« first day (2526 days earlier)      last day (2331 days later) » 

12:01 AM
:(
 
12:29 AM
So much Blue...
 
and red laser eyes
 
Now all we need is a white guy... then AMERICA
Eh?
 
 
2 hours later…
2:36 AM
-2
Q: A program to convert MyCard number to information of birth

unknown99Would you help me to write an algorithm using array structure for the question above?

 
3:19 AM
> The least-liked person at the table takes the first turn as the Villain.
this is how friendships die
(context - superfight card game)
 
3:36 AM
CMC: given positive integer n, output every positive divisor m of n such that n/m is a square
 
3:47 AM
So... Given 12, output 3?
Given 16, output (1, 4)?
 
rip 12 and 16
 
is there a discord?
 
Can I upload an implementation of a language to Pastebin and then claim the language has an implementation even though nobody can find it?
 
@JesterTran For PPCG? There are two
 
4:03 AM
@DJMcMayhem yeah
 
Anonymous
@LeakyNun Actually, 13 bytes: ;╗÷⌠╜\;√≈²=⌡░
 
@JesterTran discord.gg/h2EyGF for general PPCG (not super active) and discord.gg/mVY6Ep for gaming
They're both unofficial though
 
thanks
 
@Mego nice
 
4:32 AM
CMC: Write a Python prelude such that the line getLine >>= putStrLn actually does what it would do in Haskell (print(input()))
 
> Python
:(
 
Were you expecting to be able to use Haskell?
 
@ATaco Wait, how does that not raise an error for undefined variable putStrLn?
Okay, change: I'm allowing any language for which an empty submission would be invalid
So that excludes Haskell, Idris, etc.
 
@ATaco For your challenge, can the last test case be in any order? the key is the same for each
 
Anonymous
4:43 AM
@EsolangingFruit My approach (done without looking at your solution)
 
@Mego You could just have putStrLn = print
 
Anonymous
@EsolangingFruit Oh right duh, I'm tired :P
 
Anonymous
I went with that implementation because it more closely matches how Haskell works
 
@EsolangingFruit Don't question Funky.
 
Whenever somebody complains that Python doesn't have [insert functional language feature], remind them that it has >>=.
 
4:45 AM
@ConorO'Brien As per the challenge spec, the key must be assumed to repeat.
So who, what, when still must XOR with 3 at the o, a, e.
@EsolangingFruit In particular though, Funky doesn't care too much about Errors or undefined variables. It's perfectly fine with the idea of a bitshift with undefined.
@EsolangingFruit Funky has a and= b, which I think is the most absurd assignment.
 
by default, can we swap truthy/falsy
i.e. falsy for valid, truthy for valid (in [decision-problem])
 
I don't think so.
 
the challenge doens't specific t/f so I think I can (just valid/invalid)
 
I think in that case, maybe.
I'd recommend asking the OP.
 
already did but low hopes
hasn't been seen since dec 23
 
4:59 AM
I love how that Python looks with the Syntax Highligher...
 
@ATaco is this a userscript?
 
Yeah.
 
not bad
 
uses google highlighting right
 
Used to, now uses Highlight.js
Link or, if you want the version which supports Funky Link
 
5:02 AM
request: option to disable shading
>_>
 
Shading?
 
shadows sorry
 
I don't see shadows, What Browser are you using? Can you send a screenshot?
 
latest chrome, os x
 
Oh, I see that issue. Might take some work, but I'll try to fix it.
 
5:05 AM
thanks
oh it's probably the comment thing trying to grey and italicize
 
CMC: return 1 for a string which contains only non-uppercase letters, 0 otherwise. (or true or false)
 
so all lowercase? or anything but uppercase letters
usage of non is unclear, VTC
 
@Riker Patched, You may need to re-download, but it should no-longer show the ghost.
 
thanks
will do tomorrow
 
@Riker the latter
 
5:08 AM
just finished an answer, g'night all now
 
5:19 AM
@ConorO'Brien Funky, 24 bytes s=>0|!s::match"[A-Z]"||0
 
5:55 AM
@ConorO'Brien Jelly, 3 bytes ⁼Œl
@Mego What if I think both way makes sense? (reply to Please don't post an answer unless you actually agree with it.) Then don't post any?
 
I wouldn't recommend posting any, fence sitting is a bad habit.
 
