« first day (2650 days earlier)      last day (2209 days later) » 
00:00 - 21:0021:00 - 00:00

ngn
9:08 PM
 
@ngn Huh, why didn't I think of that. Also, remove the .
 
ngn
@Adám because you are not ngn :)
@Adám why didn't I think of removing the ⍞ ...
 
@ngn ⍨ I mean, I thought of comparing the letters at each index, so why not just the length.
@ngn Because you are not Adám ;-)
 
ngn
@Adám that's what I did first too, but ⎕ucs is simply too long
 
@ngn 3-'bd'⍸4∘⊃
@ngn Wait, 'ec'⍳4∘⊃ is enough. Tied!
 
ngn
9:18 PM
@Adám 1+3|1+≢
 
@ngn Ouch.
 
@Adám or...he had the solution stashed until now to make you feel like that ;)
(hey, that's not friendly competition though)
 
@EriktheOutgolfer I know ngn uses tactics like that. He wrote about it after the codegolf.co.uk tournament.
 
Is anyone here an HTML/CSS wizard? I need a bit of help.
How do I blur out a <div> with text in it?
 
@Pavel Blur the text? You can make the text transparent and add a not-offset shadow.
 
9:25 PM
That might work
Actually, if I think about it, there can be all kinds of stuff like images too.
 
@Pavel #mydiv {filter: blur(5px);}
 
Oh, ty
 
@Pavel Only doesn't work on IE.
 
ngn
@EriktheOutgolfer what is there for me to gain from that... ?
 
@ngn doing that is a 100% no-no; this site is supposed to host friendly competition, and using tactics like that means, in the long run, making other members quit the site, so it would not be friendly anymore
 
9:29 PM
@Adám Good thing I'm building a chrome extension :P
 
ngn
@EriktheOutgolfer but I did not stash my solution
 
@EriktheOutgolfer I don't think ngn does that here. During the tournament, it was a tactic to make people ignore a problem until it was too late.
 
ngn
@EriktheOutgolfer look at my first attempt - it even had an extra ⍞
 
@ngn oh, didn't say you did that here
@Adám I'm not sure how people would have "ignored" a problem; you can't go to the next one until you solve the current
 
ngn
@EriktheOutgolfer in the Optima tournament I used such tactics, but afaik, it was not against the rules
 
9:32 PM
and, since you can't ignore it, you try your best to golf it...right?
 
@EriktheOutgolfer No, but you could think you've reached the optimal solution because your competitors have same byte count, and then not spend time looking for a shorter one.
 
ngn
@Adám sorry, corrected
I also submitted some deliberately wrong submissions that only passed the tests but didn't solve the problem in general
 
@ngn that's why the tests shouldn't be visible; heck, they shouldn't even be hard-coded!
 
ngn
to make others think that I've outgolfed them and waste their time
@EriktheOutgolfer well, scoring must be repeatable
 
@ngn huh? if you mean comparable, then isn't the code size the same for any test case it can be tested on?
 
ngn
9:37 PM
@EriktheOutgolfer I think it's good that tests are visible. The right way to counter dirty tactics like mine is to let anyone suggest a valid test.
 
@ngn but...it didn't let that, did it?
 
ngn
@EriktheOutgolfer yeah, code size is the metric, but the tests tell you whether you passed or not - that should be consistent across test runs
so you can't have randomised tests, for instance
@EriktheOutgolfer no, the system didn't let us suggest new tests
 
@ngn ah, I meant having a confirmedly valid program hidden ultra-securely in the server generate the tests and then compare those to the submission
(maybe not even in APL)
 
ngn
@EriktheOutgolfer ah, I see. Well, I'd still prefer open tests, but it's up to the organisers.
@EriktheOutgolfer there are dirty tricks you can do to learn the input data, one bit at a time :)
 
9:43 PM
For the student competition we don't publicise the tests.
 
