The Nineteenth Byte

Anush

18:06

How you iterate over all lists on

How you iterate over all lists of 10 non negative integers whose sum is 10 in an efficient way?

I feel like this must be known

pxeger

Consider: how do you iterate over all lists of 9 non negative integers whose sum is 9 in an efficient way?

How do you iterate over all lists of 8 non negative integers whose sum is 8 in an efficient way?

How do you iterate over all lists of 7 non negative integers whose sum is 7 in an efficient way?

...

Mayube

did you mean to change the list lengths for those 2nd and 3rd ones?

pxeger

How do you iterate over all lists of 0 non negative integers whose sum is 0 in an efficient way?

@Mayube no, but I do now :Þ

Mayube

How do you iterate over all lists of n non negative integers whose sum is n in an efficient way?

hyper-neutrino

regarding the idea to scan all chat messages - unfortunately there is no way of distinguishing a message that's been deleted or is in a private room from a message ID that is yet to exist; thus, just iterating up in message ID won't really work

one way to circumvent this is to have a bot periodically post a message in the sandbox and report its ID so we know where the boundaries are, and every 404 before it is just hidden

pxeger

18:12

well, actually, {all lists of 10 non negative integers whose sum is 10} = (map (1:) {all lists of 9 non negative integers whose sum is 9}) ++ (map (2:) {all lists of 9 non negative integers whose sum is 8}) ++ ... ++ (map (10:) {all lists of 9 non negative integers whose sum is 0})

so yes, @Mayube I did mean that

@hyper-neutrino context?

hyper-neutrino

@pxeger we were discussing the easiest/best method to scan all new incoming chat messages from all rooms to do a smokey-like scan in TL and I asked here and Redwolf came up with a few ideas

one of them that I also had is to just while True: x++; get(chat.stackexchange.com/message/x)

I doubt it'd really do much. just a "fun" (for some definition of fun) little idea

pxeger

@hyper-neutrino does the "in TL" apply to "we were discussing", or "do a smokey-like scan"?

hyper-neutrino

the former

pxeger

if you only want to scan all new incoming chat messages, you could just join all rooms, couldn't you? or is the goal to avoid doing that

Redwolf Programmed

There's currently 800 unfrozen rooms lol

And that tends to be the average from what I've seen

pxeger

18:17

but there isn't any technical limitation on joining that many rooms, is there?

hyper-neutrino

not really. but having a random account exist in every room is sure to raise questions

pxeger

can't you join silently, as happens when you're reading the live transcript?

Redwolf Programmed

No, that just checks the /events every few seconds

Which, for 800 rooms, is also called a DoS attack lol

pxeger

wait what? lmao

thejonymyster

18:44

@Mayube im trying this with desmos, i dont know if theres a way to turn a list into a smaller list, so im going to have to think of how to mathematically extract this value :thinkingemoji:

hyper-neutrino

lists of X non-negative integers that sum to Y is just [k, lists of X-1 non-negative integers that sum to Y-k] for all k ∊ 0 .. Y

oh wait pxeger already said that but differently lol

Neil

@RedwolfProgrammed ... which it is...

Redwolf Programmed

*which it can be

But it's annoying

thejonymyster

are any words improved by having their plural form be the same as their singular form

hyper-neutrino

is the singular not datum

Redwolf Programmed

18:48

It was, yeah

pxeger

it used to be

thejonymyster

besides buffalo for lore reasons

pxeger

but thanks to the social construct that is language...

