The Nineteenth Byte

zyabin101

4:00 PM

._.

Helka homba would mean Hello.

Solver

it is lonely

no one is here

zyabin101

Yes, it is.

Solver

the night is cold

Sherlock9

Hey zyabin, what are those pants now?

Solver

silence persists indefinitely

Cᴏɴᴏʀ O'Bʀɪᴇɴ

4:10 PM

Helka Homba Homba Helka Homba Homba Homba Helka! Helka Homba Homba Helka Homba Homba Homba Homba! Helka Helka Homba Helka Homba Homba Homba Helka!

Sherlock9

Also, may I ask what the origin of your username is ?

zyabin101

@Sherlock9 Electric Unipants!

@Sherlock9 My actual name.

Solver

conor o brien is incomprehendible

Sherlock9

Marvellous!

zyabin101

Electric Unipants is the same Unipants, but reinforced with an electric suit!

Solver

4:11 PM

if that was spelt correctly

zyabin101

@CᴏɴᴏʀO'Bʀɪᴇɴ What would that mean?

Sherlock9

Thanks for answering

Solver

unipants? this will not stand. fractals.

Sherlock9

@Solver incomprehensible

English is weird

Solver

never mind

Cᴏɴᴏʀ O'Bʀɪᴇɴ

4:11 PM

@zyabin101 "no."

RΜGiEd+RΜGpq~Y~@H8'0Ed.["Helka","Homba"]H' '!S

Solver

@CᴏɴᴏʀO'Bʀɪᴇɴ So it's binary?

zyabin101

@CᴏɴᴏʀO'Bʀɪᴇɴ Pyth?

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@Solver Good job

@zyabin101 Jolf. ಠ_ಠ

I don't know much pyth

Helka Homba Helka Helka Homba Homba Helka Helka!
Helka Homba Homba Helka Homba Homba Homba Homba!
Helka Homba Homba Helka Homba Homba Homba Helka!
Helka Homba Homba Helka Helka Homba Homba Homba!
Helka Homba Homba Helka Helka Homba Helka Homba!
Helka Homba Homba Homba Helka Helka Homba Helka!

That looks really nice, yeah.

Solver

this is nonsense. pyth is nonsense. python is nonsense. javascript is the light

Cᴏɴᴏʀ O'Bʀɪᴇɴ

HelkaHomba enters the room

zyabin101

4:13 PM

Homba 42 42 42 42 Homba!

Cᴏɴᴏʀ O'Bʀɪᴇɴ

oh shoot an error

42 is jolf's error code

OH! Because I haven't pushed Jolf yet

Doorknob

@Solver please do not use code formatting improperly.

Solver

@Doorknob Okay.

Sorry.

:(

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@zyabin101 Try again? i.e. refresh the page

zyabin101

Homba 42 42 42 42 Homba!

Cᴏɴᴏʀ O'Bʀɪᴇɴ

4:16 PM

uh oh

Solver

@CᴏɴᴏʀO'Bʀɪᴇɴ it's the end.

zyabin101

How much does it take for GitHub Pages to update page?

I thought 10 minutes, but it seems like forever.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

Idk, I've seen it be as fast as a minute and slow as ten

Solver

@Doorknob if I put a comma in front of an @someone, will it still ping them?

PhiNotPi

Use Helka and Homba to write binary lambda calculus.

zyabin101

4:18 PM

,@Solver Yup.

Solver

I mean @zyabin101,

Yup.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@zyabin101 Try again, it's working for me now

zyabin101

Helka Homba Helka Helka Helka Helka Helka Homba!

\o/

Cᴏɴᴏʀ O'Bʀɪᴇɴ

:D

Solver

@PhiNotPi Good luck with that.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

4:26 PM

@zyabin101 Now I just need to write an inverse

zyabin101

@CᴏɴᴏʀO'Bʀɪᴇɴ Replace every occurence of Helka with 0 and Homba with 1.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@zyabin101 Yeah, but in Jolf

brb in ten hours.

kk bai

jk

4:28 PM

oh lol

Lembik

@TonHospel I will try it out now!

