« first day (2806 days earlier)      last day (782 days later) » 

5:23 AM
@Scrooble welcome to hell
5:51 AM
@quartata do you have suggestion on how to do spam classification for Axtell besides training a crap RNN
6 hours later…
11:44 AM
A: Sandbox for Proposed Challenges

maxbBe there, for the square Sometimes when you're lying in bed and reading a message, your phone screen will pop into landscape mode right in the middle of a sentence. Only being able to read left to right, you find yourself incapacitated, unable to process the text in front of you. To ensure that...

12:42 PM
I have a stupid question y'all. My prof just breezed through an argument and I'm lost af
We're working in Haskell, and he said the code (1::Integer)+2*(3::Double) causes a type error, which can be revealed through trivial analysis of (+) or (*).`
I don't get this
@RushabhMehta You can't add Integer and Double.
that part I got
but what is the second way
like what analysis of (*) tells you there's a type error
(*) alone doesnt cause issues
Exactly what I thought!
But clearly my prof thinks otherwise
I asked him about this, he was like "that'll be some good thinking for you"
@RushabhMehta maybe he meant "both"?
because you have to analyze the "*" to figure out the return type
12:48 PM
Here's the exact wording
Consider the expression (1 :: Integer) + 2 * (3 :: Double). This will cause a type error, because the two different type constraints are inconsistent. One bit of trickiness is that the inconsistency could be revealed either in the analysis of (+) or (*). Provide both analyses.
I suppose the "analysis" here is by hand.
oh, I think he's saying "It could be in the (+) or it could be in the (*). Analyze both to identify which one it is"?
"either" could mean "xor" or it could mean "or"
Huh. That would be a really longwinded way of saying something a lot more simple
12:50 PM
yeah, possibly? That's the only way to make the question make sense
I suppose it's like this
1. Consider the (+). --> (1::Integer) must have the same type as (2*(3::Double)) --> 2*(3::Double) has the type Integer. (etc)
2. Consider the (*). --> (2*(3::Double)) must have the same type as (3::Double) --> 2*(3::Double) has the type Double. (etc)
Either way, I think it's simple, don't overthink it.
@user202729 @NathanMerrill I've literally been stuck for 3 hours on this stupid shit
but thank you!
1:34 PM
Unpopular Topic Time -- Is PPCG dying? 19th Byte seems to have intermittent chat rather than the near-constant stream it used to have. We're getting maybe one or two challenges per day rather than the 5/day we used to have. Prominent figures are leaving or have left without a word. Even the Sandbox seems to be more sluggish than usual.
The chat sure have seemed a lot more dead than usual recently. Answers to challenges seem to more in line with what it's usually been. Challenge frequence seem to have gone down some yeah, but in fairness, the more challenges we have the harder it is to come up new unique ones
I am a contributor to all 3 of these issues as well
From what I've seen, the nature of the challenges have changed greatly over the past few years. It seems like people want to post harder and harder challenges, and that people avoid posting to avoid having questions marked as duplicates.
As a user who only started recently actually using this site - it seems fairly active to me, in everything except maybe new questions posted, which can be attributed to reasons above. But answers are plentiful
Also, the more challenges are posted, the bigger the chance for dupes, I guess?
Also regarding the chat: I had been visiting this site for over 6 months before I even heard about this chat. I'd say that it needs some advertising to attract new visitors, and maybe some examples of things that are on-topic for the chat?
1:45 PM
So people just think "oh this is a good challenge" and bam, dupe.
I don't feel as if the difficulty of challenges have increased in any meaingful way. There is still a healthy mix there I feel, with a bias towards easier challenges since those are easier to write and tend to get more answers
@J.Sallé Of course, but you can always put a twist on it to make it unique enough. I posted this challenge, which was almost a duplicate of this one before I put a twist on it
I tend to agree with Emigna -- I don't think the difficulty ratio of hard to easy challenges has changed, I just think there are fewer challenges overall.
Writing a good challenge isn't easy. And making one that isn't a duplicate or close to one is not trivial with the amount of challenges we have in total
At least that's my impression. I know there are people here who make it seem easy
^ agreed.
1:53 PM
Another thing could be that people end up here because they like coding and golfing, not because they like creating challenges. I agree that it's much more work to create a challenge compared to coming up with a solution (especially since you can get inspiration from other answers). But it's up to all of us to post challenges, and to leave feedback in the sandbox.
I try to have one challenge in progress most of the time, but the sandbox doesn't get the attention that it needs most of the time.
Personally, I think we could start "reviving" select challenges (in a language-of-the-month fashion or some such thing). People could suggest old(er) challenges they find interesting with a low number of answers and bump them (maybe after some editing so the challenge is up to current standards).
@J.Sallé something like this?
(up to 1 answer, at least 60 days old, no accepted answer, open, unlocked)
@EriktheOutgolfer sounds good. I think I might make a meta post asking for opinions on that
@J.Sallé and another approach could be answers:..4 created:..60d closed:0 locked:0 (ignoring whether an answer has been accepted, as well as upping the bar to 4)
2:09 PM
Yep, I'll make a meta-post on that so everyone can chip in on what would be the best parameters
that can't really be done with just a meta post...
nobody wants to dig through the comments just to get an average
Well, I meant presenting the idea properly to the community to see if people would, first of all, be on board with it
the parameters and such could be decided after that
@J.Sallé I guess a CW-locked question would be fit
and the answer having a bunch of code snippets
those snippets will contain numbers, and each user should be able to add one number to every one, if they want to
(and not delete other numbers)
after enough are added for each, the average is taken (the question then is: to floor or to ceil?)
I'm making a meta-post presenting the idea. If people think it's a good idea, we could make that in another post?
also, you might want to consider the Community bumping factor
2:24 PM
What is that?
the Community bot bumps questions from time to time, although not all of them are what this is about, and not all such posts are ever bumped
Ah, I see. I'll mention that as well
I'm really sorry for being super help vampirey this morning, but I've been so frustrated. I'm getting an indentation error in ghci (haskell) on the line allp ::...
module Divisibility where
-- | This module is filled with functions aimed at finding divisble numbers.