I mean, it still is, but "data" is also now an uncountable

thejonymyster

wait how can you use data as a singular anyway

pxeger

so it is a singular, and it doesn't have a plural

thejonymyster

18:49

what even is "a data"

Redwolf Programmed

What's a datum either?

pxeger

@RedwolfProgrammed a data point, or a piece of data

thejonymyster

it works the same as the word "information"

Redwolf Programmed

^

pxeger

@thejonymyster it's uncountable, so you can't use it with "a", but you'd still say "this data is" (grammatically singular) instead of "these data are" (which is the old / "technically correct" version)

Redwolf Programmed

18:50

Uncountable is so much better

thejonymyster

why on earth would ever say these data are? these information are

Redwolf Programmed

@pxeger No such thing as "technically correct" in english

pxeger

@RedwolfProgrammed "uncountable" isn't even a noun, let alone an uncountable noun, so you can't say "uncountable is" :Þ

Redwolf Programmed

What's technically correct is what people understand

thejonymyster

@pxeger [having it be] uncountable is...

pxeger

18:51

@RedwolfProgrammed I'm aware, I should have said "technically correct" rather than technically correct

@thejonymyster joke

thejonymyster

nod

but yea no nobody should be debating how to use the word data because theres only one sane way

Neil

@RedwolfProgrammed no, it has a singular

Redwolf Programmed

It had one :p

If you still use "data" as plural though, you've lived in a cave for the last few decades

Neil

@RedwolfProgrammed I'm going to need more data to substantiate your claim

thejonymyster

wahts the singular form for "information"

pxeger

18:54

@RedwolfProgrammed it still does, because, as this datum shows, some people still use it that way

@thejonymyster "information" is singular

thejonymyster

whats the plural form for information

pxeger

there isn't one

because "information" is uncountable

Redwolf Programmed

This is a pretty pointless argument though

Mayube

data is the plural form of datum

@RedwolfProgrammed or you work in academia

thejonymyster

what is one datum genuinely because i thought data was like, by definition uncountable

Redwolf Programmed

18:55

Same

pxeger

5 mins ago, by pxeger

@RedwolfProgrammed a data point, or a piece of data

thejonymyster

oh, thats fine

a datum of data

Mayube

a datum is one piece of information or numerical result

data is the plural of datum

thejonymyster

now in my head its not a weird debate about plurals, its just "theres this funny word that means a single point of data"

Redwolf Programmed

lol

pxeger

18:56

thejonymyster

data is still uncountable, datum is just its own thing (to me at least)

pxeger

I'm sorry cartograpghers and engineers, but what the fuck?

Mayube

so 3 is a datum. [2, 3, 6] is an array of three data

thejonymyster

@Mayube that would be incorrect :D

thats three datums ^_^ hope this helps

Mayube

datums is not a word

thejonymyster

18:57

@pxeger yes it is :P

Mayube

data is the correct plural form of datum

Redwolf Programmed

"correct"

pxeger

@Mayube "datums" is a word, because cartographers and engineers use it as a word

language is a social construct

thejonymyster

@Mayube but thats just begging the question

Mayube

@thejonymyster in the context of cartography and engineering specifically, sure

thejonymyster

18:58

@Mayube right sorry that one was a joke, but i think if were going to have the word "datum" it cant be part of the same word as "data" anymore, "data" has moved on and become a strong independent uncountable

hyper-neutrino

@Mayube it's still a word, just maybe not the correct one to use in other contexts

Mayube

@thejonymyster it's not fighting a losing battle against change, because nothing's changing. data is the plural of datum. data is also an uncountable noun. Words can have multiple usages depending on context

hyper-neutrino

so is xnopyt

Redwolf Programmed

The only correct usage of "data" as a plural is to refer to the character from star trek

thejonymyster

@Mayube but thats so bad

Redwolf Programmed

19:00

(Don't star that it's an overused xkcd)

thejonymyster

@RedwolfProgrammed star of shame

Mayube

@RedwolfProgrammed But there's only one Data. Datalore is not Data

@thejonymyster Why?

thejonymyster

because its one word for two closely related things in the same context

which leads to unnecessary confusion

like monads

Mayube

welcome to the english language

Redwolf Programmed

@Mayube Personally I dislike "data" as a plural because it's inconsistent with "information", and I find the concept of a single datum kind of pointless

