« first day (4111 days earlier)      last day (108 days later) » 
00:00 - 15:0015:00 - 00:00

3:26 PM
@user I've just realized--QBasic is a counterexample. It is statically typed, and your standard Hello World program is simply PRINT "Hello, World!" (or ?"Hello, World! if you're feeling golfy).
Not that I would recommend developing software in QBasic. ;)
 
Got a 5 on my calculus mock AP exam :D
 
nice
 
And I just submitted my AP seminar essay, which I've been worried about finishing in time for a while
 
cool
 
(so worried I spent every night for the last week playing hypixel bedwars lol)
 
3:35 PM
off topic but we should play together sometime :P
 
We should, yeah. Ping me if you're ever able to at any point between like, 4:30 and 10:30 in my time zone (I forgot which way the timezones increment so converting it is left as an exercise for the reader :p)
 
I'm (almost) always in the cgcc gaming chatroom so I'll ping you there if I have the time (if you're unpingable I'll just abuse the reply trick lol)
 
I'm surprised that room's still unfrozen lol
 
it got frozen for like a month or two; I somewhat recently just unfroze it to dump a conversation about minecraft from here into there :P
 
4:04 PM
Well this is weird. I'm on Archive.org's QBasic, trying to get it to draw black pixels on a white background, and it won't draw anything. It's like it thinks color 0 is transparent instead of black. 0_o
 
4:17 PM
@DLosc Oh, TIL QBasic is statically typed
 
> Lisp stands for "List Processor,"
TIL
also hi
 
Oh, it even had an IR and IDE and everything
 
@user Yeah, it's easy to forget because you usually don't have to declare types for variables in simple programs if you use type sigils instead: s$ is a string variable, n# is double precision (and i! is single precision, but that's the default anyway so you'd just use i). However, you can also DIM x AS STRING and then x is a string variable.
 
Would DIM X AS STRING AS NUMBER (idk what the type for numbers is) error or try to parse it as a number?
 
Pretty sure that's a syntax error, let me check
 
4:28 PM
I mean if you cast a string that could be parsed as a number to a number
 
Oh, I see. Well, depends what you mean by "cast".
If you just try to add a string to a number, that's an error
If you use VAL(), that converts strings to numbers
 
S$ AS NUMBER, I suppose
@DLosc Ah, that's the one I was looking for
 
The DIM [var] AS [TYPE] is a variable declaration, like int x; in C. Doesn't work for converting stuff.
 
my school had qbasic in 5th grade (about 4-5 years ago)
 
oic, I thought DIM X AS STRING meant "get the length of X, then convert that to a string"
 
4:33 PM
@PyGamer0 Cool!
Then there's also the TYPE statement, which creates a user-defined type (basically a struct). I believe you have to use DIM if you want a variable of a user-defined type.
 
now the syllabus has changed sooo no more qbasic
 
F
 
so my brother didnt learn qbasic
they have scratch but they didnt learn that either because of the reduced syllabus because of covid...
we will probably have normal school starting from this year..
 
4:47 PM
Are we still deleting sandbox posts after they're posted to main? Because I'm seeing a lot of them that aren't deleted. Should I comment? Flag?
 
Most people don't delete them, but just edit them down to a stub (containing only a link to the post)
 
@DLosc I've seen people comment before
 
If people haven't done that, it's a good idea to edit it
but I don't think it necessary to delete them
especially since it doesn't make any difference to 10k+ users
 
Well, it does now
Since deleted answers only show up at the bottom
 
Oh yeah I forgot about that
 
4:50 PM
^^
 
Well I still like to keep my past sandbox posts around for people to go back and look at the feedback if they wanted
 
5:16 PM
0
A: Sandbox for Proposed Challenges

ophactWhat does the text talk about? code-challenge machine-learning string note that the machine-learning tag will be new META: this is far from being done. I also understand that this challenge is based mostly on manual opinion about the "type" of a piece of text. Hence, if you take issue with that, ...

 
This is going to be a major undertaking...
 
I think it certainly looks very interesting. Let us hope that there is no golfing language with a builtin!
 
