The Nineteenth Byte

petStorm

12:20 AM

@ChristianSievers I suspect that the person is serial-downvoting. They're downvoting every answer to the question.

Lyxal

1:24 AM

@petStorm every answer? Judging by the first page of answers, only five answers were downvoted.

Actually, there is only one page

So only ~27.77 of answers are downvoted

petStorm

1:42 AM

@Lyxal How's your progress on prime checking in naz? Is it easy or hard?

Lyxal

I haven't started writing a program yet, but I think it should be relatively easy

For numbers in the range [1, 127] that is

Numbers bigger than 127 are a no-go using my method

petStorm

You know, there's an extra limitation. You can't do arithmetic between two accumulators.

Lyxal

My idea focuses on using ~3-4 variables, modulus and the register

@petStorm 5-6 variables then

pppery

Isn't one of the benefits of Python that a script can act as a function? @pppery Looking at a lot of the answers in this answer, that seems to be the consensus among most answers. Is there a community definition of a "program or function" I can reference in the future? — GetHacked 7 hours ago

Could someone please answer the question in that linked comment? I'm having some trouble figuring out where exactly on meta the rules on what makes a program/function is defined.

petStorm

@pppery A script can be treated as a function if command-line arguments can be taken and the program outputs via a return code. This snippet does nothing analogous to a function.

pppery

1:48 AM

@petStorm The second part of the question was where exactly on meta (or the help center) that rule is defined, which is what I think GetHacked was asking.

petStorm

@pppery Here

Lyxal

@pppery This outline of what is expected from Python answers by xnor should help

> Nor can the output just be the value of an expression like in the interactive shell.

If code placed in Try it online! doesn't produce output, and it isn't a callable function, then it isn't valid IMO.

It can be turned into a lambda for 16 bytes

pppery

Could one of you post that as a comment on the linked question?

Lyxal

@pppery looks like @petStorm and I both did

petStorm

2:04 AM

CMC: restricted-source Repesent the Fibonacci Sequence in the Set Theory. You should not use natural language or ellipsis. Most importantly, you can't use the Fibonacci formula to calculate.

Here is a tutorial on the Set Theory.

Ad Hoc Garf Hunter

2:27 AM

Your set theory is a little odd.

petStorm

3:00 AM

@AdHocGarfHunter Why is that?

Ad Hoc Garf Hunter

Universal is a rather uncommon set for consistent set theories.

2

petStorm

Note that I am describing the naive set theory. I pretty much added that just for completeness.

Bubbler

3:29 AM

@petStorm Can I use matrices? {x|n∊N,x∊[1 1][0 1]^n}

Or fixpoint and max? fix(S→S∪{max(S\max(S))+max(S)},{0,1})

Ad Hoc Garf Hunter

max is just union in the von neumann encoding

Bubbler

Or Haskell-style self reference? ∪F|F={{0,1}}∪{{y,x+y}|{x,y}∊F,x≤y}

Ad Hoc Garf Hunter

Oh wait now I see how it works.

Bubbler

3:46 AM

Or is it acceptable if I define the formula using just integers? {x|n∊N,x=(a^n+b^n)/c,a^2-a-1=0,b^2-b-1=0,a>b,c^2=5,c>0}

petStorm

The self-reference is really clever.

Bubbler

I think the self reference could actually work as a mathematical notion of "the set F should satisfy this constraint".

:54031058 It is needed because without it, you'll get {1,3} from {1,2}={2,1}, and then everything is screwed.

petStorm

So it's pretty much ungolfable? What's ∪F| for then?

Bubbler

It is for flattening the set of sets.

petStorm

So is the result {0, 1, 2, 3, 5, 8, ...} instead? Does it handle duplicate 1's? (Sorry, I'm not questioning you.)

Bubbler

3:59 AM

Yes, it is.

It's a set after all, so {0, 1, 1, 2, 3, 5, 8, ...} = {0, 1, 2, 3, 5, 8, ...} by definition.