But this argument's still pointless :p

thejonymyster

19:01

>datum
>pointless

Mayube

@RedwolfProgrammed because you don't work in statistical acadamia

Redwolf Programmed

Yep

It's fine for a word to mean different things to different people, the worse english is the more I like it :p

thejonymyster

well as long as it makes for interesting word puzzles i guess

Mayube

English is actually just an esolang experiment to make an esolang so inconsistent that the only way to build a parser for it is to evolve life

thejonymyster

every day we plow through syntax errors

wait lol programming language descriptivism

hyper-neutrino

19:05

oh no

Mayube

syntax error*
`error` is the plural of `errum`

thejonymyster

the "correct" syntax is whatever the programmer uses

hyper-neutrino

for those of you who are annoyed by how many chat flags math.se chat generates

you'll love to hear that math.se is starting an election in 3 days

Mayube

@thejonymyster isn't that just JavaScript?

thejonymyster

personally i think every plural should be the singular form with an "s" at the end

dogs, cats, fishs, datums, etc

hyper-neutrino

19:06

personally I think we should just not have singular/plural forms, like mandarin does

Mayube

@thejonymyster Can you think of a plural noun that becomes a singular noun when you add an s?

there might be multiple in english, but I know of 1

thejonymyster

i cant, is there one? :O!

Mayube

princes

hyper-neutrino

lol

thejonymyster

ohhh hehe

Mayube

19:07

plural of prince, add an s it becomes princess

thejonymyster

~~new rule: no words can end with s other than plurals~~

Mayube

sheesh

that's gunna mess up some words

thejonymyster

mezz*

Mayube

the sheer mass of that change would cause quite a lot of stress

thejonymyster

who could guess

but i digress

Mayube

19:09

also what about continuous verbs?

walks eats becomes writes codes

thejonymyster

i tried to read that as a single sentence haha

Mayube

hahaha

new rule: code is a single lexicographic token, codes is a complete program

we don't write code we write codes

thejonymyster

codum and coda

Mayube

codums

but only in cartography and engineering

low-effort call backs are one of my favourite forms of comedy, up there with non-sequiturs and post-irony

and of course puns

thejonymyster

i like it when the low effort call backs keep accumulating together like a big dumb joke katamari

hyper-neutrino

19:15

> 塊

Redwolf Programmed

@Mayube Bring back and repurpose walketh, eateth, becometh, etc. as continuous

hyper-neutrino

that's a funny looking kanji

Mayube

@hyper-neutrino if you squint real hard it looks like a person with a sword and a swollen leg

hyper-neutrino

... interesting

thejonymyster

i wish i looked like that

Mayube

19:20

"My pronouns are he, not him.. cos I'll never be him"
塊

Don't you hate when you spend like half an hour writing a super detailed explanation of your code, then notice multiple ways to golf it and have to rewrite the whole thing?

hyper-neutrino

writing explanations for my code is actually where a lot of my golfs come from

because as I'm typing an explanation for it I realize "wait this is dumb"

Mayube

same, but when it's a ~460 byte interpreter in python it's a lil soulcrushing

hyper-neutrino

lol

yeah

a lot less annoying for a like 10 byte jelly answer to rewrite an explanation that took me like 4 minutes

caird coinheringaahing

@Mayube I do that with basically every Whispers/Add++ answer, which may be why I don't really use them :P

Mayube

There've definitely been a few times where I have a few hundred byte JS answer with a beautiful explanation, then I significantly golf it and just give up and remove the explanation :(

caird coinheringaahing

19:29

I do remember one Jelly answer where I did "indices; powerset; index into input" and HN pointed out that was just "powerset" :/

@Fmbalbuena You're learning Add++?

