« first day (2494 days earlier)      last day (2342 days later) » 

6:00 PM
@ETHproductions Well, mine and some other public colleges here use eduroam directly.
 
6:11 PM
My current school has "<school name>Visitor" with no login required and "Air<Sports program name>2" with a login that's notoriously hard to set up, and then a departmental "Secure" network that somehow plays a role maybe. I've lost track.
 
6:28 PM
wow... my comp sci professor's homework rules are kind of dumb
"Documentation is not given any weight for this homework. However, you must do proper documentation, as it must be in your programming habit by now. The absence of documentation may lead to penalty."
You don't earn points by commenting your code, but you can lose points by not commenting... isn't that exactly the same though?
 
@ETHproductions I had a similar dumb interaction with my comp sci professor. Comments should just supplement readable code. You don't need to comment every line, just confusing lines. And yet she took away points for not enough comments until I had a comment on every line.
 
Yeah, it's even worse when the expectations on paper are different than how submissions are actually graded
 
@DJMcMayhem well then don't golf your homework :p
 
I've earned 7.5 out of 7.5 points on each homework so far, simply by getting everything working, making my code neat, and commenting appropriately
Meanwhile, I'm pretty sure about half the class is struggling to just get it working
 
You write code for fun. Of course it'll be slightly easier for you :P
 
6:33 PM
oops, read that the wrong way at first, haha
previous experience definitely helps, even if it's not with the same language
Anyway, I've got to catch a class. Talk to y'all later
 
Lol... Our CS teacher was asked by a pupil to recommend him another programming language to learn (we learn C++ (eugh!) in school). They said that Python is among the best choices. Two weeks later (today), the boy asked them a question about Python classes. They said they don’t know Python......
The same professor lost our test papers a couple of weeks back
 
Learning C++ isn't terrible, but the professor not knowing Python is... Well, ^ there you go. 5/7 standup professor, would take classes from again
 
Our french teacher 2 years ago lost our exercise books 2 months into the school year, then tried to fake it through the rest of the year by giving us worksheets and watching french movies :P
 
What are those important tests that you have to take at the end of the semester called in English?
 
6:41 PM
Pfff not what I was thinking about. Maybe they don’t exist in England
 
We have mock exams in January/February, but don't really have exams at the end of terms
 
Oh, “thesis” I think
Anyway, we had the <enter “fancy test at the end of the semester” name> In geography
 
Finals?
Thesis is closer to an (academic) article than a test
 
7:02 PM
0
A: Sandbox for Proposed Challenges

Jakob LovernDNA Quine Problem Description Design a quine that outputs its own source code, but encoded into amino acids. Take your source code and expand it into base 4. Each digit now maps to a nucleotide like so: 0 -> A 1 -> C 2 -> G 3 -> T A set of three nucleotides produces an amino acid. You can f...

 
7:21 PM
CMC: output ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz
 
+rG1G
(pyth)
 
@ConorO'Brien Brachylog, 5 bytes: ẠụwẠw
 
Anonymous
@ConorO'Brien Actually, 4 bytes: ú;û+
 
@ConorO'Brien Japt, 4 bytes: ;B+C
 
05AB1E, 2 bytes: žn
 
Anonymous
7:27 PM
Quick strawpoll for those present: what would you say are the top 3 stack-based languages on TIO for golfing (in terms of short code across a lot of different challenges)?
 
1. rotor 2. rotor 3. rotor
 
05AB1E surely is first?
 
11/10
 
1. 05AB1E 2. Neim 3. Actually
 
But seriously I would probably put MATL first or second, some of its builtins are surprisingly versatile
 
7:28 PM
@ConorO'Brien V, 6 bytes: ¬AZ¬az
 
It just doesn't have a codepage is all
 
@Mego stacked obviously /s
 
@ConorO'Brien (?A..?z).each{|i|$><<i if i=~/[a-zA-Z]/}
 
Anonymous
I'm looking for serious answers for a project I'm working on
 
@ConorO'Brien 4, 67 bytes: 3.60001601066216562202822620268205210212100120200091222200021210194
@Mego 1. 05ab1e, 2. MATL, 3. idk
 
7:32 PM
@ConorO'Brien alternate pyth: srB1G
Serious answers? 3. Seriously.
 
Anonymous
I'm just going to assume all "idk" votes are for Actually, since that's what most people think when they see Actually programs
 
@Mego 1. Brain-flak. 2. Brain-flak. 3. Brain-flak.
 