@ConorO'Brien Perl5, s/[A-Z]//g. Returns an empty string (falsy) if there are only upercase letters, otherwise returns a truthy string.
 
6:10 AM
@Pavel You're suppose to return falsy if there are any uppercase characters, not only.
 
Oh? I misread then.
 
(That's intentionally using the same method as your Perl answer, not a valid answer to the CMC)
 
@ATaco So I just looked at the RR standard library, and I found if. How does that work? I can't figure it out just by looking at it
Oh! I see now.
I feel it's a bit overcomplicated though, with the base conversion and all.
 
6:28 AM
That’s done so you can pass decimal arguments
 
Couldn't you just match 0? to be falsy and anything else as truthy
 
Most math is done in Unary, so if you’re passing anything to it, it’s likely going to be Unary
The decimal variant is just a convenient overload
 
But what if I want to pass in text and check it if it's empty?
It should accept foo as a valid truthy value instead of rejecting it entirely.
 
Surprising, I never considered string methods
You could probably make a length function and do IF<length<foo>,
 
Yeah, it's a bit odd that in a language about regular expressions with no concept of numbers most of the stdlib functions are about numbers.
 
6:35 AM
The stdlib was designed because of a lack of number support
 
True. I do think length should be in there though.
 
Maybe, I should probably write some more libs
 
Can't tell what language the syntax highlighter is guessing at.
 
Base2 wouldn’t hurt, neither would a string library. maybe a table library..?
 
I think all three fields got a different lang, actually.
 
6:37 AM
yeah, that’s possible
I didn’t unify their guessing, but it should use the correct language if it recognises the languageId tio uses
I’m on my phone right now, so I can’t see what the guesses look like
 
TIO uses weird language ids, unfortunatly.
 
Mmhm
 
Things like ecpp-cpp should get C++ highlighting
Here, I'll paste a screenshot.
 
It’s not bad at guessing atleast, usually
 
@ATaco Would it help to create a json linking TIO language IDs to highlight.js language ids? Including null for things like ReRegex which don't have a reasonable highlighter.
 
6:42 AM
Nope, no idea what it’s guessing. I think python for the last one..?
And yes, it would help
 
<-- doesn't actually know json and will probably run a python dict through jsonify
 
JSON is just JavaScript objects that complain If you try to do things the easy way.
 
<-- doesn't know JS
 
That’s fine, neither do I
 
6:58 AM
@Riker I didn't get to say it earlier but If you have any questions about RPI, I'd be happy to answer them. I don't have any obligations to make the school seem good so you can know my answers are honest ;)
 
@ConorO'Brien Dyalog APL, 7 bytes: ~∨/⍞∊⎕A
 
@ATaco I've discovered the version of Highlight.js that's publically hosted provides these 23 common langs, and requires you to build a version of it yourself in order to support all the langs it's able to provide.
I believe simply doing that what allow for much better language detection.
 
 
1 hour later…
8:12 AM
0
Q: Play tic-tac-toe and never lose

l4m2(There exists some challenges that require to use the best startegy, but here we don't. Even if you are able to win, you are allowed to make a tie) Play tic-tac-toe as the second player, and you're not allowed to lose. I/O method may be: Input may be the current chessboard(always your last re...

 
@Pavel common langs ≠ PPCG
 
8:44 AM
36
Q: Find the first duplicated element

Thomas A. AndersonGiven an array a that contains only numbers in the range from 1 to a.length, find the first duplicate number for which the second occurrence has the minimal index. In other words, if there are more than 1 duplicated numbers, return the number for which the second occurrence has a smaller index th...

Just read ^, may be related.
It does not implies it's possible, though, because the restriction in the question is stricter.
 
9:11 AM
CMC: every number can be represented as a point in n-dimensional cartesian space (where n is large enough for us to not care) such that each point is unique in the number they represent. The axes are the prime numbers so that every number can be decomposed as a multiple of them. For example, the number 24=2^3×3^1×5^0×7^0×... will be represented in this space with the coordinates (3, 1, 0, 0, ...). Given an input integer X, find the distance between X's point in this space and X+1's point.
a number here means a positive integer
1 is the point (0, 0, 0, ...)
 
@Cowsquack n = infinity?
 