Fmbalbuena

yes

Anush

CMC iterate over all arrays of 10 non negative integers whose sum is exactly 10 without creating 10^10 arrays and filtering

caird coinheringaahing

@Fmbalbuena what can I help with? :)

Fmbalbuena

Try it online! Why i cant use?

thejonymyster

@Mayube solved in desmos :D! this was fun even solving it, ill have to golf it when im not at work. input with i, output is o. Everyone should click this i think

2

caird coinheringaahing

19:33

@Fmbalbuena You need to separate the On and Ox commands with ,,: Try it online!

hyper-neutrino

@Anush tio.run/##TcnBCoQgEADQ8/YVc9TWoKFbUD8iHoyywt0xRND5eqNb1/…

Mayube

Impressive, good job!
How do we score desmos answers? :P

caird coinheringaahing

The indentation and newlines aren't significant in Add++, so it ignores them and tries to run OnOx, which isn't a command

@Anush I think you can do it with less than 10^10 arrays, while still filtering :P

thejonymyster

@Mayube you score it based on the code you paste into desmos to make it work

in its current state, it isnt actually scoreable

Anush

@cairdcoinheringaahing how much less?

thejonymyster

19:35

im just excited there is a solution :]

caird coinheringaahing

Well, we know trivially that [0,0,0,0,0,0,0,0,0,0] doesn't, so, we can say that 10^10-1 is an upper bound, at least :P

Mayube

@thejonymyster that's awesome

thejonymyster

ty ^_^

Fmbalbuena

@cairdcoinheringaahing thx but how can execute a variable as Python

caird coinheringaahing

@Fmbalbuena Like, if a variable has a string, how do you run that string as Python code?

emanresu A

19:37

@Mayube YES. It's so annoying.

thejonymyster

@Anush i dont think i will be solving this one in desmos :P

caird coinheringaahing

You'd have to use functions for that, as Python exec isn't a command in vanilla mode

Try it online!

Mayube

@emanresuA just realised toward the end of writing up my explanation that 4**9 is both fewer bytes and a larger value than 2**15. A small change, but I have to change it in 4 places: Code in answer, code in TIO, ungolfed code in TIO, explanation

caird coinheringaahing

@Anush alternatively, I can do it by creating a lot more than 10^10 arrays and filtering :P

Fmbalbuena

@cairdcoinheringaahing and i want to upvote the language

caird coinheringaahing

19:41

@Fmbalbuena Well, if you have a Github account, you can star the repo. You can also upvote its LOTM nomination instead :P

Tho that does mean I have a month to finish v6 :?

Fmbalbuena

I already upvoted LOTM

And i star your github repo

caird coinheringaahing

In my mind, that counts as upvoting the language :P If you're interested in using Add++ on the site, there is a open bounty for answers in it

Fmbalbuena

@cairdcoinheringaahing Try it online!

Mayube

Alright it's finished. I'm gunna cry if this gets significantly golfed and I have to update the explanation.

Also @cairdcoinheringaahing some of that code might look familiar ;)

Anush

@cairdcoinheringaahing :)

No lovers of Knuth here :(

caird coinheringaahing

19:52

@Mayube Does that use the Brainflump strategy of indexing into a list of code? :P

Mayube

@cairdcoinheringaahing yes. It also uses Brainflump's strategy for handling loops

rak1507

@Anush twitter.com/rak_1507/status/1455658193040908293?s=20 it's no knuth november :P

4

Fmbalbuena

@cairdcoinheringaahing See you later

caird coinheringaahing

o/

Mayube

really it was just an adaptation and further golfing of brainflump's interpreter to handle volatile instead :P

thejonymyster

19:54

cmegac given a list of integers output the coolest one

Mayube

@thejonymyster JS, 14 bytes: a=>a.sort()[0]

Braingolf, 1 byte: x

Anush

@rak1507 noooooo

emanresu A

@Mayube Sorry for golfing, but why are you assigning C when you only used it once?

Mayube

@emanresuA it's used 4 times

emanresu A

I mean capital C

Mayube

20:01

oh uppercase C

emanresu A

C=u[i]
if c==C:f+=1

Mayube

good point :P

emanresu A

And before you try it, f+=int(c==u[i]) saves no bytes.

Also, loop#2 can become for c in a, then you can get rid of the assignment to c

Mayube

no it can't

i is assigned for loops

can't go backwards in for c in a

emanresu A

Oh.

Mayube

20:08

happen to know of a 3-byte expression that evaluates to >= 2**15?

emanresu A

9e9?

