The Nineteenth Byte

:O react is magic

it automatically put css file in HTML :O

Who needs frameworks when you can write your own?

The .jsd framework isn't very popular yet.

because jsd don't suport pug & sass

Yet

I have no idea what sass is.

6:15 AM

@HyperNeutrino in general you shouldn't flag a moderator for something that's going to happen anyway

user165474

@ais523 Okay. Makes sense, thanks.

I now need sass in my life.

although you should be aware of where flags go; if a flag doesn't go to a mod queue, then feel free to use it to speed up community purposes

(e.g. close flags and VLQ flags are handled by the community)

user165474

Ah, okay. But typically mod flags shouldn't be used for things that aren't urgently in need of mods?

it's OK to use them for anything that will inevitably need a mod, whether urgent or not

user165474

6:17 AM

Okay.

for example, I flagged a mod for an incorrect red tag on an old question on Meta

that clearly isn't urgent, but on the other hand, it clearly needs a mod to fix and fixing it's still to the net benefit of the site, thus it's an appropriate flag (and the mods agreed)

user165474

Okay. That makes sense then, thanks.

I refused to use Express so much that i wrote my own cookie, then session handler so that I could use OpenID.

I don't reinvent the wheel, I rebuild the wagon.

And it's a terrible wagon, it falls apart in like, 3 days.

@ATaco OOCQOD

user165474

I don't reinvent the wheel, I build a wagon that doesn't even need wheels

6:24 AM

That's a sledge.

#tobogganmasterrace

user165474

sled vs sledge vs toboggan

@MartinEnder: did you just ninja-delete a comment of mine, when the comment it was replying to was deleted? not that I'm opposed to that, but it was so fast it left me a little disoriented :-)

or did I somehow miss the comment submit button?

@ais523 sorry, I didn't know how long ago you had posted the comment, I just checked my phone and saw the notification, and your comment made mine useless so I deleted both of them

you managed to delete it something like 2 seconds after it was posted :-)

I know why it's done, it's fairly common to delete answered questions and hope that the reply gets deleted too, and mods have a shortcut around that

(I've unfortunately seen the opposite: I've deleted an obsolete comment and had someone respond with a comment of their own listing the original context to their reply…)

6:33 AM

I don't really see the purpose of deleting comments in the first place.

what's the purpose of leaving useless comments around?

Stack Exchange rules say that all comments should be intended to suggest improvements to the post and eventually deleted; these rules are widely, and rightfully, ignored

that doesn't mean that all comments make sense to leave around permanently, though

user165474

> crossed out 44 is still 44

user165474

lol improvements

The SE philosophy seems to be "keep things around so that other people can see it if they have the same question".

And oftentimes people are just asking for clarification on very small things,

6:34 AM

no, the SE philosophy is that comments are transient and anything that's important to keep around should be edited into the relevant question or answer

actually I think SE was invented as a collaborative FAQ-building project, the whole "get an answer to your question" thing was intended to encourage people to contribute questions

on SO, that didn't really work out

it might work on some of the smaller sites

Dennis

@Challenger5 That doesn't apply to comments. Thousands of comments get deleted every month.

on PPCG, we aren't even trying that, except possibly with tips

The issue is on PPCG people ask for clarification on things that really don't need to be in the post, like "Can I join most lines with \n except with a trailing \r\n?"

Well JSD now supports sass.

user165474

6:36 AM

@Dennis how many of those are self-deleted vs mod-deleted?

@Challenger5 ideally that'd be something that was covered in a Meta question; the only thing more ridiculous than having tons of those corner-case questions is making the OP of every question decide all that stuff for themselves individually

Dennis

No clue.