Anonymous
@DJMcMayhem Minimalist brainfuck-like langs need not apply :P
 
Is this for PPCG v2?
 
CMC: output ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz
 
7:38 PM
@StevenH. V, 3 bytes: ¬Az
Uhm... wat
 
@StevenH. 4, 39 bytes: 3.6000162065621578215200202000121210094
 
@DJMcMayhem wat
 
@DJMcMayhem That looks like maybe the same user editing their own answer?
 
the heck
 
@DJMcMayhem That's ais523
 
Anonymous
7:41 PM
@Pavel Nope, just a neat little thing I'm doing
 
@DJMcMayhem Accepted, pretty sure that's ais523
 
@cairdcoinheringaahing they deleted their account again I guess?
 
I don't get why they would go to such great lengths to not get reputation to prove that they don't care about reputation
4
It doesn't make any sense
 
I don't understand why not CW
 
Or do they just create a throwaway account for each answer now?
 
7:43 PM
@ETHproductions Yeah, they're doing a thing where they post answers, then delete the account to avoid the rep gains
 
@StevenH. (?A..?z).map{|i|$><<i}
 
@DJMcMayhem I think it's more complicated than that
it's not just that "they don't care about reputation"
it's that reputation brings them bad feelings and whatsoever or something
 
I would rather have him do that than lose half of the world's Brachylog users :p
 
Here's a conversation that explains it
 
Anonymous
Or, crazy thought: just ignore the rep?
 
7:44 PM
mainly because they hate the mod tools and such they don't let themselves gain reputation
@Mego or even just anonymize every answer, but I'm not sure if that's possible, since it likely involves SE staff and such
 
@DJMcMayhem Thanks! That was my 50th edit review :P
 
btw can anonymous accounts actually get privileges?
 
Anonymous
@EriktheOutgolfer Nope
 
@EriktheOutgolfer No
Of course, ais523 could explain it all himself :P
 
Anonymous
They don't have actual profiles, and no rep tracking is done
 
7:46 PM
I wouldn't have to delete the account so often if not for the fact that you can't delete an account immediately unless you have only one logged action
so the way account deletion works actually forces me to use a different account for each post
 
@Mego then I don't see why not just create an anonymous (i.e. not registered) account and post all answers from that one, since they don't get the privs anyways
 
there's one I forgot to delete, let me try to find it
IIRC it has a profile page
 
So do you keep around @ais523 for chat
 
also an anonymous account is purely cookie-based and my browser doesn't store cookies long term
@Pavel actually I tried to delete @ais523 generally but SE sucks at deleting accounts :-P
 
@ais523 What's this one?
 
7:48 PM
@ais523 I think you can restore it via e-mail address, or make an exception for PPCG's cookies
so that way you don't have to hassle yourself with deleting with every answer you post
 
@EriktheOutgolfer there are meta posts that imply you can restore an anonymous account cookie via the email, but I tried, it only gave me the option to convert it to a regular account but not to restore the cookie, so something changed at some point
@ETHproductions that's the one I forgot to delete
thanks for finding it for me
 
@ais523 still, you can make exceptions for cookies, right?
 
I have two browsers
one of them, my regular browser, is locked down heavily and can't contribute to SE because it doesn't run things like JavaScript
(the web is much less frustrating without JavaScript)
I use the other one for JS-heavy sites, like contributing to SE
but it's set to wipe everything very frequently (not just cookies, but things like history too)
 
Any python golfers got some way to shorten this?
 
@ais523 Except where nothing works without JS.
 
7:51 PM
Note, I don't think backticks can be used as we don't want a pesky L at the end.
 
@Pavel surprisingly many sites actually work better without JS
although I guess it depends on what sort of sites you visit
 
@FlipTack .endswith is nasty
 
I belive PPCGv2 is being built with noscript in mind.
 
@cairdcoinheringaahing Not as nasty as the alternatives
Splicing is longer
 
@Pavel dunno if that's possible at all
@ais523 oh
 
7:52 PM
actually, even SE works fine JS if you're just viewing it (the only things broken are contributing and expanding hidden comments)
 
How to tell if a post is ais': 1) it must be extremely well golfed 2) it's most likely in Brachylog 3) it's most likely from an unregistered user 4) it has a nice explanation 5) if the language has a quirk that makes the answer longer, the language is subtly being criticised at the end for that :P
 
@EriktheOutgolfer Why not?
 