Hopefully...
my greatest fear is that this is a duplicate because then I wasted a lot of time
not yet because it took me only half an hour to write the sandbox post but if I end up compiling 500 long articles and then I find a duplicate...
that would be a waste of time
 
I feel that. If you need helping writing some articles, I'd be happy to pitch in.
 
Machine learning is something that there isn't too much of here on CGSE
 
5:24 PM
if I find your answer great enough I could actually use it in my app
 
There is already a tag, btw
 
but neural network challenges are... rare
 
they aren't necessarily the same thing IIRC
 
which could probably be made a synonym of tbh
but not the other way round
 
they are not the same thing
according to google
please don't star that
 
5:25 PM
neural networks aren't really used for anything other than ML, AIUI?
but there's ML which doesn't use NN
 
oh, didn't notice "but not the other way round"
Apr 28 at 18:25, by pxeger
Whoever's starring random messages, please stop
> Starring things like [...] messages in ordinary discussions is an abuse of the starboard
 
Are you referring to "they are not the same thing"? I thought that was sufficiently informative to be worthy of a star. Sorry, I will try to update my starring heuristic.
 
@Pyautogui Yes
 
@Pyautogui It's not just about whether it's useful, more about whether it's sufficiently useful and/or funny without context to be worth putting on the starboard.
 
Got it. I will remember that and apply starring more carefully. Thanks for correcting me.
 
5:30 PM
It's useful to highlight there's a distinction, but without the context of the surrounding conversation it's not that useful
But don't worry too much we've done much worse
 
Wow, a full 129 stars. That is pretty incredible.
 
I wonder if it's the most starred message on all of SE chat
 
Fun fact: About 10% of people who have entered TNB have starred that message
@pxeger It almost certainly is
 
I wonder if it will remain so, and if so, for how long? I suppose without a concerted effort to beat it, it might remain the most starred for the duration of chat.SE's existance.
 
Most likely will remain so, yeah. A hundred people is more than most rooms will ever see.
 
5:34 PM
well, there's a world record for us to hold
 
Did TNB chat have that many active users in 2015 or did it somehow go viral?
 
And TNB's got a rare mix of off-topic fun that leads to that level of starring, and on-topic activity to keep it alive this long
 
speaking of off-topic, I wonder if we could persuade the devs to give us one of these
maybe data rain from the matrix?
 
Or compressing everyone's messages with Jelly...
 
I think that's a bit too obscure
but eh
 
5:36 PM
yeah, maybe
 
Or maybe an automatic rot13? For avoiding spoilers and adding some laughs?
 
I have lots of text files. Each one has three ints as the first line. I want to find the file which has the smallest first number in its first line. Looking for a nice golfed way!
This is in Linux
head -1 gets the first line of a file
 
@Ginger Persuade who? Chat doesn't really have devs, as evidenced by its lack of updates for the last decade
 
idk, Jon Skeet?
 
@graffe Seems like a job for zsh... @pxeger?
 
5:46 PM
@graffe sort -nk1 | head -1
 
Lol, nice
 
why star?
 
Typo, sorry. I am working on it. Really apologize for my tendency to star everything I can see. I might just not participate in chat for a while, unlearn the habit.
 
:|
@all thoughts on me making a meta post similar to this?
 
Imagine if there were a user named "all"
 
6:01 PM
there probably is
 
excuse me when did YouTube get a download button
 
@Ginger good idea perhaps
 
since like 2019 or something?
 
and why do I have to pay money to use it
 
@Ginger it's youtube. they intentionally make the worst decisions possible
 
6:02 PM
@Ginger that was made by Jeff Atwood, founder of SE
You're unlikely to get a response
 
@ophact Well, there is, allxy. The first three letters are all you need to ping someone.
You'd get pinged by @oph, I'm pinged by @rad, etc.
 
What happens if there are multiple users with the same first 3 letters?
 
They all get pinged
 
Come to think of it, what happens anyway if there are multiple users with the same username?
 
@
 
6:04 PM
How does chat decide which message it's in reply to?
 
The most recent one? Same as a normal ping
Oh wait...
Hmm
 
 
1 hour later…
7:32 PM
CMQ: Is UTF-32 a reasonable default for the string type in a language?
 