(note: "only thing" there is hyperbole, more ridiculous things probably do exist)

I know a decent proportion of my flags are comment flags

user165474

ah ok

mostly because obsolete and thus now misleading comments, unlike obsolete everything else, actually have a fairly simple resolution

6:37 AM

@Challenger5 if this is not a standard policy then it should go into the post. People shouldn't have to read the comment chain to find all the rules about a challenge.

also because some people insist on going offtopic in comments

now I'm tempted to make a meta post "is it legal to output an array of strings using inconsistent newlines?"

but I'm not sure the problem comes up often enough to need a ruling

I don't want to have to ask a question on Meta every time I have one of those concerns and it hasn't been handled before.

another thing that annoys me: there's no way to see an equivalent of the reputation history for Meta; just because reputation doesn't exist there doesn't mean that knowing when you've been upvoted and downvoted isn't useful

I know the system's tracking it because I got Mortarboard on Meta a while back

@Challenger5 that's why you ask the challenge author in a comment. but the clarification is just as relevant to anyone else solving the challenge, so the answer should go into the spec if it isn't already covered.

but not even /reputation will show you the events-that-would-generate-reputation on sitewide metas

user165474

6:40 AM

@ais523 I've found that quite annoying too. I think upvotes and downvotes should be logged into your profile (though they'd still be anonymous).

Mendeleev

@ais523 You make a lot of esolangs.

after a while you get good at it

it's reached the point where I can define single-purpose esolangs for single PPCG posts

You seem to like Minsky machines a lot.

quite a few of my recent esolangs are minsky-based; it's a good place to play around with the borders of Turing-completeness

basically because to have a working Minsky machine, one of the few things you need is the ability to count

and most languages are capable of counting, even if they can't really do anything else

I guess I like languages which don't even natively have numbers at all, though

like Underload or But Is It Art?

I really need to write a BIIA? interpreter at some point; there's a pre-existing interpreter written by someone else but I'm not entirely convinced it actually works

Mendeleev

CMC: Interpret Brainfuck in Tomato

> But Is It Art?

That language is very interesting.

6:46 AM

I can find at least 12 esolangs listed on your user page that mention Minsky machines.

oh wow, I just stumbled across a duplicate tag: tiling and packing

I bet I don't have the rep to start a tag duplicate vote, though; they're both pretty rare tags

At least 3 more reference The Amnesiac from Minsk, which at the first level is essentially a Minsky machine.

actually, I think the TAFM references are explicitly minsky-based, whereas most of the others just used minsky machines for the TCness proof as it was easiest

@ais523 You could ask @Dennis

And then I'm going to assume Burn and Feather are Minsky-based until proven otherwise

6:48 AM

if you're screwing around on the boundaries of TCness, you're normally using minsky machines, tag, or cyclic tag

user165474

lol I was going to suggest synonym but I need 5 score in the tag >.<

Feather's functional, it's not based on Minsky machines at all

user165474

Oh wait, I have enough in packing

@HyperNeutrino there's a tag synonym I've been meaning to suggest for ages (after getting a Meta consensus) but don't have the rep

user165474

okay I started a synonym vote apparently?

6:49 AM

the great thing about synonym votes is, nobody's told they exist unless you tell them yourself

so some of them have been going for years

user165474

oh I see

here's the vote, for anyone interested

@ais523 oh wow, I've been meaning to design a minimalist 2D declarative esolang, but I couldn't really make any progress on it. BIIA looks really interesting

fun fact: there's a tag synonym vote ongoing that was started by @MartinEnder, which must therefore have predated the current set of mods (otherwise it'd have been modhammered)

oh, which one is that?

6:51 AM

I've been considering writing a wimpmode BIIA too in order to make it golfier

You know all the good composers didn't come up with names for their songs; they just number them.

You should number your esolangs. You already have 7 and 90.

user165474

lol I answered a singular question in packing and it has enough score to let me dupevote

@MartinEnder optimized-output → optimization

user165474

there should be a review queue

I'm seriously unsure whether I agree with it or not (certainly the two have been used fairly interchangeably…)

6:52 AM

@ais523 I honestly hate both of those tags

I think they could be considered "too broad"

"optimization" seems very vague. Optimizing what? Source code size? Practically every challenge on PPCG is optimization of some sort.

you could say the same about code-golf, though

come to think of it, metagolf is also in the optimization family, but at least it has a much more precisely defined meaning

user165474

code-golf seems quite clear to me...

it's clear, it's just broad

user165474

6:54 AM

oh true

There was a meta discussion about the use of code-golf.