Anonymous
@EriktheOutgolfer It is possible. We're doing it. We'll have basic form submission instead of fancy AJAX as an option for noscript.
 
@Mr.Xcoder there's at least one by me that I think people generally didn't realise was me, but that's at least partly because it wasn't in Brachylog
 
You lose things like the page autoupdating when there's a new post, but that's not really necessary.
 
Anonymous
7:53 PM
Of course, that means smashing the F5 key to view new posts instead of using a websocket, but that's the tradeoff that has to be made
 
@Mego ninja
 
@ais523 Link please ? :P
 
let me try to remember where it was
 
@ais523 Was it this?
Or was it not in Jelly either?
 
that one was mine but it's not the answer I was thinking of
(I thought people would guess that was me)
it's this one
which is a trivial answer to a trivial question, so there's less room to show through
 
7:55 PM
Yeah I could have told immediately that was you, quite obvious
 
huh, that answer pretty much looks like yours
 
That was the next one I was going to suggest :P
 
I'm not really trying to hide
 
Because it has a huge explanation for such a triv answer
 
@ais523 In case you don't get notifications for your deleted accounts, I golfed a byte off that answer. Do you mind me editing?
 
7:57 PM
I have no objections to people editing posts made by deleted users, whether or not they happen to be me
actually, part of the reason (but a very minor part) I deleted my account was in the hope that people would start seeing my work more as belonging to the community and be less afraid to edit it
anyway, given that I'm starting to participate more and more frequently again
I might consider creating a new account, but am unsure of the morality of repeatedly dropping 500-point bounties on random things
 
@ais523 Have you ever considered simply posting everything as a community wiki? That would accomplish both of your goals (not getting rep, freely editing)
 
yeah the bounties should have a point
 
perhaps I should just go up to 20k and delete the account again, but I'm not sure I could cope with all that much rep
@DJMcMayhem huh, that probably works actually
I'd need enough rep to be able to CW posts though
 
@ais523 There was a monthly underappreciated posts bounty, if you want to take that method
 
But no more
 
7:59 PM
where's the threshold for that? maybe I can get it from edits
 
You need 10 rep
 
(the problem with getting it from answers is that if I CW the answer I'll lose the rep from it and not be able to CW answers any more)
oh, that's 4½ edits, should be fairly easy
 
What is python's sys.maxint? I was under the impression that python had arbitrary precision integers.
 
@Pavel probably the largest integer storable in a system word, given what module it's in
 
@ais523 no, 5 edits
 
8:00 PM
I don't know for certain though
 
@ais523 10.
 
@Pavel It does, but after a certain point they're long integers
 
@EriktheOutgolfer You start at 1 ಠ_ಠ, not 0
 
@ais523 Or you take the association bonus
 
Anonymous
@Pavel In Python 2, it's INT_MAX (from C)
 
Anonymous
8:00 PM
It doesn't exist in Python 3
 
Or you could post an answer and convert it to community wiki after it gets an upvote (or do you need more rep to do that?)
 
@Mr.Xcoder well, how do you gain 9 rep starting from 1 rep to begin with?
 
4½ edits :P
 
5 edits and a downvote?
 
@EriktheOutgolfer 5 edits and downvote an answer
 
8:01 PM
you need 125 rep to downvote though
 
@DJMcMayhem DV at 125 ಠ__ಠ
 
it's easiest to get an odd rep boost from a question upvote
 
Yeah, then just cast 115 DV :P
 
you can't downvote at more than a 2:1 ratio
so you'd need 230 upvotes too
 
I was kidding
 
8:02 PM
that could take a while :-)
 
@ais523 Why?
 
You'd also lose DV priveleges
 
@DJMcMayhem remember you lose the privileges if your rep falls ;)
 
ninja'd
 
duh ninja
 
8:02 PM
hehe
 
@cairdcoinheringaahing something about stack exchange not liking negativity
 
Even maxing out your votes each day, that'd take 8 or 9 days, right?
 
also I think I got the ratio backwards
 
As amusing as this conversation is, you could simply use you puzzling account to take the association bonus, then you can edit and post CW's
 
and yes, optimal voting would be to upvote questions and downvote answers I think, and even then you can only get 50 votes per day if you cast them in the right order
 