-- | Return a sublist comprised of the elements of a list that satisfies a predicate.

filter :: (a -> Bool) -> [a] -> [a]
filter p [] = []
filter p (x:xs)
| p x = x : filter p xs
| otherwise = filter p xs

-- | divisibleBy checks that one number evenly divides the other through mods.

divisibleBy :: Num a => a->a->Bool
let divisibleBy d n = (n `mod` d == 0)

allp :: [Bool] -> Bool
let allp [] = True
@J.Sallé I like this idea. +1 from me once we figure out the parameters.
@Downgoat Use SmokeDetector's blacklist. It's not as SE-centric as you might think.
In fact, Medium used it successfully.
I think a system where users could see domains used in recently-deleted posts and then vote to blacklist could help to expand any blacklist also.
2:32 PM
Trivia question of the moment: What was Gumby's horse's name?
ding ding ding
have a star
I see what you did there
A: Sandbox for Proposed Challenges

AdmBorkBorkASCII Schedule / Calendar (title suggestions welcome) code-golf date ascii-art Given a list of activities and their start time/date, output an ASCII-art calendar showing the activities on the appropriate days. All activities are guaranteed to be in the same month, no two activities will be on ...

2:55 PM
@RushabhMehta let can't be used on the top level like that
@J.Sallé I've tried to do this
perhaps there was too much process with my method, but it failed pretty pooly
Q: Tour to Improve the Discoverability of Old Challenges

Nathan MerrillWe've got some pretty great challenges here. However, after a month or so, challenges get little to no activity. Since we aren't a Q&A site, we rarely get traffic via searches. I would love it if we could introduce users to high quality questions that they may not have found on their own. I b...

(I didn't end up getting enough challenges posted/voted on)
The reason I chose the one-time "Tour" method vs a "challenge of a month" or something is due to the duration required. We've had so many programs here on PPCG to promote making challenges/answers, and none of them have lasted a long time
but since my method involved editing the questions themselves, it would naturally lead to discovery as people came across these questions
Anyways, I'm not saying all of this to dissuade you, but rather to take into consideration how your method should work
3:18 PM
Turn ppcg into something like Project Euler where we have lists of known-good challenges
/me shrugs
Right, that sounds great: The process is the problem here
how do we create that list?
3:59 PM
Q: Supreme Sum String

GammaGamesSupreme Sum String Given an input string, return the word with the highest sum of each of its unicode characters. Rules The input should be seperated by whitespace The value of each word is based on the sum of each character in the word's UTF-16 code The output should be the first word with ...

4:10 PM
A: Sandbox for Proposed Challenges

nwellnhofLength of a UTF-8 byte sequence Determine the length of a UTF-8 byte sequence given its first byte. The following table shows which ranges map to each possible length: Range Length --------- ------ 0x00-0x7F 1 0xC2-0xDF 2 0xE0-0xEF 3 0xF0-0xF4 4 Notes: 0x80-0xBF are continua...

Q: Should we start "reviving" older challenges?

J. SalléRelated. Earlier today, AdmBorkBork posted this question on TNB, asking whether or not PPCG was dying. In the following discussion, I have suggested we, as a community, could start "reviving" (for lack of a better term) select challenges, based on suggestions given by us. The reasoning behind i...

^ Sniping New Meta Posts because I don't remember if that still exists.
yes, it does exist
4:28 PM
Q: Should we start "reviving" older challenges?

J. SalléRelated. Earlier today, AdmBorkBork posted this question on TNB, asking whether or not PPCG was dying. In the following discussion, I have suggested we, as a community, could start "reviving" (for lack of a better term) select challenges, based on suggestions given by us. The reasoning behind i...