@ais523 right, but [optimization] is not a winning criterion

and also makes a complete mockery of dupehammers

I think the consensus was "as long as it doesn't cover every question on this site, it's fine"

interestingly, that's one of a few ways for a question to become untagged: it's originally given a tag X, and then migrated to a site where every question is about X

normally, if you try to make an untagged post, the site won't let you

6:56 AM

So does SE figure out which tags to delete?

there's some way of marking a tag as a "pervasive" tag, i.e. applying to every post on the site

I'm not sure of the details

That seems like mod magic

I guess a good example that applies to PPCG: Puzzling has a tag open-ended, which basically means "there isn't a single correct answer in mind"

that applies to almost every post on this site (with the exception of programming-puzzle)

There are some things that would be unreasonably hard to beat

Like any challenge that has a 1 byte solution in a golfing language

you can still look for 0 byte solutions :-)

if you find multiple 1 byte solutions, in a range of languages, which work via very different algorithms/mechanisms

then a 0 byte solution frequently does exist

incidentally, I'm still upset that PPCG won't accept answers in the 1-7 bit range

7:00 AM

I guess that's bad for 7

I don't think it was agreed what would happen if someone stored the program using a filesystem that could store a file in fractionally many bits

I've already started work on it, but other things are a higher priority

Fatalize

@ais523 Really? Why wouldn't 0.75 byte be accepted?

I don't think any filesystem can handle fractional bits, but I understand your point.

Destructible Lemon

well no more answers on my challenge yet ;_;

@Challenger5 it's possible: you can store two 2½-bit programs and a 3-bit program in a single byte, for example

7:02 AM

@DestructibleLemon m8 can you drop it already

Destructible Lemon

no

;_; it's too sad

No one will answer your challenge because you complain about it's lack of answers constantly.

however this does open up something of a loophole: imagine a language like Reciprocal Lenguage, where you take the reciprocal of the length of the program and interpret it as a program…

user165474

Fractional bits don't even make sense...???

I have a question that only has one answer.

user165474

7:03 AM

When quantum computers are created the bytecount system will be screwed over

user165474

I have a question with no answers

that'd always score < 1 byte, and be shorter the more complex the program

And that's fine.

Leaky Nun

I think someone here confused bits with bytes.

@ais523 I like what you mentioned about defining some encoding for each language that encompassed every possible program, flag, etc.

Leaky Nun

7:03 AM

0.75 bytes makes sense, but 0.75 bit is just madness.

@HyperNeutrino here's a simple way to think about it: suppose you have a language with only three programs; you can store 5 of those programs in a single byte

user165474

Hm.

thus, the number of bits taken up by the program is somewhere between 1 and 2

user165474

Okay... that kinda makes sense

@HyperNeutrino I think that quantum computers are disallowed per this meta consensus

user165474

7:05 AM

aw :(

user165474

I mean, after they become widespread

another good reason to disallow quantum computers is this one

user165474

but then we first have to fix RSA

user165474

(i mean replace)

AFAIU, quantum computers have a chance of providing any possible answer to any possible question, and a program manipulates the probability distribution of the answer to give the highest possible chance to the correct answer

then you check it on a regular computer, and try again if it was wrong

user165474

7:07 AM

hm.

With quantum computers, you can't stop your messages from being read, but you will know if was

user165474

That doesn't sound like it would ever be practical enough then

It means that you can generate a set of one-time-pad keys (which are impossible to decrypt, even with a quantum computer)

You don't need quantum computers for quantum cryptography.

(interestingly, the most probable wrong answers tend to be either close in value to the correct answers, or else somehow resemble a correct answer; for example, with factorization, if you try to factorize 25 you might get 2 or 3 as the answer because 2½ is a factor of 25)

7:08 AM

and then just keep sending new ones until you know nobody read your message

user165474

@Challenger5 so you can sit there helplessly looking as your messages are read and you can't do anything :D

at school, for A level (qualification taken when you're 18), I implemented a quantum computer simulator and ran Shor's algorithm on it

Wait, how does that work?

the simulators are necessarily incredibly inefficient compared to the real thing, so they don't give any gain over just writing the algorithm conventially

Don't you need an infinite amount of memory to store a qubit?