9:52 AM
Jelly, 6 bytes
 
10:05 AM
Cowsquack just gave what is tantamount to a bijection between N and N^infty
 
10:34 AM
@user202729 I suppose so
@user202729 I originally meant for one integer to be given as input, and for you to calculate the distance between that integer and the next one, but that is still fine. Can you extend your solution to find the distance between two positive rational numbers?
0.8 = 4/5 => (2, 0, -1, 0, 0, ...)
@LeakyNun do you think this can be extended into the positive reals?
 
@Cowsquack not sure
 
sqrt(2) = (0.5, 0, 0, ...) and cbrt(6) = (1/3, 1/3, 0, 0, ...), so I guess positive real algebraic numbers follow this bijection
 
@Cowsquack no they don't
 
oh whoops, that'd just be nth roots of numbers
 
correct
 
10:50 AM
what about 1 + sqrt(2)?
 
@Riker use an anchored look ahead
 
@Cowsquack exactly
 
Any final feedback on this (The Riemann sum integral approximation Sandboxed challenge)?
 
@Mr.Xcoder if $f$ guaranteed to be integrable?
nvm, read it
 
f is guaranteed to be continuous between a and b (no jumps, no holes, no vertical asymptotes).
Oh ok.
 
11:04 AM
do x^2*sin(1/x)
 
As an example? Ok...
 
right
 
Why u do this to me? :P
 
sorry, I meant x*sin(1/x)
x^2*sin(1/x) is differentiable
 
Ok thanks. This one is much nicer.
Oh no I take my words back... THIS one is much worse
 
11:09 AM
@Mr.Xcoder f is supposed to be in a black box
 
@LeakyNun I will consider adding one example with a trigonometric function later
 
11:48 AM
@user202729 Better?
 
Looks fine. Let me write a program to check the values...
 
Ok thanks for that. On mobile rn so it wouldn’t be too pleasant to post / write a program to verify the test cases
 
BTW you do realize that you can upload the images independently and image-link to them, right?
 
I do, but I am not completely sure I want to
IMO the images should not depend on external sources, I want the challenge to be somewhat self-contained
 
Just upload it to imgur and then link to it normally. Does SE allow that?
And SE images are hosted by imgur anyway...
 
11:53 AM
umm, do they take too much space?
 
Hmm... this has been discussed on meta SE.
 
Link?
 
And, you can add the prefix "m" or "s" before the ".png" of imgur link
to make the size smaller, and it can still link to the full-resolution version.
 
i don’t have the full resolution images anymore :-/
 
Anyway... it's not really important.
The number is blurred a little, but still readable.
 
11:55 AM
Ok...
I should have kept the Grapher documents :-/ Now I could have edited in with smaller pictures at better resolutions
@user202729 I think I can check using this.
The first two are correct.
The third one seems wrong
 
12:50 PM
@LeakyNun the bijection stops being a bijection once you get into $\doubleR$ space
 
@Cowsquack the bijection isn't even defined
 
what do you mean?
 
what do you mean by the bijection?
 
you can't just "extend" the bijection given to R
you need to define it
 
12:59 PM
so how would you define it, say for Q instead of R since it works fine with Q?
 
does it?
 
ah, I understand now, (1/2, 0, 0, ...) is in Q^n space, but doesn't map to any number in Q
@LeakyNun but technically, nothing maps to 0
 
@Cowsquack Which 0...?
If you meant (0, 0, 0, 0, ...) there is 1.
 
@Cowsquack right, N>0 and N^infty then
@user202729 the natural number 0
@Cowsquack precisely
 
So it should be "nothing maps from 0", or is my English grammar incorrect? ...
 
1:05 PM
the former
but you're just being pedantic
 
I was actually confused...
 
the mapping happens both ways, so (1, 0, 1, 0, 0, ...) <=> 10
so I don't think it should matter if you map from 0 or to 0
 
Why does SE share link contain user ID?
 
@user202729 for the "announcer" badge
 
CMC: Given a number, output a list of its digits in base -2.
@LeakyNun I don't think we have that badge...
 
1:14 PM
@Giuseppe It seems perfectly valid to me: Try it online!
 
Well, "I didn't test it" means "it may be invalid"...
 
Looks like I can just get a lot of free server, or dynamic DNS...
 
4
Q: Approximate definite integrals using Riemann sums

Mr. XcoderLeft and right Riemann sums are approximations to definite integrals. Of course, in mathematics we need to be very accurate, so we aim to calculate them with a number of subdivisions that approaches infinity, but that's not needed for the purposes of this challenge. You should instead try to writ...