Ton Hospel

@lembik: it only matters for small m. in fact now m ~ n/2 is the hardest case for the program

Cᴏɴᴏʀ O'Bʀɪᴇɴ

Inverse:

RΜZhG~pAi["Helka","Homba"]"01' "E'!dpAΙHE

Ton Hospel

@lembik: so it won't increase the score

Downgoat

Can 4 other people with enough rep on SO close this as "caused by a simple typographical error"?:

1

Q: Grunt + Babel successfully runs but doesn't do anything

Lembik

4:33 PM

@TonHospel ok that's interesting. Are there other improvements in the paper that it would help if I asked about on math.se?

@TonHospel I have no shame :)

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@Downgoat you want to close your own question? :P

Downgoat

@CᴏɴᴏʀO'Bʀɪᴇɴ yup

Cᴏɴᴏʀ O'Bʀɪᴇɴ

gg

Ton Hospel

@lembik: Yes, the other 2 modular transforms. Which make no sense to me

Lembik

@TonHospel ok.. if you help me draw up a question I am happy to post them. I would get the question wrong if I did it myself. It's all outside of my area of math

zyabin101

4:35 PM

Ton Hospel

@lembik: in fact the first one has to be wrong unless det A=+-1, and the second one is not convenient for my code

Lembik

@TonHospel ok.. let me see if we can draw up a question now for the first one

@TonHospel are we in section 7?

zyabin101

@CᴏɴᴏʀO'Bʀɪᴇɴ @CᴏɴᴏʀO'Bʀɪᴇɴ @CᴏɴᴏʀO'Bʀɪᴇɴ v

> BUGFIXES GALORE!

Ton Hospel

@lembik: simply considering n=1 you can see it makes no sense onless |a|=1

Cᴏɴᴏʀ O'Bʀɪᴇɴ

o_o

Lembik

4:37 PM

@TonHospel sorry can you point me at the exact part of the paper we are discussing please

Ton Hospel

@lembik: yes, equation 38

zyabin101

@CᴏɴᴏʀO'Bʀɪᴇɴ sry, wrong clipboard contents. Now you can do o_o.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

Oh, I see. o_o

Lembik

@TonHospel \Gamma_2 ?

Cᴏɴᴏʀ O'Bʀɪᴇɴ

My github messages are almost as beautiful as @Adnan's

Adnan

4:38 PM

Heh wat?

Ton Hospel

@lembik: No, \Gamma_1 (i also think the a \omega a^-1 should be a \omega a^T

@lembik: but the paper has lots of minor typos of that type

Lembik

@TonHospel isn't that (37)?

Ton Hospel

@lembik: Mm, we probably have slightly different versions of the paper then

Lembik

@TonHospel I am looking at page.math.tu-berlin.de/~bobenko/papers/2003_Bob_DHvHS.pdf

Adnan

@CᴏɴᴏʀO'Bʀɪᴇɴ You mean the issue box?

Lembik

4:40 PM

@TonHospel which version are you looking at?

Adnan

Or the commits

Cᴏɴᴏʀ O'Bʀɪᴇɴ

^ that one

IIRC you do the best ones

Adnan

> BUGFIXES GALORE!

Hahahaha

Ton Hospel

@lembik: I believe I'm using arxiv.org/abs/nlin/0206009. In your version it is indeed equation 37

Lembik

@TonHospel yours is much later!

Ton Hospel

4:43 PM

@lembik: Still essentially the same. Nothing substantial is different

Lembik

@TonHospel ok

@TonHospel 2min 50 seconds for n = 24.. close :)

Ton Hospel

@lembik: Yeah, but compare the old and new versions for m=1 ....

