« first day (2707 days earlier)      last day (2129 days later) » 
01:00 - 22:0022:00 - 00:00

1:38 AM
CodeMirror considers PHP to be 'clike' but not JavaScript :/
 
 
2 hours later…
3:17 AM
remmber when i said webaudio was Goood... guess what it;s bad again
now that im doibg something that isnt plugging together three builtin nodes
it makes no guarantees as to sample rate... i have to resample in software and pray i dont mis an animation frame....
it blows, its just terrible
and you all know im the authority when it comes to something being bad... you know this..,
 
@cairdcoinheringaahing Done. There were some things I wanted to do anyway.
 
Is there any language that comes to mind in which @@ is a special keyword/thing
 
3:41 AM
@Downgoat Mathematica
 
4:18 AM
Anyone know of a system where a newline is 2 consecutive equal byte?
(Windows is 0D0A, not count)
 
4:31 AM
@user202729 Markdown
 
Good idea...
but * in markdown is list element.
Done it anyway.
 
5:06 AM
is there any language in which embedding the input would not be possible for a given answer
 
@Downgoat ?
What does that mean
 
e.g. in python for lambda x:" ".split(input()) i could do lambda x: " ".split("the input text")
 
In what language would that work
What if the input contains quotes
 
you can escape them
newlines is a better example
 
@Downgoat Maybe sed?
 
5:19 AM
@Pavel But who sed that tho'?
 
IDK, I don't actually know enough sed
 
0
A: Sandbox for Proposed Challenges

xnorEarth mover's distance code-golf Dupe check: Do we already have a challenge to compute the Earth mover's distance between two 1D arrays?

 
6:05 AM
@Downgoat If the input is an array, Add++
 
6:30 AM
@mınxomaτ Confirmed :p
 
7:04 AM
-1
Q: Interesting coding

B. KnappSo, I'm learning C++ programming language, and I'm on "for loops". I've been experimenting with it a little bit, and I found that something weird (to me) happens when you put a value above 5 on my code. #include <iostream> #include <vector> #include <string> #include <fstream> using namespace ...

 
posted a new challenge to sandbox codegolf.meta.stackexchange.com/questions/2140/…
any comments gratefully received
 
7:24 AM
0
A: Sandbox for Proposed Challenges

AnushA fast algorithm for the spike distance Given two sorted arrays of floating point numbers X and Y, we can define the spike distance as follows. The distance is defined as the minimum cost associated with the transformation of one point pattern X into a pattern Y by deleting, adding, and moving p...

 
 
2 hours later…
9:16 AM
0
Q: If you don't log in with FaceBook, Google, or the new SE-only account, act now!

Peter TaylorIt hasn't been very well publicised, but StackExchange is going to drop support for most third party logins in less than a month from today. This could affect a lot of the old-timers, because back when we joined the site it didn't have its own login accounts. Therefore this is a public service an...

 
0
Q: Draw a distribution diagram!

Ivan TalanovChallenge You are given an array of three integers as input - [min_value, max_value, sample_size]. Your task is to output a diagram in any convenient format, that shows the frequency of numbers in the min_value-max_value range showing up when a random number is picked sample_size times. With lar...

 
9:39 AM
> AssertionError: expected 0 to be at least 10
2
oh no
 
 
3 hours later…
12:24 PM
Wow, that question was blasted into oblivion (the s/mime email challenge). Just because a challenge requires some domain-specific knowledge doesn't mean it's a poor challenge. Not every challenge needs to be accessible to every person.
 
underspecified...
 
It's only "underspecified" if you lack that domain-specific knowledge.
I find it perfectly understandable, but then again, I'm an email administrator so this is literally within my job description.
 
1:13 PM
0
A: Sandbox for Proposed Challenges

Luis felipe De jesus MunozStrongest Sentence Given two sentences, calculate and output the strongest one. The strongest sentence is the one which the sum of all it characters decimal codes is the biggest. Example given Hello World and Hi Maria Input => Hello World H e l l o W o ...