@ngn there is a site code-golf.io which uses such a system, and I don't think that you can do that for all holes in there, only for some (and even then, it's shorter just not to hardcode)
 
@Pavel Fun way to make people squint: *:hover {transition: 30s; filter: blur(.5px);}
 
Ah shit I just started up two instances of SRCDS and locked up my VM
@Adám what's the difference between the vegan and vegetable soup?
 
ngn
@quartata $2
 
@quartata I suppose the vegetable soup may have dumplings made with eggs and/or milk.
 
9:50 PM
Ah, makes sense
I was trying to think where eggs would come into play
 
@quartata Fashion.
 
the real answer
 
ngn
iVegan soup - $5
 
iVegan? is that like a soup you can vape
 
I think it's a mock towards Apple...oh, sorry, I misspelt it, should've said App£€ instead
 
ngn
9:52 PM
@quartata ^ :D
 
@EriktheOutgolfer ah, that's Micro$oft's rival right
 
VRegan soup $20
 
Valvegan soup $50
 
but always on -90% sale
 
ngn
@ASCII-only ¥es
 
9:55 PM
@quartata that only applies for Valvegan soup with orange things carrots
 
Blockchain soup $1 $2 $3 $1 $2
 
how far can we take this joke
 
Actually, my tofu & mushroom soup was really good.
 
@Adám $10 $100 $1000 $10000 $100000 $0.01
 
hungry people will riot way before that...
 
10:01 PM
— Waiter, there's a fly in my vegetable soup! — That'll be $1 extra.
 
Does tofu even have a flavor
 
yes?
 
@Pavel I marinated it for an hour or so in tamari with garlic and cayenne, then baked it on a pan with a layer of olive oil, and finally cooked it for a while in the mushroom soup.
 
nice
 
@Pavel I ate tofu a lot growing up. There are two main ways we had it: Scrambled with veggies, cumin and paprika (good, but not my favorite) or broiled to crispiness in soy sauce. (My favorite)
 
10:09 PM
@DJMcMayhem hmm. the scrambled tofu actually looks kinda nice
 
It is. And there's a lot of different ways to do it (I bet some broccoli in there would be delicious)
 
1
Q: Should we stop accepting answers on meta?

dylnanI ask this question because I worry that if an opinion regarding a certain rule holds a majority among users at one point in time, an accepted answer signals that the rule can no longer change to anyone who later considers the question. There is no reason, however, that the rules we put in place ...

 
@DJMcMayhem I'm considering switching from two fried eggs every morning to just adding silken tofu to my huge all-in-one smoothie.
 
What kind of smoothie has fried eggs in it
 
I don't think those are in his smoothie...
 
10:11 PM
I really like the unexpected culinary side of TNB lol
 
ngn
@DJMcMayhem "the 19th bite"
6
 
Earlier today we were talking about all the different culinary uses of Spinach and Cauliflower in #nethack on freenode
 
@DJMcMayhem clearly we've been affected by the abundance of food-related esolang names
 
Cauliflower in Russian is Цветная Капуста, which literally translates as "Colorful Cabbage".
I have no idea why this is.
 
@Pavel My smoothie has 1 dl of raw oats, 1 apple, 1 orange, 1 banana, and ¼ l unsweetened coconut milk. But I eat the two eggs (usually fried or plain omelet, sometimes boiled) on the side. I'd like to just have everything in the smoothie.
 
10:14 PM
@Pavel If anything, it should be "uncolorful broccoli"
 
@Pavel Because white has all wavelengths mixed?
 
I don't think people knew about light wavelengths back when they named it :P
 
checks out to me
 
I know those are all perfectly normal, but to me it looks like someone found wild cabbage growing in a radioactive fallout area
 
ngn
10:21 PM
@Pavel from ru.wikipedia: "в течение длительного времени называлась сирийской капустой. В те времена она ... имела небольшую кремовато-зеленоватую головку."
 
"кремовато-зеленоватая" still isn't more colorful than regular cabbage.
 
ngn
@Pavel true
@Pavel perhaps цветная comes from flower (цветок) rather than colour (цвет)?
 