1 hour later…
5:31 PM
A: Let's allow newer languages/versions for older challenges

MD XFTrust the community and add a standard loophole I think the community can handle the add-a-builtin problem (I may be very wrong). I think that if someone adds a feature to their language that is clearly meant to only solve one particular challenge, the community will probably downvote that answe...

Q: Draining pools against an opponent

FrostbyteWelcome to the Winter Pools challenge! Game In this challenge, you have to write a Python script in which you compete with your opponent to drain pools. There are three pools- yours, your opponent, and a neutral one. You take it in turns with your opponent to drain water from one pool and put ...

@dzaima I tried searching for like 15 keywords and none of them are in the title D: thanks
5:51 PM
@AdmBorkBork I think previously a lot of the 'constant stream' used to be noise and as we cleaned up on that it's just that now we're seeing more signal over noise
quality over quantity kinda thing
@quartata looking at the code for spam detection (findspam.py) it seems to be pretty hardcoded for SE if I'm not mistaken?
@Downgoat blacklisted_websites.txt. It's a list of regexes.
@Downgoat That's a fair point, and likely a contributor as well. Thanks for bringing that up.
what should happen in case of spam? Should it just be rejected from being posted?
not sure how SE handles it
6:07 PM
I thought they just let it all get posted and then let charcoal take care of things
yeah, that's what charcoal does but not sure if for Axtell we should automate any part of it
7:02 PM
So both Charcoals are in one place now...
Q: Stagger, stack, sum

Luis MendoInspired by this Stack Overflow question. The challenge Input An array of square matrices containing non-negative integers. Output A square matrix built from the input matrices as follows. Let \$N \times N\$ be the size of each input matrix, and \$P\$ the number of input matrices. For clar...

7:30 PM
How long is your MATL solution for this? — Giuseppe 5 mins ago
ah, when you start associating users with languages...
@AdmBorkBork Maybe. It's hard to tell.
I've done a couple of queries
Our monthly post count is going down
However, the average number of views those posts are getting is about the same
@NathanMerrill Oh, that's an interesting metric I hadn't considered.
@AdmBorkBork I was originally going for traffic count, but stackexchange only shows you the current numbers, not the history
This is a cheap substitute, as it doesn't include traffic on old (before the current month) questions, and doesn't include traffic not on questions
it also overestimates older questions, as they accumulate view count over time
8:00 PM
Quick question: Are "other" python libraries(such as numpy) allowed in codegolf as long as we import them, or no?
8:22 PM
@Quintec yes, since you count the import, and they define a "Python + numpy" language, for example
Got it
One weird rule about libraries is that (at least so far right now) vim libraries plugins carry no byte penalty.
I mean, what if I created a new language named PyNumPy
I think so far with our rules, you totally could. But it would be really cheap and not very popular
8:26 PM
Interpreter: exec(“from numpy import *;” +input())
(^^ number of messages per week, thanks to tnbde)
@Quintec we encourage intra-language competition, that is, competition among answers in the same "language", as defined by PPCG's rules; "cheating" with an external (non-built-in) library isn't going to work there
(and, well, it isn't really that fun)
also, your "reference interpreter" won't really work :-D
from numpy import*\n then :P
8:46 PM
the thing is, you shouldn't be altering the file itself
I’m going way too far for a joke now
Nah too lazy to write out dir(numpy) etc
Q: The max() is not enough

ElectricWarrIntro A friend posed this question today in a slightly different way - "Can a single [Python] command determine the largest of some integers AND that they aren't equal?". While we didn't find a way to do this within reasonable definitions of "a single command", I thought it might be a fun probl...

9:12 PM
@NewMainPosts 32 bytes lambda l:sorted(set(l))[1:][-1:] if IO was more lenient
yeah, outputting singletons is allowed by default...
Yeah but it breaks several rules of the challenge even though effectively doing the task
@DJMcMayhem Golfable in Py3
Ooooh duh {*l}?
9:17 PM
Or even [*{*l}][1:][-1:]
Yeah I was looking for the set[1:] thing but realizing it’s hard to type brackets and asterisks and whatnot on a phone. :P
Psssh, I write V and jelly from my phone all the time :P
@Quintec Also, as it stands your answer is currently invalid
Bah, I was going to write apl, then I realied I couldn’t
Need a AltGrrrrrrr key
@DJMcMayhem thanks
Don't you mean max(eval(x))? :P
I'm just deleting this mess
9:53 PM
@DJMcMayhem No. SpamRam blocks most of all the spam, actually.
just by simple IP blocking
10:14 PM
@quartata what is a SpamRam
10:50 PM
@Downgoat Like a goat but angrier
SpamRam is SE's first line of defense against spam. The exact details of how it blocks spam aren't public knowledge, but IP blocking is a factor. Just by blocking IPs that have posted spam that got red-flagged, a lot of spam never makes it onto SE.

« first day (2806 days earlier)      last day (782 days later) »