 
@AdmBorkBork I tried just getting an example or a quoted part of the RFC so that the question was self-contained but the author responded that it was easy and you shouldn't need the RFC and then he proceeded to remove the RFC link from the challenge...
I don't think what I was asking was unreasonable and I think it would have made peopel chill
that being said I didn't downvote or vote to close
there's a reason i haven't posted this challenge yet
It will certainly be received poorly until I take the time to explain the main parts of the ttf format
 
1:51 PM
I think it is just as specified as codegolf.stackexchange.com/q/76379/42963 (Simple TCP Server), which is one of the more-highly-regarded challenges on the site.
 
I agree with Adm
 
@AdmBorkBork I think this is a perfect example of a certain type of bogus question: the one which can only be answered by using by using a library or builtin to do most of the work.
It says "do X" without really defining what exactly is needed to do qualify as doing X.
Thus the only way to be sure is to pick a library advertised as doing X that everyone can agree does X.
There's no scope for shaving it down to the absolutely minimal functionality because the boundary is not clearly drawn.
 
Yeah, I'm not a fan of questions that ask you to conform to a enormous protocol are basically "Does you langauge have a library for X, if so then call the function that does this task".
 
and what counts as "pick a library advertised as doing X that everyone can agree does X"?
 
@cairdcoinheringaahing re (But forcing a different result in a tie makes my example program increase to 12 bytes): lol, my approach is 5 bytes and outputs both in case of a tie... But I guess you use some other quick like Þ for example :P
 
2:06 PM
@NewSandboxedPosts really surprised to see this isn't a dupe
 
When did you change your avatar to a Pokemon trainer trying to go Super Saiyan?
 
This challenge should be downvoted and closed for the same reasons as listed above, yet was upvoted instead.
 
@AdmBorkBork Eh? The poor question with TCP had +95.
This Ethereum one at least looks like the author put a lot of work into trying to precisely specify it.
 
I'm not a huge fan of that one either (the ethereum), but at least it has a reference implementation and seems like it would be easier to verify the task is done successfully. I honestly don't know how to verify an email was signed properly by looking at the code at all, and I'm not positive the OP does either, or the question on SO wouldn't still be open.
It also doesn't help that AFAICT, the OP's only responses to questions have boiled down to "this is easy".
 
Well, the author says it's "something as trivial as downloading a web page over HTTPS" which I would understand to mean that it's very complex.
 
2:16 PM
the author might mean that "it's as trivial as using __import__('urllib').request [or Jelly's ŒG]", that's not how trivial is defined :P
(note: using urllib.request has always been a hell for me)
 
On an unrelated note, SO's new right-nav menu looks terrible.
 
@Geobits Left-nav?
 
Ah crap. That's what I get for looking at my screen upside down >_>
Yes. That
I mean, umm, no, I meant the right-nav. It's all squoze up now that there's a left-nav (which also happens to look bad).
 
2:39 PM
The ethereum challenge and tcp challenge both have test cases and the ethereum one even has a reference implementation. that's literally all i was asking for in the email challenge
OP couldn't write an implementation himself but was touting that it was easy
@feersum It most definitely can be answered without a library or built-in. How else do you think that library or built-in was created? Just because it's not a 5 byte solution doesn't mean the question is bogus
I'd like to see more practical challenges like that rather than "implement this oeis sequence"
it actually forces answerers to think about the problem and opens a lot of opportunities for golfing instead of just using some golfing language with 1000 built-ins
 
@Poke brb, making Mthmtca
@Geobits Ah, I see what you mean.
 
2:59 PM
@Poke How? I think it's quite the opposite.
For "implement the oeis" you have to think. for that one you have to use that exact builtin.
Although yes, that kind of challenge offers a different "challenge" than the "implement the oeis" ones.
 
@user202729 one of my previous points was that there may not be an exact built-in
if might actually be a challenge instead of just a contest
 
@Poke Or a sequence of the exact builtin.
 
sure
 
@Poke (assume I understand exactly what you mean) a challenge is just a contest, but harder.
 
that's more or less what i'm saying i gues
like maybe i write a tcp server by using winsock.dll directly or something wild like that
instead of a native socket library in the language i'm using
and sure you might need another library to invoke the dll
but it's that type of thought process that an oeis challenge lacks
many oeis sequences even have an algorithm right on the page already
 
3:06 PM
And yes, I agree that those challenges should not be closed.
Anyone want to make a meta post?
"Please do not close challenges which use technologies you can't understand but with links to complete specification"
or something like that?
 