Lembik

@TonHospel argh.. I have to go sorry. More soon I hope

thanks again for all this

quartata

@El'endiaStarman pffft

New Main Posts

1

Q: Find the last position in a matrix walking like a spiral

Introduction Given one matrix A x A and a number of movements N. You will need to walk like a spiral starting in (1; 1): right while possible, then down while possible, then left while possible, then up while possible, repeat until got N. Challenge You can only take as input the numbers A...

code-golf matrix

Cᴏɴᴏʀ O'Bʀɪᴇɴ

4:58 PM

Anyone want to help edge-test-case or whatever Dyia?

zyabin101

a =: 0

\/

Token[a]
VerbToken[=:] dyadic
NumberToken[0]

Token[a]
NumberToken[0]
VerbToken[=:] dyadic

@CᴏɴᴏʀO'Bʀɪᴇɴ Is the test parse tool supposed to output both forms?

Cᴏɴᴏʀ O'Bʀɪᴇɴ

Yes, it is.

It's for visual comparison of the tokenization and shunting

Agawa001

@trichoplax mmm now i recall, i solved some of his challenges, the pristine was an ordeal

zyabin101

+ -4

\/

TypeError: _op is undefined

Agawa001

anyways do u have an idea how to handle this tool called vi ?

Cᴏɴᴏʀ O'Bʀɪᴇɴ

5:06 PM

@zyabin101 Subtraction isn't defined yet.

(-4 isn't a number; it's the monadic application of - to 4)

zyabin101

@CᴏɴᴏʀO'Bʀɪᴇɴ Then what is a negative number?

Cᴏɴᴏʀ O'Bʀɪᴇɴ

There aren't any right now :P

puts on todo list

zyabin101

Oh, I see.

+ _4

\/

NumberToken[1]

Cᴏɴᴏʀ O'Bʀɪᴇɴ

o_o

Oh, right.

I need to implement _ too :P

zyabin101

Apparently :language C translates Vim to the C locale.

Is it a version of an existing language or something?

quartata

5:12 PM

@zyabin101 pubs.opengroup.org/onlinepubs/009695399/basedefs/…

C localization functions

In computing, C localization functions are a group of functions in the C programming language implementing basic localization routines. The functions are used in multilingual programs to adapt to the specific locale. In particular, the way of displaying of numbers and currency can be modified. These settings affect the behaviour of input/output functions in the C Standard Library. == Overview of functions == C localization functions and types are defined in locale.h (clocale header in C++). == Criticism == C standard localization functions are criticized because the localization state is stored...

You're probably looking for :set filetype or :set syntax

Optimizer

^.o

zyabin101

@quartata Nope.

I was actually looking for a way to translate Vim to a specific locale.

quartata

Why would you want to set your locale to the C locale? Generally you only do that if there is no other locale on your system...

zyabin101

:28905447 My LANG equals ru_RU.

quartata

Do you want to change it in vim or everywhere?

Zgarb

5:16 PM

@CᴏɴᴏʀO'Bʀɪᴇɴ Should adverb chaining work yet? 1 % 2 gives 0.5 and 1 %~ 2 gives 2 as they should, but 1 %~~ 2, 1 %~~~ 2 etc seem to all give 2.

zyabin101

@quartata In Vim.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@Zgarb Ooo nice catch. I forgot about that, thanks!

quartata

Then use :language

Put it in your .vimrc

zyabin101

I wanted to show Vim to Agawa (who left). Then I thought she doesn't understand Russian, so I was finding a way to change the locale.

@quartata Then gvim would be weird. The interface is all in <insert current vim locale> but the menus are in Russian.

quartata

:set langmenu

zyabin101

5:18 PM

@quartata Ah.

quartata

set langmenu=en_US.UTF-8 for instance

Zgarb

@CᴏɴᴏʀO'Bʀɪᴇɴ Should parentheses work?

Cᴏɴᴏʀ O'Bʀɪᴇɴ

@Zgarb Not yet, no.

Zgarb

Ok.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

Also, fixed adverb chaining.

@zyabin101 Fixed _ in numbers.

zyabin101

5:23 PM

a =: _3
a + 3

\/

NumberToken[0]

You made negative numbers! \o/

Cᴏɴᴏʀ O'Bʀɪᴇɴ

Yup, that's correct. I currently haven't broken anything out of tokens for output yet.

@zyabin101 :D yup!