 
@NewMainPosts Too slow...
I often strip it from the URL instead.
(that may be the reason why I didn't get it)
 
1:16 PM
@LeakyNun haha, a lot of the links on the first page of the announcer badge are to "Build a working game of Tetris in Conway's Game of Life"
 
@user202729 Jelly, 3 bytes Try it online!
It might be invalid, though
 
Actually, I just thought that the algorithm can't work for negative base. Let me check...
 
Jelly is very well implemented
 
So... Dennis actually handled that case.
if digit < 0:
	integer += 1
	digit -= base
The -= is obviously intentional.
BTW...
5 hours ago, by user202729
36
Q: Find the first duplicated element

Thomas A. AndersonGiven an array a that contains only numbers in the range from 1 to a.length, find the first duplicate number for which the second occurrence has the minimal index. In other words, if there are more than 1 duplicated numbers, return the number for which the second occurrence has a smaller index th...

Anyone think it's possible or it can be proven impossible? It had generated a lot of comments.
The same problem with n+1 elements and without the restriction the second index must be minimal can be solved within the constrains.
 
2
A: Find the first duplicated element

Leaky NunPython 3, 94 92 bytes O(n) time and O(1) extra memory. def f(a): r=-1 for i in range(len(a)):t=abs(a[i])-1;r=[r,i+1][a[t]<0>r];a[t]*=-1 return r Try it online! Source of the algorithm. Explanation The basic idea of the algorithm is to run through each element from left to right, keep ...

 
I’d be interested to see a golflang answer to my challenge.
 
Which challenge...? Riemann sum?
Ok, Jelly should be easy enough.
 
Yes of course
I’d be able to do this in Pyth I think
 
But I think I will do other things now... like trying to install Linux.
 
Installing Linux.... hm Good luck!
 
1:32 PM
What happens if the created partition overlaps with the existing ones? wiki.archlinux.org/index.php/Fdisk#Create_partitions I just want to create a new one without any data loss.
 
1:43 PM
Also, how much memory should I let the partition be? 20GB?
(@NieDzejkob )
 
2:18 PM
This is completely out of left field, but I just want to let you all know that you're absolute geniuses and despite having absolutely nothing useful to contribute to the community at all I'm constantly in awe of everyone here. Just felt like I needed to show some appreciation for the years of entertainment and learning I've gotten from this community :)
17
 
2:44 PM
@user202729 20 GB should be enough... in fact, my system fits in 20GB (if you don't count /home), with a ton of crap installed and a comfortable amount of space left
sorry for taking so long to reply
 
Can I partition in Windows first instead of using Linux tools "as in the guide"? That would be easier because of UI.
 
Anonymous
@user202729 I don't see why not
 
Just to be extra sure. No problem.
 
Anonymous
Just be careful if you have any ext4 partitions already existing, because Windows doesn't know how to play nice with them
 
Looks like I have to defrag it before be able to shrink it significantly.
"unmovable files"?
And I can't shrink any FAT32 drive. The only NTFS drive (C:) can be shrink at most 6631 MB. Should that be enough?
(6631 MB ~ 6.5 GB, doesn't feel good)
 
3:06 PM
Hey guys!
1
A: Sandbox for Proposed Challenges

Bruce ForteRhombus sequence Tags: code-golf math sequence Imagine enumerating the elements of rhombi that which grow [1],[1,3,1],[1,3,5,3,1],… (only odd numbers such that they align nicely). This would look like as follows, note that you always begin enumerating with 1: 01 1 ...

Any final feedback for that challenge?
 
Anonymous
@user202729 You shouldn't need to defrag NTFS
 
Anonymous
Unless you're adjusting partitions so that they're contiguous
 
@Mego you're thinking ext or hfs
@Mego NTFS is prone to the same fragmentation (but less so) that FAT drives are
and if you don't regularly defrag once a year ish on WIndows, you end up not being able to properly shrink partitions
at leas,t not the way Windows wants it :p
 
Anonymous
Windows 7+ automatically runs defrag for you, so you shouldn't need to manually do it
 
@BruceForte rhombi that which?
I have never seen any such grammatical structure, though it might be correct (?).
 
3:19 PM
@Mr.Xcoder: Yeah that's a typo :)
I'll fix that, is it otherwise alright?
 
Still reading
 
Annoying! Only 6631 MB that can be shrunk, still. No idea why.
 