That sounds fairly condescending worded that way ;)
 
If a challenge (or even just a question on any SE) is put on hold, can we disable voting?
 
@Geobits ?
@Poke Just delete it.
@Geobits "condescending"? How?
 
@user202729 sometimes the author isn't around to do that, though
> you can't understand
maybe rephrase
"you are unfamiliar with"
 
@Poke I think there has been a case where I've upvoted a challenge I've VTC'd (as dupe IIRC, before getting the hammer)
 
3:08 PM
@Poke Sounds bad this way .....
 
@Poke Did you read the next 3 messages after that one you replied to?
 
@Poke ... not a bad idea?
 
@feersum uh... no. sorry about that
i agree with your points
 
Is it considered bad practice in Java or C# to define an enum type with two or three members just for use as options to a single method?
I know it's a common practice in F# but idk
 
@NewSandboxedPosts And nobody says "
Apr 3 at 16:19, by caird coinheringaahing
@NewSandboxedPosts ಠ_ಠ That's just lazy
Obviously it's not a problem. Some users (possibly zero, so this is offensive to nobody) just hate some user group/rep group.
About the challenge... I realize that it has nothing to do with a sphere.
 
3:20 PM
considering implementing a Follower Leaderboard to rank the loyalty of my followers....
 
@quartata How would you measure that
 
would this motivate you [Y/y]
 
@user202729 nah, I don't think this applies here
 
@Pavel how often they like my shit
truly loyal followers like my stuff even when it's bad
that's how it worked in medieval europe...
 
@EriktheOutgolfer I don't really want to discuss that, I'm not the downvoter anyway
 
3:23 PM
@quartata Well it seems to be working
 
ok clearly you need some more time on the hypnopaedia
 
*thinking*System.Process.Start is cross-platform and throws Win32Exception on failure
 
(random note. I hope that closed questions are immediately VTD-able)
 
@user202729 Oh god no. I'd never get another solution to anything I ask on SO ever again.
All my questions are closed immediatly but someone generally writes out a super detailed answer in comments
 
@Pavel ......
...?
How?
How are the questions closed? Too broad? Something else?
 
3:27 PM
Too broad a lot of the time, yeah
I mean, how often have you asked a well-recieved question on SO?
It almost never happens
 
@Pavel Hm...
I have almost nothing to ask.
Mods: For challenges with a lot of obsolete commands, should I flag them all or do something else?
 
I have to use Windows for work again and I think I've determined that the thing I miss most is the fact that I can't summon fancy progress bars with dnf upgrade
Getting a list of all my outdated software about to become new is like the most satisfying thing ever
 
chocolatey?
 
It's not the same
It never gives you several hundred TeX packages to updated
Speaking of which, Chocolatey's python3 became Python37 just now
 
@Pavel even more satisfying than running sudo apt autoremove --purge? ;)
 
3:37 PM
Same sort of thing
@quartata Also, chocolatey packages don't have progress bars when you download them
 
doesnt powershell have a builtin for that
 
Write-Progress
 
No one seems to use it for Chocolatey packages though
 
But that a) assumes the program is running interactively, and b) assumes that the author of the package implemented it.
 
3:59 PM
@Pavel You and I have very different experiences of SO.
That seems completely alien from my experience
@JoKing Hey I was thinking about making a Lost answer to this, but I'm hitting some conceptual walls. Since your the lost expert and seem to always be up for a tough challenge, I was wondering if you wanted to collaborate some time.
 
thinks about it... ouch
 
4:13 PM
I can never remember my trig, atan is the opposite of tan right? IE atan(tan(a)) == a?
 
if the a stands for arc then yes
 
the a in atan? it does
 
CMC: arctan(a)
 
That's either a builtin or not a CMC
 
@Adám Mathematica, 6 bytes: ArcTan
 
4:21 PM
js, 9 bytes: Math.atan
C#, 16 bytes: System.Math.Atan
 
Alternatively, Solve[Tan@x==#,x]&
 
@Adám APL, ¯3○⊢ (side note: I'm liking the extended tool tips for the glyphs in the new ride)
 
Oh yeah, I still haven't applied for a dyalog license for my new laptop
 
I don't think you re-apply for the same OS
 
A) You do B) this is Windows
 
4:29 PM
I've had an experience with A) once, and I got my application rejected as a dupe
 
You must have done it really quickly after the previous time
The license is still valid for a couple weeks
 
I eventually followed different steps to solve my issue (re-enabling MyDyalog)
yes, I meant after 2 weeks
also, the license doesn't expire, it's MyDyalog that does
 
@Cowsquack Yeah, as promised, we gave all the tool tips a complete overhaul in 17.0, and RIDE 4.1's language bar is built from the interpreter source.
 
