounties)
 
Still, 8.4k dollars sounds pretty cheap for 32.7k answers.
Almost thirty-three thousand answers. O.O And I thought I had an addiction...
 
How old is SO? 6 years? 8 years?
A more important question is, when did Jon Skeet sign up?
 
7 years, 4 months ago
JS~SO
 
grc
Did you know, that if you count the legal positions in Go, you get 20816819938197998469947863334486277028652245388453054842563945682092741961273801‌​537852564845169851964390725991601562812854608988831442712971531931755773662039724‌​7064840935
 
That's a lot of Go positions.
 
grc
3:02 AM
apparently google's AI is just starting to beat humans at Go
 
Which humans? Beating me should be pretty easy, as I have no idea what Go is.
 
grc
s/humans/human champions/
so this is a bit outdated now
 
But they're still working on Calvinball.
 
> Because Mathematica itself does all the heavy lifting, this is a community wiki.
ಠ_ಠ
CW is not a rep waiver. (╯°□°)╯︵ ┻━┻
 
Programming languages are typically responsible for most of the heavy lifting on this site.
 
grc
3:07 AM
we should make an account for Mathematica and let it post answers itself
 
@El' Have you looked at the Encoding page for Crayon?
 
I glanced at it yesterday (and just now).
 
I have some ideas for the visible substitutes for these chars (mostly arrows). I'll update it in a sec
 
Crayon?
 
My new language for ASCII-art challenges. I'm currently working on the spec. github.com/ETHproductions/Crayon
 
3:14 AM
Oh, neat. Before I started with Jelly, I toyed with the idea of making an ASCII art language.
 
@Dennis I commented on the question following that one, and it turns out the author allows inpainting builtins so it doesn't even need to be Community Wiki. Some people just seem to have a CW reflex
 
I just found out that CodeForces sorts accepted solutions by length.
So we could totally play code golf on it.
 
Oh shoot, I have like 20 notifications for the PPCG redesign project
 
@Dennis I mean "doesn't even need to be CW according to people who use it as a rep waiver". There's some argument for making it CW because two answerers were discussing collaborating, but even then I'd rather it not be CW.
 
Anonymous
Everyone should check out the TryItOnline chat bot if they haven't already. It runs code in chat. It's a little clunky at the moment, but it works!
 
3:18 AM
I don't think this should be a community wiki. — Dennis ♦ 1 min ago
 
If we upvote that comment enough maybe it will change back
 
Can one unwiki one's own posts?
 
I've never wikied one so I can't check
 
How is it offensive that someone makes their post Community Wiki?
confusde
 
3:20 AM
I assume it's mod only as otherwise you could trick people into improving your post and then take the rep...
@feersum In this case there is nothing wrong with their answer and they should get the rep for it. The offensive situation is when someone posts knowing their answer is invalid, and tries to get around deletion by making it CW
 
@feersum It's not offensive, but CW means this is a collaboration effort and you're free to edit it, not I don't want the rep.
 
It used to be incorrectly used for unabashed breaking of rules, like this
 
looks through commit history, randomly finds one commit on its own branch
 
Anonymous
Hey @Dennis could you pull Seriously for TIO?
 
@Mego Pulled.
 
Anonymous
3:33 AM
@Dennis Thanks! I was wondering why Data wasn't executing the code correctly :P
 
This seems to be the one problem with TIO: it has to be updated manually.
 
Ruby's OptionParser is just amazing.
 
It's never a bad time for some Bach: youtu.be/dYV3d-UVBKg?t=8m8s
 
Anyone here heard of Walk Off The Earth?
 
@PhiNotPi oh shoot I've been Bachrolled
 
3:45 AM
@ETHproductions Of course
 
@AlexA. I'd never, until yesterday
 
TIL that LTE is just as fast as 3G if your service provider caps the speed at 3 Mbps.
 
They have some pretty awesome stuff
 
@Doorknob Well, you've heard of Roll Over Beethoven...
@Dennis .__.
 
@ETHproductions That's not too much of a problem. I presume @Dennis doesn't want to allow people to run arbitrary code on his servers (which they would be able to do if he automatically pulled new changes).
 
3:46 AM
^
 
True.
And buggy versions would be uploaded anyway and TIO would break until the bug was fixed.
 
@Dennis No, a non-moderator user cannot remove the wiki
 
What about a sporadic one?
 
ಠ_ಠ
 
That's the face Alex makes to show appreciation of pedantic word humour
3
 
3:50 AM
A user can only remove CW if they cannot be parsed with a regex.
 
I cannot be parsed with regular expressions, only sporadic expressions.
 
If I'm made of HTML, does that qualify me?
 
Regular HTML?
 
No, sporadic HTML.
 
@AlexA. Sporadic expressions: Like regular expressions, but summon nasal demons if the time since the epoch is divisible by 10.
 
3:54 AM
wtf are nasal demons
 
boogers
2
 
@Calvin'sHobbies Okay thanks
 
@Doorknob o.c
 
I hate to think what will happen to anyone who enters this room now and glances at the starboard
 
3:56 AM
I love to think what will happen to anyone who enters this room now and glances at the starboard
 
How do I wrap the whole stack in an array in CJam?
 
] i think
 
@ThomasKwa is the guinea pig
 
In GolfScript it's ]
ninja'd
 
@Calvin'sHobbies s/the/a/
@Calvin'sHobbies s/guinea pig/hampster/
 

« first day (1828 days earlier)      last day (3015 days later) »