Yeah, nah
 
why not utf8
 
i hate javas use of utf16
(before java 17)
 
@RadvylfPrograms I like an approach similar to Python: Use UTF-8 unless the string has any non-ASCII characters, in which case switch to UTF-16, unless the string has any non-BMP characters, in which case switch to UTF-32
 
7:37 PM
why not utf8 everywhere tho
 
Massively saves memory bandwidth compared to pure UTF-32
 
its still the same amount of memory
 
But you get O(1) indexing
 
how
 
@Seggan No? UTF-8 is smaller for ASCII
 
7:38 PM
i meant for non-ascii text
 
Well, UTF-8 can still be smaller
 
well you reencode to UTF-32 if the string has any non-ascii characters added to it
 
"абв" is still the same size in all variants
 
@pxeger That seems like a bad idea performance-wise
@Seggan Not true
 
I like an approach where the language takes care of everything and I don't have to think about it. :P
 
7:40 PM
@RadvylfPrograms i meant in utf 8/16
 
It's 6 bytes in UTF-8, 6 in UTF-16, and 12 in UTF-32
 
@Seggan it's 6 bytes in UTF-8, 8 bytes in UTF-16, but 16 bytes in UTF-32
 
@DLosc Either that, or CP437 ;P
 
@DLosc in this case we are being the language
 
my point is you dont save memory using larger utf variants
so go with the smallest one
 
7:41 PM
@RadvylfPrograms well, if you use immutable strings, then any time you add characters to a string you have to copy anyway, so it's not a huge difference if you also reencode
 
@Seggan Saving memory is almost never the point of using UTF-32
 
@Seggan but you don't get O(1) indexing with UTF-8 unless you can guarantee the string is ASCII-only
 
how would it be O(1)
 
array[index] is O(1)
 
yes
 
7:43 PM
but in UTF-8, the characters are not always the same size, so you have to walk the whole string
 
@pxeger But whether they're immutable for the programmer or not won't necessarily mean it's not mutated under the hood for performance reasons. I'd have a hard time believing that if I slightly modify a string and then never reference the original again that it makes a copy of it.
 
but if it's ASCII-only, then the characters are always the same size, so you can just do array[index]
 
hmm
interesting
 
> if I slightly modify a string and then never reference the original again
 
Plus, UTF-32's much easier to work with since there's no validation or overlong encoding studd to worry about
 
7:44 PM
How do you check if you never reference the original again?
 
By being a compiler
var str = "Hi! My name is Ryan. [5 million more characters]";
var str2 = str;
str2[4010129] = "z"; // convert to american spelling

printf(str2);
 
this just in: radvylf is a compiler
 
@RadvylfPrograms Not true, because unicode characters can't be larger than 0x10FFFF
 
Okay, s/no validation/much easier validation
And I guess the UTF-16 pair thingies
 
surrogate pairs, yeah
@RadvylfPrograms Fair enough. I was still thinking we were talking about Python's model
(which might actually be able to detect if you never reference the original again, anyway, because it has runtime reference counting instead of a pure GC)
 
7:48 PM
Although idk, from what I've experienced using JS, what does and doesn't get optimized away has a lot to do with the phase of the moon, color of the table your laptop is sitting on, and background radiation on pluto
5
 
@pxeger i think its possible in tree gcs too, like java, just need to see if there is only 1 connection to the root
 
8:48 PM
I don't think it's possible in Java
If str2 holds the only reference to str, someone with a reference to str2 could still do black magic to get ahold of str at some point later
Although it's possible in a language with private variables to whom access is actually regulated at runtime
 
@user There are some fingerprinting techniques that still work, but the most powerful ones do not. And very few techniques work if the security slider is set to the highest.
 
@emanresuA I would think this is fake, although not sure what the motive is. No way Mac OS has such an ugly warning sign with that 2012 gradient.
 
@forest There's a security slider?
@RadvylfPrograms It doesn't look at all ugly to me, your yellowphobia may be influencing you :P
 
Any color of that would be hideous
 
@user Yes. There are three settings. The lowest enables JIT for performance and allows a number of potentially insecure but useful features. The medium setting disables JIT and disables certain fonts and makes media click-to-play. The highest completely disables JavaScript, external fonts, SVG images, etc.
 