8:04 PM
the problem is that, well, people will soon start to complain about "serial downvoting" and other such minor stuff but that's minor :p
also btw there's no practical difference between 10 and 11 rep
 
just make sure all the answers are by different users
 
... or you could just downvote what is downvote worthy
 
yeah that too there's always plenty of that stuff
 
it could be because I learned voting behaviour at Slashdot, but it took me ages to get the votecap badge
basically, I saved up questions I wanted to upvote until there were 50 of them
then upvoted them all on the same day
somehow I suspect this is not the intended effect of the badge
 
8:08 PM
I've actually never gotten it
 
it's one of those badges that's easy to get if you want to and very unlikely to get by chance
 
I didn't get Vox Populi either
Although I wasn't allowed to vote anymore for like 3 days in the past, dunno why I didn't get it
 
I don't think you get vote restricted for being terrible at voting, so it must have been a consequence of hitting some sort of cap
 
Yes, I have hit vote cap and didn't get the badge
 
I'm going to give up trying to explain SE software because a) I can't and b) trying only makes me angry
hmm, if I do create a named account, what display name should I use? would user62131 increase or reduce confusion, for example?
 
8:13 PM
How about ais523?
 
why not just ais523
 
that's the other main option
 
ais62131
 
Or how about ais524?
 
ais523 is a username I generally use in programming topics
especially esolanging
 
8:15 PM
Does anyone know what is wrong with this function r=sum.map(read.(:[])).show?
 
jimmy523
2
 
I'm trying to make a short haskell function to sum the digits of a number
 
anyway, I should go to bed
no point making decisions when tired, it rarely works out well
 
good night
 
Adding r::Int->Int seems to work, but I really don't want to pay to add a type signature if I could not.
 
8:23 PM
@WheatWizard I remember having this problem before. But it works for me now. Also (:[]) can be pure.
 
@H.PWiz It works if you call it on a number, but not if you just declare it, by itself.
Now I'm wondering if answers that require being called to work are valid.
 
Oh... interesting I can define it in the repl
 
Best I have is plus 4 bytes for (+0) or (*1). I'd love to know what's wrong though
 
The problem is that the compiler can't figure out the type for the function, because show has multiple instances, once it sees it needs to be a number it figures out the type you want.
 
8:38 PM
@WheatWizard sneaky compiler flag?
 
That's longer than the other options
 
1
Q: Are functions that only work when called valid as answers?

Wheat WizardIn Haskell it is possible to define functions that will cause a compiler error unless they are called elsewhere in the program. For example the following code causes a compiler error r=sum.map(read.(:[])).show main=putStr"" Try it online! while the extremely similar r=sum.map(read.(:[])...

 
@DJMcMayhem My Question
 
0
Q: Save American Society!

tfbninjaStory You are an American spy, tasked with saving your country from the inevitable release of extremely sensitive documents being held a maximum security base, deep underground in the middle of the Pacific Ocean. If they are released to the public, society could collapse. You have been tasked wi...

 
Hey there it is
 
8:51 PM
Cool
I'm looking for that one Jelly Quick I was thinking of
 
@ConorO'Brien APL (Dyalog), 9 bytes: ⎕←⎕A,819⌶⎕A
 
@DJMcMayhem I just breezed through the Jelly docs..... wow
 
@StevenH. APL (Dyalog), 11 bytes: ⎕UCS 64+⍳58
 
@ConorO'Brien SOGL, 3 bytes: Zz+
 
CMC: Sum the digits of a number. (You may take input as string).
 
9:00 PM
@StevenH. SOGL, 5 bytes: ΖAzΝ∑. The last char can be removed if the characters can be on separate rows
 
@DJMcMayhem Do you mean ƭ (cycle through atoms)?
 
Yes (see my answer)
 
Yeah, just saw it
 
@Adám SOGL, 3 bytes - encode to base 10, sum that
 
Hmm, three answers deleted in 15 minutes doesn't suggest clarity in the challenge
 
9:05 PM
Even though I deleted my answer, I found a cool trick.
O%32 could be used in place of ŒlO_96 which is super convenient
@Adám Surprisingly terse in Brain-flak: 30 bytes ({[((((()()()){}){}){}){}]{}})
 
This is why edge cases such as spaces shouldn't be included in code golfs; my Jelly answer is 10 bytes longer because of it
 
@H.PWiz I get the same error when adding (+0) or (*1): Try it online!
 
@Laikoni at the end of the line
 
Oh right, the show is causing the problems.
 