Strange, no OEIS sequence
@BruceForte It looks fine to me, but when it comes to challenges, I usually allow three i/o methods: take n and print the nth term, take n and print the first n terms and outputting the sequence indefinitely. Of course, that's your choice.
Maybe the nth term suits this challenge better, though.
 
@Mr.Xcoder: I guess printing the first n terms makes sense, not sure about the sequence indefinitely
 
Yeah, definitely up to you.
 
3:26 PM
But I'll add it anyways (I always liked those challenges that were that lax with IO)
 
After refresh it worked!
(unmovable files)
I think if NieDzejkob says 20 GB is more than enough (I guess that's what "tons of crap installed" means?), I will use 16 GB.
 
I found a very interesting property of your sequence. If you analyse each side length as a separate sequence you get something pretty intriguing: 1 -> 1, 2, 4, 5, 9, 10, 16, 17 (+1, +2, +1, +4, +1, +6, +1, ...)
 
"Not enough space available" - why?
 
@Mr.Xcoder: I changed IO and added a few more test cases
 
Anonymous
@user202729 For reference, I have a 17GB Xubuntu install with lots of server stuff (MySQL DB, Redis, memcached, a website, and lots of Python and Node stuff)
 
3:35 PM
What side lengths, sry if I'm stupid?
 
Sorry, I meant column index in each rhombi
For instance you have (1), (1, 3, 1), (1, 3, 5, 3, 1), ... Flatten this and take the indices of 1: 1, 2, 4, 5, 9, 10, 16, 17,... it corresponds with the sum of each of those columns (the columns that have exactly 1 element)
 
Yeah true
But not sure if you can use this for the other elements
 
Lol... If you take three, the indices are: 3, 6, 8, 11, ... The sums are three times that 9, 18, 24, 33, ...
I think Jelly is the right tool for the Job
@BruceForte It is always the column total index times the column index in the current rhombi.
\o/ 5 gold badges
 
Lol, I didn't notice that when writing up this challenge -.- Well done! Guess you're the first to post a solution (as so often :P)
 
@BruceForte I don't think I am going to be the first to post a solution. I think that implementing the algorithm I described is a bit harder than it may seem
 
3:48 PM
Just posted it.
 
I saw, +1'ed immediately
 
I noticed
Btw. I'll keep your format of those three options for challenges, if you don't mind?
 
Sure go ahead.
 
How can I make a regex that matches a and b independently, but when they are concatenated matches ab rather than a, b?
a|b|ab will match them separately when concatenated
a(b)?|b seems messy
 
@BruceForte Turns out I am still the first to answer :P
 
4:00 PM
^^
Before @MainPosts could catch up (deja-vu)
 
Ok, looks like a quick dirty solution is ab|a|b
 
2
Q: Rhombus sequence

Bruce ForteImagine enumerating the elements of rhombi which grow [1],[1,3,1],[1,3,5,3,1],… (only odd numbers such that they align nicely). This would look like as follows, note that you always begin enumerating with 1: 01 1 02 03 04 1 2 3 4 05 06 07 08 09 … ...

 
4:27 PM
@Downgoat I was able to get it anyway because I was being dumb, but thank that might help golf
 
Anonymous
CMC: Given the filename of a file containing many, many, many lines of text, return a random line (chosen uniformly) from the file. (use TIO and /usr/share/dict/words for the file for timing)
 
it's gshuf on os x if anybody wants to try it there
 
Anonymous
Python 3 (uses reservoir sampling)
 
4:48 PM
0
Q: Disentangle doubly linked data

JordanA doubly linked list is a data structure in which each node has a value as well as "links" to both the previous and next nodes in the list. For example, consider the following nodes with values 12, 99, and 37: Here, the nodes with values 12 and 99 point to their respective next nodes, with val...

 
ofc it's slightly faster in bash/C then in python :p
 
5:05 PM
@Riker you can remove the space before the 1
 
@Cowsquack fastest-code
but yea I can
 
>_< I did not see that
 
0
Q: Tuple addition in pointfree

Wheat WizardWhat is the shortest way we can express the function f(a,b)(c,d)=(a+c,b+d) in point-free notation? pointfree.io gives us uncurry (flip flip snd . (ap .) . flip flip fst . ((.) .) . (. (+)) . flip . (((.) . (,)) .) . (+)) which with a little bit of work can be shortened to uncurry$(`flip`...