8:55 PM
Ah
 
9:12 PM
jesus I leave for an hour and Radvylf transforms into a compiler
 
@RadvylfPrograms It also differs greatly based on the particular JS engine.
 
9:25 PM
I just pronounced vyxal "vucksel" send help before lyxal ends my existence with a particle cannon
 
I can send over some grenades
To blow yourself up before lyxal gets there
 
ಠ_ಠ
hey radvylf, want to see something blasphemous?
import logging
from sys import stdout
from multiprocessing import Process, Value
from flask import Flask, render_template, redirect, request, url_for, make_response
import stackexchange
from stackexchange.models import User
import time
from werkzeug.routing import BuildError
import json
import requests
import logging
import constants
import kothdb
look at all of those libraries
muahahaaaaa
 
That is normal Python
 
that's the joke
*subdued evil laughter*
does anyone else dislike python files without import statements at the top?
 
9:46 PM
i dislike the ordering on those import statements
 
:|
 
probably personal preference but i always put global import ... at the top, then from ... import ..., then local import ..., then finally local from ... import ...
 
^
just looks better and makes more sense
although in the case of
import stackexchange
from stackexchange.models import User
i'd probably introduce some blank lines so those can be together in a logical unit
 
okay, just a second
wait, I just realized I import logging twice
oops
import time
import json
import requests
import logging

import stackexchange
from stackexchange.models import User

from sys import stdout
from multiprocessing import Process, Value

from flask import Flask, render_template, redirect, request, url_for, make_response
from werkzeug.routing import BuildError

import constants
import kothlib
better?
 
10:01 PM
FWIW, PEP 8 (the official Python style guide) has some things to say about imports
 
doesn't PEP 8 also say that you should use snake case?
 
...yes?
you are doing that i hope
 
NOPE!
snake case bad and nothing you can say or do will change my mind
 
what are you using instead
 
camelcase ofc
 
10:06 PM
Anyone else write JS avoiding the var and function keywords?
 
Snake case is the only proper case.
 
are you really using camelcase in python
 
yes!
 
for just normal ass variables
 
That's absolutely horrifying.
 
10:07 PM
^^^
 
I oughta permaban you for saying that. ;)
 
@UnrelatedString yes ofc
well, I obviously lowercase the first letter of non-class names
 
consider how prevalent underscores are just by default
and how overwhelmingly lowercase python is
 
I'm not that uncivilised
 
camelCase =/= PascalCase
 
10:09 PM
I do not care, I will use camelcase till the day I die
and nobody can stop me
 
hacks Ginger's computer and edits all his source code
 
you fool, I'll never let you! sudo rm -rf / --no-preserve-root
 
att
why not TitleCase
 
that's for class names
 
^
 
10:11 PM
Why not SCREAMING_SNAKE_CASE?
 
that's for constants
 
Yes. And snake_case is for everything else!
 
actually, that is the only instance I ever use snake case: in constant names
 
att
who needs consistency
 
^
 
10:12 PM
Why not backwardS_hybriD-snakE_kebaB-cameL_case?
5
 
besides, as bad as snake case is SCREAMINGCAMELCASE is far worse
 
att
well, that one actually has a reason
those are minus signs
 
my pliers just fell apart
 
att
foiled by the parser
 
10:13 PM
inb4 someone comes up with "space case".
 
ooh, I can do that
spaceutwentycase
so "some long ass name" would be someutwentylongutwentyassutwentyname
 
space%20case
 
@UnrelatedString camelCase ~= PascalCase tho
 
I am getting distracted
 
@user Is that a regex match?
 
att
10:15 PM
did you know mathematica has a special space character that can be used in variable names
 
If a language uses camelCase for its variables, it probably uses PascalCase for classes, etc.
@emanresuA No, a crappy approximately equals sign
 
who wants to sacrifice their KOTH to my test software?
huh, I was expecting a swarm of volunteers
 
att
though, it displays in notebooks as a faint underscore
 
KOTH?
 
makes sense
 
10:16 PM
King Of The Hill
it's a thing we do where bots that people make compete against each other
 
@forest I have to disagree there, it makes variable names longer and just looks ugly, but that's subjective and I don't feel too strongly about it. Anyone sane would write Pytohn code with snake case because that's how everyone does it. Ginger is a psycho
> bots that people make
As opposed to what?
 