Well, I've had to reapply every time
 
I wish I don't have Dyalog installed illegally right now...
 
4:32 PM
I end up reapplying a lot because I do stupid shit with my computers and have to reinstall the OS all the time
It always takes me a couple tried to fill out the form
 
lol VirtualBox/VMware Workstation Player exist
 
@Pavel Just email Karen instead (keep your license number somewhere outside your computer).
@EriktheOutgolfer What?
 
@Adám like, I wish Pavel doesn't know of some new license terms that suddenly make my install of Dyalog illegal, but I don't think that's the case
 
Oof, company policy prevents me from installing Dyalog IME
That's dumb
 
also, I really do want to renew MyDyalog, but I'd rather be a little bit patient until 17.0 comes :P
 
4:35 PM
@EriktheOutgolfer Oh. No, don't worry. Your license for whichever version you've got is forever.
 
Although in fairness, I should be working right now and not installing APL
 
@Pavel Can you install my keyboard layout instead?
 
@Adám Probably not, it doesn't seem to be letting me install layouts at all
I can install the official MS ones but that's it
 
@Pavel OK, well, the IME is a real program, my layout is "just another layout".
 
4:37 PM
I'll try
Though probably not right now
 
@Pavel Hm, we should get MS to include APL layouts out of the box, just like Linux has it.
 
Hmm, I wonder how well doing my assignment in Dyalog would go over
 
@Pavel Are you allowed to choose language?
 
you'll most likely get sent to a psychiatric institute :P
 
Yes, but now that I think about it it needs to be .NET and crossplatform
 
4:38 PM
> .NET and crossplatform
 
So I'm limited to C#, F#, Q#, and VB.NET
 
@Pavel Could you install RIDE? You can always use this in the browser.
 
Ooh, yeah, that'd work
 
@Pavel Dyalog is .NET and crossplatform (but not crossplatform .NET).
 
definitely do it in VBA
 
4:39 PM
VBA isn't crossplatform
 
like, MS Office?
 
It's not just office, it's a bunch of other things, but yeah
@Adám .NET Core when :3
 
4:54 PM
._. everywhere I've worked (this has happened 3 times now) I've been given admin access to critical repos
Why would you give a temporary intern admin access to anything
 
Wait ... ngn is watching my RAD repository. OwO
 
#!/usr/bin/env bash

# build.sh

echo 'Visual Studio is required to build this project.'
exit 1
k
 
5:20 PM
._.
 
@Zacharý remove cumulative reduce and wait for the reaction :p
 
@EriktheOutgolfer No! Now that I actually have a working computer, it's time to work on the difficult parts of RAD.
 
don't worry, you don't have to remove it entirely, just from the repo ;)
 
@LeakyNun @flawr Just in case you have some time to take a look
 
No! I actually have a decent language for once.
 
5:34 PM
@Mr.Xcoder Sorry, I was editing at the same time!
 
Np, your edit is more relevant anyway
 
@Pavel I didn't realize people would try to game the Follower Leaderboard
serial voting...
i know this to be the case because no person born after the 60s would likr my "leave it to beaver" subtweet... it was actually a honeypot
 
I'll be completely honest I haven't read them all yet
 
one of the more revolutionary features of my incoming wsbsite, which is being lovingly crafted in hand made C by Oompa Loompas in the tarpits of hell.., is that there is a "random tweet' button which can be used as a more consumer-friendly way to enjoy my tweets...
 
@LuisMendo You wrote "f(0)=1 implies", but you first said "f(0)=0"
(I assume you meant the latter.)
I think you also assume that the derivative exists everywhere.
@Mego Oh cool, it seems I haven't been in TNB in a while :/
 
5:51 PM
@Pavel Why don't you bug support@?
 