7:09 AM

but it's interesting to see the result in action

"Only" an exponential amount.

and the value of the set of all qubits in a program is basically a probability distribution, except the probabilities are complex numbers

quantumplayground.net/#/playground/5191954035900416

this is a sort of object that classical computers can represent, although doing so is crazily inefficient because you have to associate a complex number with each combination of bit values

user165474

I mean, in theory, classical computers also have a chance of giving a wrong answer, just a very small chance, if I understand correctly?

7:10 AM

so you're basically storing a list of complex numbers of length 2**(number of qubits)

I guess this is why I'm not a quantum physicist...

@HyperNeutrino A Turing machine doesn't

@HyperNeutrino not idealized ones; idealized quantum computers would though

because I think the only way to get the probability of the correct answer as high as 1 would be to use only classical operations

at which point the quantumness of your computer doesn't really help

hi all

user165474

Okay

It worries me that as technology gets more advanced, the barrier to entry gets higher and higher

In terms of understanding the architectures

7:12 AM

Whaddya mean, surely quantum computer is much simpler than x86 or whatever.

Maybe "architecture" wasn't the correct word there

Although understanding how to program it could be tougher.

Then again, understanding the architectures becomes less important.

there are universities which have teams of quantum computer programmers, even though they can't actually test any of the programs

I guess that's why we have compilers, yeah

7:13 AM

rather, they prove they work

the idea is to get a head start when someone figures out how to build a usable quantum computer

Not to imply that it isn't valuable knowledge, but nowadays it's possible to do actual work without knowing how a processor works.

luckily there will never be a working quantum computer that does anything useful :)

That's not very positive thinking.

So instead of a zero-day attack on RSA, we get a negative-day attack on RSA!

That's great.

Quantum computing will greatly improve our basic arithmetic!

7:15 AM

quantum computing defies the typical logic of arithmetic, though

@ATaco Well.. to be more accurate it is not clear either way if there will ever be a quantum computer that does anything useful

Basic arithmetic is important, given how everything we do is derived from it.

I think Q.C.s will only be used for resource-expensive number crunching, not for day-to-day use.

e.g. if you have two mutually exclusive events A and B, the probability of (A or B) being true can be lower than the probability of either A or B

@Challenger5 the problem is that they don't know how to do even that yet

7:16 AM

also, quantum computers would probably eventually end up as coprocessors

like graphics cards are today

why do people here think they will ever work?

Because they do. They have been built.

Because they do work.

@Challenger5 negative

ninja'd

Q: Is there a decent "largest number" question on the site?

7:17 AM

@Challenger5 there has never been a working quantum computer built that can compute anything faster than my phone

New Meta Posts

0

The question largest number printable is broken in many ways: it has a huge number of source restrictions and a strange scoring function. Many of the other largest number challenges are like this in one way or another. I was wondering if we should create a canonical question that fits the moder...

discussion

the largest true quantum computer I'm aware of had 7 qubits, and was a single atom in size; there were lots of copies of it made :-P

@Challenger5 There was a time people thought computers in general would only be used for that task.

@ais523 but even those are not what they seem

IIRC they were used to factorize 15

7:18 AM

They were wrong.

@ais523 Nope, they got 21

QCs require fault tolerance. 7 non-fault tolerant qbits is useless

And then they switched to a different algo and got 56153

and the factorization experiments are close to being fakes

@Challenger5 neat; I imagine you could do both

@Lembik it was very fault-tolerant, given that there were a huge number of computers involved; just take the consensus results

7:19 AM

Apparently they were trying to factor 143 and the algorithm somehow also factored larger numbers ?

read dabacon.org/pontiff/?p=6913 for example

it's all not quite what it seems :)

Looking at the people in the room rn, before I spoke, the first 4 people all don't have icons.

who knows, maybe one day they will make something that works

but they haven't yet

Wikipedia: "In 2012, the factorization of 15 was performed with solid-state qubits. Also in 2012, the factorization of 21 was achieved, setting the record for the largest number factored with Shor's algorithm. In April 2012, the factorization of 143 was achieved, although this used adiabatic quantum computation rather than Shor's algorithm. In November 2014, it was discovered that this 2012 adiabatic quantum computation had also factored larger numbers, the largest being 56153."

Y'all need icons.