9:20 PM
CMC: Reverse the digits of an integer with constant memory. (I have no idea if this is possible but I feel like it is)
 
@ConorO'Brien So technically, converting to a string would not be constant memory because the size of the string changes?
 
correct
 
@ConorO'Brien This?
 
I don't haskell >.>
but if that uses recursion, I'd think not, since that allocates memory on the call stack
 
I'll try testing it
 
9:26 PM
(which is proportional to the number of digits roughly speaking)
 
@ConorO'Brien With bounded 32-bit integers the memory can easily be constant. With unbounded integers I don't think it's possible.
 
@Laikoni I wouldn't say unbounded, just type independent
 
@ConorO'Brien Oh wait, do you mean by constant memory that no further memory is allocated apart from the initial memory for the integer?
 
@Laikoni I mean a constant amount of memory is allocated, independent of the input
 
You can potentially get constant memory including the input if the input is given in a stream.
 
9:29 PM
@ConorO'Brien Yes, memory is increasing with more digits
 
since unbounded integers can emulate integer lists, I wouldn't consider those since that's besides the problem
 
Well not for this particular problem but for some problems.
 
right, good idea, I was thinking from terms of a stream of digit characters, but I decided to generalize a bit
this would be helpful in many esoteric languages, all of which I see utilize linear memory for custom integer input, which becomes bothersome for further generalization for, say, brainfuck into higher level languages
 
why isn't the traditional answer correct for this
 
int out = input % 10; while (input /= 10) {out = 10*out+input%10;}? Pseudocode of course
 
9:31 PM
^
 
oh, neat
 
That was mine, minus the recursion instead of while
 
codegolf might be getting to you if that wasn't your immediate answer, haha
 
import math
def f(n):
    l = int(math.log10(n))
    new = 0
    for i in range(l + 1):
        new += int(n / (10**i) % 10) * (10 ** (l-i))

    return new

print(f(123))
@ConorO'Brien ^
It does make an integer, but ignoring arbitrary size integers, that should still be constant memory
 
@Poke I call it coding in nothing but BF and similar for a little over a week
 
9:35 PM
@ConorO'Brien yep, that would do it ;)
 
@DJMcMayhem length of a number is not constant memory
 
he's not allocating arbitrary memory tho, it's still only single integers
 
If it was in C it would be constant
 
(ignoring the range of course)
 
There's a true-constant-memory algorithm though
even for arbitrary sizes
 
9:37 PM
it's the equivalent of reversing a list but with digits
 
@ConorO'Brien it looks like python3 in which case range doesn't precompute a list
 
@Poke I don't know how python optimizes stuff but that makes sense. python2 needs xrange for that right?
 
mhm
 
@ConorO'Brien It's basically a generator rather than a list
 
9:39 PM
ah, I see
 
Do you know what a generator is?
 
yeah, they have those in JS and ruby
 
PSA: New tag created (see meta), would really help if people could begin slowly retagging where applicable (but not so rapidly as to flood the front page)
 
@DJMcMayhem You know I thought I knew what generators were but it really didn't click until now. Perhaps I knew what they were but now I understand them.
 
Haha, well I'm glad I helped
@WheatWizard BTW, your current avatar is terrifying, as usual
@ConorO'Brien You know, {({}<>)<>}<> is arguably a valid brain-flak answer
 
9:47 PM
@cairdcoinheringaahing RIP all the effort we put in our controlled assessment, tasks were leaked online and Ofqual is planning to remove it from the overall mark.
"Exams regulator Ofqual plans to pull this chunk of the qualification from the overall marks as it has been seen by thousands of people."
 
@DJMcMayhem Thank you; I do my best.
 
@ConorO'Brien Does this count? APL (Dyalog), 9 bytes {⍞←⍺}/⍞,0 (prints to STDERR)
 
import math
def index(num, location):
    return (num // (10**location) % 10) # 0-based
def replace(num, location, digit)
    return ((num // 10 ** (location+1))*10+digit)*(10**location)+num%(10**location)
def f(n):
    l = int(math.log10(n))+1
    for i in range(l // 2):
        n = replace(replace(n,i,index(n,l-i),l-i,index(n,i))

    return n
True constant memory, happens in place
Works for arbitrary size integers
Not golfed but it's a PoC
 
Given a line of space-separated integers, is there a short ruby method for finding the product of that list?
 
Why the input method as a string?
 

« first day (2494 days earlier)      last day (2342 days later) »