@LuisMendo Just to clarify: Does the power series converge in the whole domain (0,1)?
 
@Pavel When are you going to do an internship at Dyalog? We don't give interns admin access to anything.
 
@Adám I get extremely paranoid whenever I try to write an email fearing I'll make any kind of mistake and then I find myself unable to actually send the message
Emails are scary
@Adám Do you have high school interns?
 
@Pavel Then maybe bug MortenKromberg in the orchard.
@Pavel Actual level of education shouldn't matter. It just has to be legal. How old are you?
 
By the time I finish my current internship, I'll be 17.
 
5:57 PM
@Pavel When will you finish? And if you can be an intern now, why shouldn't you be able to be at Dyalog?
 
@Adám End of this summer. I assume there aren't offices in the Seattle area, so by then School will start, so I'd have to work remotely.
I just don't htink it'd work out.
 
@Adám Morten Kromberg the Unpingable
 
hi... could anyone give me a hand with codegolf.meta.stackexchange.com/a/16580/9207 please
 
@Pavel Right, I was thinking next summer. We have had an intern by my colleague in Vancouver, but I'm sure we could fly you in.
 
it wasn't loved at all when it was on main
 
6:10 PM
@LuisMendo I feel like right now there are not enough constraints, because the funciton only seems to be analytic in one point (see my question above). If the power series converges in the whole domain, I think you can just directly compute the limit of xf'(x)/f(x) by term wise differentiation and maybe applying some de l'hôpital or so
 
It seems unambiguous to me but apparently it is confusing
 
@Adám Cool. Well, we'll see.
 
6:22 PM
@flawr Yes. Also, that implies the same for the derivative
 
Then I really think you can just directly compute the quotient, let me try.
 
@flawr I think I tried that but all derivatives become infinity at 0
 
hmm...
 
I wish I could set multi-word aliases so I could do e.g. alias "pip install"="pip install -U"
 
I did something similar but for cal, I wanted cal with no arguments to display the calendar but with the current date highlighted
 
6:31 PM
hi @orlp
 
so I just aliased cal to bash $HOME/q.bash where q.bash is Try it online!
 
I didn't quite understand your comment to my cs.SE question
 
@Cowsquack I'm a try that once my WSL finsishes updating
I have no idea what Failed to connect to bus: No such file or directory implies for said update but it's getting printed a lot
 
bus wat
 
¯\_(ツ)_/¯
 
6:39 PM
@LuisMendo sorry, I just did some computations and tried reformulating the problem, but right now I'm struggling with taking the derivative of polynomials^^
 
if anyone could help with my sandbox question it would be awesome.. please :)
 
after realising my cal was bsd, I installed gcal from brew, turns out gcal already highlights the date :/
 
Lol
 
@Pavel You missed your bus and now you need to walk.
3
 
how do tio permalinks work, what’s it called?
looks like base64 with @ for pluses
but is it?
 
6:49 PM
@FrownyFrog You could ask Dennis

 talk.tryitonline.net

For general discussion and feature requests regarding tryitonl...
 
@DJMcMayhem um, I don't think it's easy to do so, IIRC I did it once and the answer was along the lines of "it's complicated, it also contains bzip"
 
Well, it's easy to ask, not necessarily easy to get an answer :P
 
you could just go look at the source code if you wanted
 
um, there's probably a reason it's not easy to get an answer
(on the other hand, Dennis also has a life)
 
@FrownyFrog Check out the TryItOnquine challenge.
 
6:53 PM
wow
thanks
 
@flawr Well, thanks for your time anyway!
 
It’s base64 with @ for pluses, plus compression
 
hmm... I will leave it an hour before begging again :)
must be no algortihms interested people in
 
> As nouns the difference between pluses and plusses is that pluses is while plusses is. -- Source
 
7:09 PM
@AdmBorkBork that is why I like minusses
 
eyes narrow I see what you did there...
 