 
5:23 PM
@Doorknob Glad to see you’re back! By the way, congrats on IOL!
 
@Mr.Xcoder Thanks very much! :)
 
5:42 PM
np == no problem == !p, p == problem. so, we're comparing p to !p. obviously np != p
 
@Mr.Xcoder IOL?
 
@Riker International Olympiad of Linguistics
 
> Participants competed to solve problems on diverse languages and writing systems including Birom, Abui, Kimbundu, Khom, Madak, and Emoji.
> Emoji
looks cool though, congrats @Doorknob
 
@Riker ahaha yeah wajah tersedu-sedu
@Riker thanks!
 
@Doorknob yep am reading
 
5:57 PM
nahuatl in 2009 iol wow
 
nahuatl is a problem committee favorite; there've been at least 3 problems on it :P
 
i didnt scroll past 2009 after i saw nahuatl
it was in 2011 and 2015 too
 
6:15 PM
I exist?
 
in 2010 IOL there is an mRNA problem, do other countries teach how they work? we learn it in turkey (not the specific codons for amino acids but how they work broadly and with the examples given a normal student would solve most of the problem in under 5 minutes and this sentence is too long for parentheses)
 
6:30 PM
oh yeah, that one made a lot of people mad not only because of the prior knowledge thing but also because some people claimed (perhaps rightfully) that it's barely even linguistics
 
> Language is a system that consists of the development, acquisition, maintenance and use of complex systems of communication
mRNA helps the DNA to communicate with the ribosomes ¯\_(ツ)_/¯
 
6:42 PM
are there any nethack people here? eh too late nvm
 
@Cowsquack Doorknob, me, riker
 
I was against a djinni in the gnommish mines, it even killed my pet
zapping it twice with a spell of magic missile didn't seem to do anything, then the djinni cornered me and killed me
 
7:19 PM
2
Q: Ternary Triangles

J843136028The idea of this is mainly from BIO 2017 q1. I got the idea for posting this challenge from my Binary Sequences challenge, since lots of people seemed to like it. Also, this is the first challenge I've posted without posting on the sandbox. I'll delete it if no one likes it. Rules Take in a se...

 
0
Q: ASCII Art Output Rules

James HoldernessI know there's a desire on PPCG for challenges to be flexible in terms of what they allow for input and output. However, I think there is an argument to be made that ascii-art is a special case, and should be stricter about the permitted output formats. As a case in point, there was a recent asc...

 
8:25 PM
0
Q: Maximum Area of a Polygon with Vertices of a Polygon

J843136028Areas of Polygons with Vertices of a Polygon Rules Given a list of integer coordinates, l, with a length of at least 4, and an integer n such that n is smaller than the length of l (but at least 3), return the largest area of an n-gon (n-sided polygon) that has all the coordinates of its n vert...

 
i love non-golfy languages
 
CMQ: Is there anything you can do in Clang which is golfier than the gcc port?
 
i dont think so, gcc is more flexible than clang
 
That’s exactly why I am asking
 
8:42 PM
0
Q: Summation of Ternary Triangles

J843136028The idea of this is mainly from BIO 2017 q1. This is very similar to my Ternary Triangles challenge. Rules Take in a sequence of digits in ternary (base 3); this could be as a string, an array or the numerical value along with the number of preceding zeros. For each row in the triangle, a row ...

 
9:15 PM
3
Q: Output the hours at 90 degrees

CharlieToday while playing with my kids I noticed that an apparently simple toy in the park hid a challenge. The wheel has a triangle that points to a number, but also has three circles that point to the numbers every 90 degrees from the first one. So: Challenge (really simple) Given an integer be...

 
9:27 PM
Is this question remarkably broken looking to you as well?
3
Q: Identifying a spider from my childhood memory

user5661Once, as a child, I found a spider in my parents' garden. I remember catching it in a jar because I thought it looked unusual, but I never figured out what kind of spider it was and unfortunately didn't get any pictures. I've been looking for it in books and on the Internet since, based on memory...

 
@Blue Yeah, seems broken here too
 
How strange
There's a comment on it which doesn't render either
 
what's so wrong with it? I don't think I see anything weird
 
 
+6142?
 
9:33 PM
@Blue nope, don't see anything like that on vivaldi
 
huh
Reloading it makes it work
 

« first day (2526 days earlier)      last day (2331 days later) »