BTW I love the large arrow \/

Zgarb

Hmm, now a lone underscore _ is 0, two or more are NaN.

Extra numbers, like * 2 3 or * 2 _, also give funky results.

a =: 2
a

Token[a]

@CᴏɴᴏʀO'Bʀɪᴇɴ ^

This executes happily:

a =: 2
1 =: a
a + 0

And 2 + 1 =: 2 gives 4 :D

Cᴏɴᴏʀ O'Bʀɪᴇɴ

:o

Did I allow for redefining of numeric constants? XD

Zgarb

5:38 PM

Not globally, 1 + 1 after that returns 2 as it should. I think =: just takes whatever as its left argument and tries to redefine it, and returns the right argument.

For numbers the redefinition just silently fails or something.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

Hm. Is that a good thing?

(I'm trying to make a sane language, not a golfing language.)

Zgarb

Depending on your exact implementation, I guess it could bite you back later on.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

Okay. So it should probably through an error?

Zgarb

Yeah, that seems the sane thing to do.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

Cool, thanks!

@Zgarb What does that mean?

Zgarb

5:45 PM

@CᴏɴᴏʀO'Bʀɪᴇɴ * 2 3 returns 2, and I have no idea why.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

o_o

VerbToken[*] dyadic
NumberToken[2]
NumberToken[3]

NumberToken[2]
NumberToken[3]
VerbToken[*] monadic

The result from the stack is shifted and outputted; I have no detection in place for extra members on the stack.

So the stack would look like [2 *3] = [2 1], then 2 is shifted.

Zgarb

Haha, this gives 1:

2 =: a 1
0 + a

Cᴏɴᴏʀ O'Bʀɪᴇɴ

BTW I fixed the non-display of tokens, so just refresh the page

Zgarb

Ok, cool.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

Should I do lists with spaces (like J) or with commas (like Jelly)?

Zgarb

5:51 PM

Whatever floats your boat. They are probably much easier to parse if you have some brackets around them.

Cᴏɴᴏʀ O'Bʀɪᴇɴ

True. Would it be a bad idea to have the comma be an operator that joins two entries, and treat it like a list?

Zgarb

I don't really have an opinion on that. APL and J have , for concatenation and pair construction, and it works for them.

On the other hand, building a list of literal expression will then require lots of parentheses, as in (a+4),(a+7),(a*2),(b+1)

But in practice that may be rare.

Solver

6:10 PM

Hello?

zyabin101

@Solver Yes.

Hello.

Solver

@zyabin101 Wow, that was quick!

Maybe electric pants make you run faster :P

3

encrypted-tbn0.gstatic.com/…

^^^ that is the funniest thing I've seen all day.

Ton Hospel

@lembik: Ah, I finally start to understand. Since both a and a^-1 must consist of integers in n=1 they must be 1 or -1, so there is no blowup

@lembik: simularly det of integer matrix must be integer, same for inverse, so det = +1 or -1

@lembik: So that transform is unimodular so it's just a lattice redefinition

Downgoat

6:30 PM

In Cheddar, should preset built-ins such as range and Pi be un-overrideable?

Lembik

@TonHospel cool!

@TonHospel I am running timing tests for n= 22 and m = 0..21

@TonHospel it gets slow around m = 6

Ton Hospel

@lembik: I assume high and low m are fast ?

Lembik

@TonHospel high is very fast.. low not so lightning fast.

well.. is m=6 low?

Ton Hospel

@lembik: not so much :-) 1 and 2 are low (they used to be even slower than m=6)

Lembik

@TonHospel ah ok :)

do you have a feeling for what the largest possible value is given a fixed n and m?

Ton Hospel

6:37 PM

@lembik: No, since I still don't completely understand. The speed is mostly determined by det(P/pi), but I don't really know yet to what extent eccentricity also plays a rolw

@lembik: det(P/pi) small => slow (the exponential decays too slowly)

Lembik

@TonHospel ok thanks

maybe no one knows :)

or I could make a challenge to create the largest sum with this matrix structure :)

(just a joke)