bots that bots make
 
att
people that bots make
 
Codex-generated bots, which have their own competition
 
All bots are indirectly or directly made by people
 
10:17 PM
i'd side with forest on the inherent superiority of snake_case
 
@att ಠ_ಠ
 
@UnrelatedString It's highly subjective
 
snake case bad
 
consider that i'm the kind of guy who actively fights my mobile keyboard's autocorrect to write messages in all lowercase
 
There's no inherent superiority in either snake_case or camelCase
@UnrelatedString Lmao
 
att
10:18 PM
ah, I do that too
 
I use PascalCase for my classes so camelCase for variables would just feel more consistent
 
I'm sorry you use pascal case for classes AND snake case for variables?!
 
and yall are calling me a psycho
 
att
how about this: you don't need to worry about case much if all your variables are one character long :)
 
10:20 PM
:|
 
att
it's just a matter of upper or lower case
 
please please please tell me you're kidding
 
You don't need to worry about case if all your variables are made of $ and _
 
att
even better
 
feast your eyes upon this monstrosity:
cursedPliers
 
10:23 PM
gorgeous
 
Did you twist them backwards?
 
A magician never tells
 
the hinge might be built to let you take them apart
i know i have a pair of scissors like that
 
me: *uploads cursed image*
TNB: *immediately starts deconstucting it*
@hyper-neutrino Is it okay if I post a test KOTH to CGSE to run GKOTH on?
 
att
I can see a use case for something like that
 
10:25 PM
how
 
@Ginger Feel free to turn this mess into a challenge
 
can't be bothered
and I can't use an existing one because it has to be from my account
 
Why? (I could let you use my one and only koth)
 
SE OAuth stuff
for obvious reasons GKOTH will only let you post KOTHS you made
 
SE lets you do that?
 
10:30 PM
yup
you have to sign in to GKOTH with your SE account
CMQ: Where should GKOTH store its databases?
 
... on your server?
 
do you have to post it as a question or can you test it by giving direct links to people?
 
on Radvylf's
@hyper-neutrino ???
 
like, why do you need to post a test koth
 
so that I can make sure GKOTH works correctly
@hyper-neutrino ^
 
att
10:44 PM
what is gkoth?
 
10:57 PM
^ what exactly is it and what would you like to post and why do you need special permission and can't find a work around
 
30 mins ago, by emanresu A
@Ginger Feel free to turn this mess into a challenge
 
@Ginger As a general rule: only post things that should be posted to the site. A "test KOTH" likely shouldn't, unless its a legitimate challenge
 
11:12 PM
if nothing else, consider how useful it could be to run it on sandboxed koths :)
but seriously there's no way you need that for testing anything beyond se integration and even that you cna probably mock up first
 
11:23 PM
could you test it by integrating your service with an existing koth or is that not possible
 
@Ginger you really are a special child. You've somehow managed to completely misread the vowel in the language name.What a dumbnut.
 
lyxal has arrived to end ginger's existence with a particle cannon, gg
 
Exactly.
It's in the last stages of charging
Any second now it'll end ginger's existence for commiting such a stupid act
 
11:43 PM
Is your particle cannon just a regular cannon, except you call it a particle cannon because it technically fires particles?
 
att
what if I read the language name in ipa
 
Still not vucksel
 
att
u is pretty much the closest you can get in english
 
Really?
 
att
ipachart.com listen to y
the x is wrong though
 
11:49 PM
fuck i actually am just pronouncing vyxal as /vyxal/ now
 
huh so according to ipa-reader.xyz i should be pronouncing it "voo-hall" now
 
Wtf
 
0
Q: Offtopic-TNB room’s rule

okieIt look like most of us are agree on having two separate chatroom to handle offtopic chat problem. In my opinion, any long running chatroom will require some standard rule to be running smoothly and prevent possible destruction behavior. Uncontrolled offtopic chatroom could also affect ontopic on...

 
oh frick
it's lyxal
I swear it was an accident
a moment of weakness
brb
 
00:00 - 15:0015:00 - 00:00

« first day (4111 days earlier)      last day (108 days later) »