7:20 AM

@Challenger5 read dabacon.org/pontiff/?p=6913

@Lembik that's clearly a parody

@Phoenix I think a memorable username is more important

it'd be more interesting to see the concrete criticisms against the algorithms being used

@ais523 exactly.. the point is that the factorization examples people are talking about are meaninglesss

the parody suggests the nature of the criticism but doesn't state it

7:21 AM

Somehow I'm able to remember ais's username despite half of it being a random number

@ais523 arxiv.org/abs/1301.7007

@Challenger5 It really isn't. IDK about you, but I identify most people by icon.

Then I shall be anonymous to you forever! Muahahahaha

I've got used to this icon by now

perhaps we need some sort of script that rerandomizes user icons at regular intervals to make them more anonymous

I get the feeling the S.E. people would not indulge us on that.

7:25 AM

@ais523 all I am saying is that there us much nonsense hype around QC. There is currently no working QC on any scale

maybe there will be one day.. but we have no idea

There is currently no working Turing Machine on any scale either.
Just because it can't be done in real life doesn't mean sufficiently accurate models can't be useful

Also we need a script that randomly replaces characters in usernames with Unicode look-alikes.

user165474

@feersum o plz no

We need a script that randomly generates and launches scripts.

user165474

@Challenger5 for(;;);

7:27 AM

@Challenger5 I disagree with the "any scale"; Turing machines with very large tapes can easily be simulated by modern computers

it's just the fact that the tape isn't infinite that's a problem

I was referring to a Turing Machine with an infinite tape

Q: Optimal game strategy - select one coin from two piles of coins.(coin can be selected from either end)

(that said, you can emulate an infinite tape for the forseeable future by using compression and automatically buying data on, say, AWS; data storage improvements are currently going faster than a Turing machine's ability to use that data, and it's possible we'll find a way to break the theoretical limit for information storage in the Universe before running a program actually hits it)

New Main Posts

0

Given that there are two piles of coins(coins have different values given). Game is turn based. In each turn player chooses one coin from one of the two piles, from either end. Thus there are four possible choices to pick a coin. Player who has maximum amount at the end wins the game. If only on...

decision-problem game

@Challenger5 youtube.com/watch?v=cYw2ewoO6c4

@ais523 No way to fit more than one bit per cubic plank length, given sizes below that make no physical sense. Pretty hard limit.

7:31 AM

I can imagine a universe where half of it is taken up by a computer so that someone can watch cat videos

@Phoenix imagine if we discover that, say, there are multiple independent extra Universes and we have some way to communicate with them

that happens in science fiction a lot; we currently have no reason to believe it to be true, but it's possible that that will change

@Challenger5 Because it's only three digits long?

@Challenger5 Apparently people are more into cat gifs these days.

nowdays gifs are videos, though

@ASCII-only Yep.

7:33 AM

I've got the 523 part down but I always remember aes523

.gifv is basically the same sort of thing as .mp4, the only difference is the typical subject matter (short, low-quality, and looped)

And no audio

@Phoenix >_> ais isn't an encryption algorithm, they're a person

I didn't understand that reference.

AES is a fairly well-known symmetric encryption algorithm

possibly the best-known, actually

(asymmetric encryption algorithms are probably more famous among programmers, though)

7:36 AM

IDK security

(that's cos they're generally better for the purposes most programmers need encryption, such as sending data securely)

(You can tell because I once basically leaked my GH password in TNB)
(And then deleted my account because I forgot changing your password was a thing)

that's adorable

Change password? Come on, everyone knows nuking it from orbit is the only valid choice in any security dilemma.

We were talking about passwords, I mentioned the pattern mine followed, and then we did some math, and we realized that pattern matched as many possible passwords as saying any two letters.

>_<

7:39 AM

My SE password is a long, random alphanumeric string

It used to be a long, random hex string, but I forgot it

I don't have one, it's bound to my google account.

this browser has codegolf.stackexchange.com and its login page as the only pages in its history, thus it's suggested as the URL whenever I even just put the cursor in the location bar

I deleted everything else and then only use the browser in private browsing mode in order to prevent anything else being added

So you're not using Chrome then

Because on Chrome if you clear your history the Settings page is always there

(it is, however, not my primary browser; I'm unwilling to let SE run its javascript on that, because I don't really trust SE and don't have javascript there enabled by default)

That seems needlessly complex to save typing the letter c

7:42 AM

nah, it is Chrome; it has chrome-urls, settings, and version permanently in the box

but near the bottom

I also changed the search engine to example.com to stop it sending things typoed in the URL bar off to sites I don't trust (which is most of them)

Okx

CMC: Given two integers as input, n and k, determine the nth number of the polygonal numbers of size-m. Can be 0 or 1 indexed.