More explicit expression would be {x|x∊f,f∊F,F={{0,1}}∪{{y,x+y}|{x,y}∊F,x≤y}}.

petStorm

4:16 AM

@Bubbler I tried to write a version that does not employ flattening. F={0,1}∪{X+Y|{X,Y}∊F,Y-X<X}

Does this work though?

Bubbler

No, because F doesn't have any two-element sets as its member.

petStorm

Ahh, subset then. F={0,1}∪{X+Y|{X,Y}⊆F,Y-X<X}

Bubbler

Pretty close, but it doesn't ban 4 from F because {2,2}⊆F,2-2<2.

petStorm

Therefore I need to make sure that X and Y are different items...

F={0,1}∪{X+Y|{X,Y}⊆F,Y-X<X,X≠Y}

Bubbler

It doesn't reach 2.

petStorm

4:26 AM

Hmm... F={0,1}∪{X+Y|{X,Y}⊆F,0<Y-X<X}

Bubbler

It still doesn't reach 2.

I think F={0,1,2}∪{X+Y|{X,Y}⊆F,X<Y≤2X} works.

Ad Hoc Garf Hunter

F={0,1,2}∪{X+Y|X∈F,Y∈F,X<Y≤2X} is shorter

Oh wait it is the exact same length.

petStorm

I guess the cartesian product of two individual items wrap both them into a pair in a set. Any uses here?

Ad Hoc Garf Hunter

{X|(X,Y)∈F,F={(0,1)}∪{(Y,X+Y)|(X,Y)∈F}}

Bubbler

5:32 AM

@AdHocGarfHunter That would work, but only if tuple is allowed.

petStorm

Tuples are not allowed.

New Sandboxed Posts

5:50 AM

0

A: Sandbox for Proposed Challenges

Sum in 2540 Sums code-golf code-shuffleboard This is my attempt to pair code-shuffleboard with code-golf. You need to write a program that sums all codepoints of the input string. Score Your base score is the sum of the ord codes of the codepoints. Your base score has to be exactly 2540. You...

0

A: Sandbox for Proposed Challenges

Output half of your program code-golf quine Output half of the characters that make up your source code, the outputted characters can be any characters of your source code. Rules If your source code is of an odd length, you need to output one extra character from your source code. The number ...

New Sandboxed Posts

6:08 AM

0

A: Sandbox for Proposed Challenges

Reverse Internal Truth Machine Ok, the previous one was too easy, because stack languages. 0 # Push input Truth machine # Do stuff However, what if instead of putting it at the start, you put it at the end? Of course, this is still trivial for ><>, but what about for other langu...

petStorm

7:23 AM

I tried to follow everyone in this organization.

As a result my index finger is sore... I've definitely clicked 2.1k+ times.

I really hope that GitHub adds a "follow everyone" button at the start of the organization... so that I can mass-follow people in an organization.

John Dvorak

You can switch to another finger temporarily :P

petStorm

7:55 AM

Now I've improved my algorithm. I can now process follows at 2p/s. (p=people,s=second)

Now I have currently only followed 1.7k people. Am I missing anybody?

Side note: can I crash ~~the GitHub server~~ my browser with my follows?

If not, I'd continue until it does.

Bubbler

It might crash your browser, but it's very unlikely that it will crash the GitHub server.

petStorm

@Bubbler During my organization-follows, I see pages of people without "follow" buttons next to them. So I start to think that it might be a problem in a server. I refreshed and the buttons still don't appear, I often need to do it multiple times for them to appear.

petStorm

8:51 AM

I did this for 20 minutes and I only followed 2.6k people... sigh.

petStorm

The Unix cat program copies its input to its output.

A copycat copies another thing to their/its own thing.

So the Unix cat program can be considered as a copycat. Hence the name.

My organization almost hit the amount of the Intel corporation!

Checking my activity bar...

But first, I'll follow everybody in the Intel organization.

My browser has 16 tabs open, a lot fewer than the people I've followed.

Now I gotta follow so many people that there's an activity every second on the GitHub main page.

Also, I am surprised that my speed is a lot faster than before. I can now process at 4p/s.