Just thought I could add something to the converstion =P
 
the puns are multiplying
 
7:25 PM
exponentially
 
This is an amazing answer over on Physics
 
7:58 PM
last try for today.. could anyone take a look at codegolf.meta.stackexchange.com/a/16580/9207 please and let me know what I can do to make it acceptable? It was disliked when it was on main
I personally think it's an interesting algorithms question
 
You should clarify: What is a point pattern? Does it matter that the array is of floating point numbers?
Can the array have repeated entries? (Just asking because you're talking about subsets.)
And how do you measure the speed of the algorithm?
 
@flawr thanks! Let me clarify now
 
Also what is dx? (I allowed myself to fix some typos/make it slightly more readable.)
Maybe you should also add a more complicated example
 
@flawr how about now?
 
@AdmBorkBork "matter tells space how to bend and space tells matter how to move." almost the first half of an H2G2 quote (actual quote uses curve)
 
8:10 PM
@flawr I think I addressed all those concerns
thank you for the typo fixes. You may have introduced a new typo in the sum but I fixed that
 
Looks better now.
 
thanks!
 
@Neil Could be a reference/misquote
 
I'm still not convinced that there is a natural measure of complexity.
 
@flawr Wouldn't it just be O(nm) for example?
 
8:12 PM
What is n and m here?
 
n is the number of points in X and m is the number of points in Y
 
well that is an important point
because you cannot really compare an O(n*m) algorithm with one that has O(|p_d|)
 
isn't that in "The input lengths of the two arrays will be n and m. Your code would take X and Y and the three parameters as input."
?
oh I see. I should say the complexity should be in terms of n and m?
 
Right.
 
done
 
8:15 PM
But then again, we have two variables we have to consider the behaviour as they grow
 
can you explain that a little more?
 
lets say we have two algorithms, one is O(n^2 + sqrt(m)) and one O(n + m)
which one is better/faster?
 
got you.. I will eliminate m :)
it has gone
 
maybe you should require the participants to express the complexity in terms of N = max(n,m) or something like that
 
@flawr I think I fixed it
I just said the sum of the lengths is n
it's better with one variable as you say
 
8:18 PM
Right that is equivalent
 
yes.. but slightly nicer :)
this has been really helpful, thank you so much
now.. any idea how to solve it? :)
 
that is a matter of taste :P
well I haven't thought about it yet
 
I will post it soon.. so maybe don't tell anyone soon
@flawr yes :)
 
but it might be a difficult one
 
good
I would hate to post anything trivial
the linked question has an O(n^2) solution it seems if you can hash in constant time
 
8:20 PM
I suggest letting it sit in the sandbox for a few more days.
 
:(
 
It is a sorted array of integers right?
 
yes
 
Do you mean integers, or maybe just bounded ints?
 
hmm
 
8:21 PM
(that can make a huge difference)
 
I don't really want an algorithm that iterates over all ints and then says that is constant time
 
@EriktheOutgolfer I hope so but that answer doesn't cover the situation we are in
which is to do with computational complexity
 
I think that loophole is general, not just for
 
but it doesn't say anything about banning algorithms that iterate over all ints that in range
 
8:24 PM
@Anush so bucket sort is using a loophole?
 
it would be to claim en.wikipedia.org/wiki/Counting_sort is linear time
at least as far as my style of question is concerned
 
Well you just have to be clear about those things.
and just banning "iterating over the rang of ints" is maybe not a good idea either
 
the point being that properly we should be measuring running time as a function of the length of the input in bits. But because that's really fussy we ignore the number of bits in an integers. But this slackness leaves you open to abuse if you are not careful
so an algorithm that run in O(p_d) time should properly be regarded as exponential time
as it is exponential in the number of bits needed to represent the integer p_d
@flawr could you explain your concern about using the word "integers" in the question?
 
@AdmBorkBork I would comment but I don't even have an account on that site
 
@flawr I added a line about the complexity to the question which I hope is helpful
 
8:41 PM
Now I think all algorithms that include the size of p_d are considered s O(exp(n)) too?
I think you have to ask someone who actually knows what they are talking about, I don't think I can come up with a good solution
 
8:53 PM
@flawr OK thanks. I feel I know what I am talking about on this topic :)
but maybe it's just simpler without the extra note
 
9:24 PM
@Neil just make one?
 
01:00 - 22:0022:00 - 00:00

« first day (2707 days earlier)      last day (2129 days later) »