Mayube

oh dang I forgot about e

emanresu A

Also (not s[-1]) can become ~s[-1]

thejonymyster

forgot about dr e

woah, i cant infix that?

emanresu A

You can, you just need spaces around strikethrough

---a--- b

thejonymyster

20:10

but then it messes up the

Mayube

@emanresuA can it? Isn't that bitwise NOT?

emanresu A

Yes but overloading/type conversion is bleh

Mayube

but if it's a bitwise NOT it won't do any type conversion. It's multiplying a string, if the result is anything other than 0 or 1 it's gunna have unintended consequences

wait are negative numbers falsey in python2?

emanresu A

Idk but it seems to work on the other testcases

Mayube

no they aren't

@emanresuA doesn't work if top of stack is negative

volatile's loops are != 0 not > 0

emanresu A

20:15

Ok nvm

Mayube

I think I had a whole-ass unnecessary line in there too

emanresu A

Dammit maqrkdown

Mayube

edit it on TIO and post the link

emanresu A

At the end you can assign e to [e,...][g+1] for -3 bytes:
https://tio.run/##NVHLbsMgEDzHX4EiRYCNnbjqpZDtF0T5AcuqLIwdJAcQfqS95NddcNMD7GrZ2Zkd3M90s@ZtXTtv78g3pg1B3531U5o04JvHlzZunggVGhY4iRGqWljAjoyFs47Qw/hKKBYDDMoIB2PROKdMm7SqQxcyM8muzFOeIN2huZB2NhOR9HzlXk2zN3mZoBnminNfiw5OCeqsRxppEyX1igxkpgG@C3AJoVHXvMug3AodwP8cFPtyncWQEv95osnjpgeF9HkgTeSX0ASwUICx6AE/eYGLTps2qNnEycCJszw9Yq7AHmQsYoIBJF8CYQRquJCG7emeLaykOCXGTmis8rLeRmC6dW/Stip/AUgE5CUNy@7UMCq@5GGDSFpmfWCrgqfxA3Tw5sTe0/QjWMqi0dvsmPr4Fs9fieK66usEqW8lkRI6CFzXJ8@LJz9mBdlu@gs

Mayube

your change didn't take in the link, but I see what you mean

in fact i can skip assigning e at the end altogether and just exec [e,...][g+1]

emanresu A

20:25

Nice!

Mayube

wait if I put e at the end I don't need the +1

a grand total of 43 bytes golfed so far. Thanks @emanresuA!

emanresu A

One more thing - Why assign g?

Mayube

20:44

because at the time it was used in the if and the array index, but now the if is gone

emanresu A

So -5 bytes by removing the assignment

Mayube

-4*

emanresu A

Do I have too many tabs open?

@Mayube I count good

Mayube

@emanresuA We don't need to be able to math and count, we have computer for that ;)

Unrelated String

emanresu A

20:46

@UnrelatedString ಠ_ಠ

Mayube

@UnrelatedString Imagine having that warning turned on in 2021

Neil

1022 is an anagram of 2021

Mayube

oh hey so it is

Redwolf Programmed

CMQ: Cool name for a chat program?

rak1507

epikchatzz9000

emanresu A

20:49

@RedwolfProgrammed Cool name for a chat program

rak1507

you can have that one for free

Mayube

@RedwolfProgrammed "Quot"

pxeger

@emanresuA cnfacp

emanresu A

Sure

pxeger

@RedwolfProgrammed do what I do when naming projects: tr -dc a-z < /dev/urandom | head -c 4

how about cvih

Mayube

20:55

with a couple of synonyms you can make that pronouncable: `Cifaca`
Cool identifier for a chat app

emanresu A

Although try pronouncing that

Mayube

See-fah-ka

pxeger

I think emanresu meant cvih

Mayube

k-vee

emanresu A

@Mayube That's exactly my point

What does that sound like?

Mayube

20:58

@emanresuA if you think it sounds like that you're not stressing the ah enough. Think the way a bostonian would pronounce "far"

emanresu A

I guess... Still sounds a bit wrong to me

pxeger

@emanresuA /kəvɪx/

Transcript for