By following people, I get the achivement of successfully poking all bubbles in bubble films.

petStorm

9:36 AM

You exercise your body -> Your body gets heavier -> You can't move your body that easily -> Your body gets less exercised. So what's the point of exercising?

Now I keep seeing the word follow as fellow.

"Follow the fellow!"

It's even more complicated now. I got 21 tabs open ...

Lots of small organizations. Lots of people to follow.

I got the TensorFlow to page 3, I'd better finish it.

I can now watch people like watching a film. But 3k isn't enough.

I said that I'd like an activity to watch every single second.

So, I'd do it!

petStorm

9:55 AM

They got 1.2k people in the Google organization? Even less than Microsoft?

petStorm

11:03 AM

Fantastic! One activity every 3 minutes!

petStorm

11:26 AM

After seeing so many people, I got an irresistible rage of following all of them.

Because of that, I have a side-effect of lots of people following me...

Only 3 people out of 3.3k people follow me. That's a weird ratio.

After that I can enjoy the consequences of that. I can find new people to add to my org more easily!

(How did they manage to have so much activity all at once... :) )

Git it! I finally followed everyone in the .NET organization.

Just 3 people left until Intel!

That organization-map seems to be endless...

I never thought that I could ever clear the queue.

From a SINGLE user! I got so many followable people.

Obvious next step: follow all of the followers of famous people.

You know, Guido van Rossum has 9.8k people! If I follow all of them ... I'd be viewing a huge increase of activity!

When they heard footsteps approaching, they approached to bed.

Whoo! Got to the second level! Now trying to get to per second.

(BTW I'm now following 3.8k people!)

petStorm

12:29 PM

Doing all that following is like playing Dino Run (the built-in Google Chrome game). It's really fun!

Coincidentally I got that feeling when I am following everyone in the Google organization.

After training (following), my hands are a lot more agile than before.

And now that no-button problem again...

Alright, it has loaded.

I click a lot faster than GitHub responds!

Whoo! Now I'm following 5.1k people.

I've finally tacked the Google organization!

petStorm

1:36 PM

One of my arms are sore, so I had to switch to the sup-optimal algorithm.

petStorm

3:01 PM

I'm doing it fairly efficiently.

The most annoying part is that I always see pages without follow buttons...

Alright. I'm signing off.

My speed is 8k/d.

New Main Posts

3:24 PM

0

Q: Chase the Easter Bunny and collect his eggs

Chat room: https://chat.stackexchange.com/rooms/106513/easter-bunny-challenge Git repo: https://github.com/AJFaraday/easter_bunny_hunt The garden is a 2-dimensional grid with 49 rows and 49 columns. The Easter Bunny™️ is in the center of the garden, minding his own business, holding on to 100 ...

code-challenge

New Meta Posts

7:16 PM

0

Q: Would randomness-based languages be allowed?

I've been thinking about creating a language where it's impossible to create a constant with a known value. Thus, it's impossible to guarantee a program will work 100% of the time (though you'd be able to achieve an arbitrarily high accuracy, at the cost of bytes). My question is, would it be al...

discussion

petStorm

9:12 PM

3

A: Sandbox for Proposed Challenges

Sum in 2540 Sums code-golf restricted-source code-shuffleboard This is my attempt to pair code-shuffleboard with code-golf. You need to write a program that sums all codepoints of the input string. Score Your base score is the sum of the codepoints of your program. Your base score has to be e...

2

A: Sandbox for Proposed Challenges

Symmetrical difference code-golf Surprised that we don't have a symmetrical difference challenge yet. Given two lists only containing positive integers, return all items among these two lists that aren't contained in both of the lists. Rules From the set definition: A set is a collection ...

I have a lot of ideas recently.

I'd be glad to see any feedback to these.

Ad Hoc Garf Hunter

You just deleted one?

How can to be giving feedback on a deleted post?

petStorm

One: the propasal with a one upvote. It's the my #1 unclear Sandbox post.

(I don't know how to clarify it.)

Also it encourages trivial solutions anyway.

Ad Hoc Garf Hunter

Yeah main=print"main=pri" would be a very easy haskell solution although I am not sure about the "encoder decoder part".

petStorm