I don't even trust example.com :P

@ais523 Given its userbase, I think people would know if SE did anything bad.

It wouldn't be SE's fault if it were vulnerable to XSS

@Challenger5 I put the search query in an anchor parameter: example.com#!query

7:44 AM

Actually it would, but it wouldn't be malicious

My personal approach to security is "I have nothing worth hacking/stealing"

@Phoenix many people hate SE's referral links, and I dislike the fact that it creates a (temporary) account for you even on just visiting the site

So why bother

and it's probably tracking you everywhere with analytics

You could write a userscript that deleted ga.js

7:45 AM

@Phoenix because then you're making life harder for the people who do have things worth stealing, as it makes them easier for criminals to identify

@ais523 Do you do this for every website you use?

Wait.
How do you run chat without javascript.

It's not possible.

the whole reason I use a separate browser for SE is that, unlike most sites, it requires JS enabled to work

the vast majority work fine without it; in fact, many of them work better without it

this browser has JS enabled, but because of that, I don't use it for anything else and wipe cookies, etc., regularly

(thus I have to log in to SE every time)

That actually seems smart
But I'd just have one browser for anything that used JS, not just SE.

A: Sandbox for Proposed Challenges

How did you get this paranoid >_>

New Sandboxed Posts

7:48 AM

0

Write an interpreter for XKCD's broken language code-golfinterpretercode-challenge Meta Is this a good idea? Is the point system good? If so, are the point values good? I based them off of how difficult I imaged each part to be. Overview Randall Munroe, auther of the webcomic XKCD, po...

same here; it's just SE that I use most often here

My paranoia stops at not saving my logins in my browser :P

I trust people not to be assholes. It's worked out pretty well so far.

I've been burned by saving logins in chrome/firefox before, plus after doing a little digging out of curiosity I realised just how ludicrously easy it is for malware to read your saved logins

I suppose it's possible something does happen at some point, but I don't think the inconvenience it will cause will outweigh the inconvenience of e.g. constantly clearing cookies.

7:51 AM

@ais523 *like most sites

@Phoenix I think it would become routine and you would barely notice it after some time

Eh, not buying it.

not saving passwords/logins helps your memory too

@Mayube Solution: don't get malware :P

I mostly clear cookies to stop sites tracking me

user165474

7:53 AM

Guys, there's an easy solution to all of these problems.

I don't see how it benefits me for the site to know what pages I've been looking at

@Phoenix Same (I'm also trusting Windows Defender to be better than before :P)

@Mayube Doesn't improve my memory at all

Windows Defender is actually much worse than it used to be

MSE > WD

Really.

7:53 AM

back when it wasn't widely used, it wasn't a high-value target for malware to attack / work around

user165474

@Mayube What does this have to do with meta-stackexchange?

user165474

:P

now that it's installed by default and always on, it's the #1 target for malware to break through, as it wouldn't work otherwise

Eh, it removed the software my school uses to spy on us, good enough for me.

thus we can expect it to, over time, become the least effective antimalware program on Windows

7:55 AM

These are the kind of things that make me wish everybody on the earth just got along

It's so much better on my planet

3

a good analogy is lysozyme, the safest known antibiotic

it's produced continuously by humans and is present in large amounts everywhere a bacterium might get into the body

as such, non-lysozyme-resistant bacteria don't affect humans at all, meaning that it's pointless to use as an actual medicine

Still, wouldn't want to get rid of it.

I guess you just have to avoid one competitor dominating all the rest

Otherwise you have a single point of failure

right, the advantage of things like Windows Defender and lysozyme is not in the fact that they're ineffective compared to the alternatives, but that they reduce the prevalence of infections in general

Should I have antimalware on my linux machine? xkcd says no, but IDK.