Woah
That makes so much sense
 
10:46 PM
@Downgoat they had to do it for SO Teams
It was rolled out to people in the Teams beta before this
 
should not a or b be parsed as (not a) or b or not (a or b) ?
 
@ConorO'Brien (not a) or b
because unary operators usually have higher precedence than all binary operators
 
well i guess not should still be lower precedence than most binary ops though
 
Thinking about it from "what would be used more", (not a) or b makes more sense anyways
also if you write it like !a or b...
it would be quite the surprise
 
10:52 PM
! a||b
 
@Pavel +1 for space-sensitive precedence :P
 
@ASCII-only reminds me of I (the first ninja)
 
@ASCII-only I has. ninja'd
 
ninja'd
 
hahahahaha
 
10:55 PM
@ConorO'Brien yes. was talking about this in another chat a few days back
 
@ASCII-only Where?
 

 new language yay

DIscussion for a possible new language
 
@quartata ye, but ! isn't an operator for negation
 
11:08 PM
It could be negation prefix and factorial postfix
 
mixed unary *fixes is kinda gross
 
Is it though?
 
yeah
esp. if ! is also dyadic
and if two adjacent entities mark statement separations
 
But is it also dyadic
 
yes
Print!!3!4
 
11:12 PM
What does that do
 
this illustrates the problems with having an extra postfix !
 
I mean, -- and ++ are both prefix and postfix operators, and no one minds
 
right, I forgot that universal rule :P
 
@Pavel well yeah but those languages have mandatory commas to separate statements
 
λ attache -ae "Print!!3!4"
\- #Token<"!", :operator, 5>
    -- #Token<"Print", :word, 0>
    \- #Token<"!", :operator, 8>
        \- #Token<"!", :unary_operator, 6>
            -- #Token<"3", :number, 7>
        -- #Token<"4", :number, 9>
yeah my parser is a bit insane
what this does is (Print) ! ((! 3) ! 4)
 
11:15 PM
So what does 3 ! 4 do
 
well it's unimplemented rn, I've been debating whether or not to ncr, npr, or something else
 
@ConorO'Brien :|
 
@ASCII-only doesn't that make sense
 
@ConorO'Brien ew ugly tree
 
@ASCII-only it works, if you want to overhaul be my guest and submit a pr :P
 
11:17 PM
@ConorO'Brien wait where did the third ! go
 
@ASCII-only they're all there
 
@ConorO'Brien ah. they are
@ConorO'Brien :| wrong order then?
 
nope
it's right
 
i would totally have posted an AST if I had shunting yard working
 
@ASCII-only why would it ever be parsed like that lol
 
11:18 PM
oh binary
hmm.
 
maybe I should say that f!x is f[x]
 
0/10, not @
 
f@x is also f[x]
 
@Pavel I had a bad idea for a language where that is actually necessary in all function calls >_>
 
@ConorO'Brien What's the difference?
 
11:22 PM
precedence
 
Ah
 
and that f@g is composition, for f, g functions
 
Why not just use $ instead of ! tho
 
because $ is either $ or $word
 
@ConorO'Brien hmm. is that a good idea
 
11:23 PM
@ASCII-only yeah
the whole "f@n is function calling" idea came a long while after @ meant function composition
 
Anonymous
@ConorO'Brien glares in PHP
 
$ indicating special variables is a good idea though
 
well
it isn't >_>
@Mego $word is an artifact from earlier versions of Attache, it's just an alias for "word"
λ attache -r
A> $glare
"glare"
A> $glare = "glare"
true
A>
it was to be used idiomatically in things like Define[$name, "John Smith"], which was before := was a thing
(this entire language was going to be a meme originally, but I got so attached to the parser I decided to run with the concept)
besides, $word is more readable sometimes, especially when specifying options
Bisect[1:5] == [[1, 2], [4, 5]] and Bisect[1:5, bias->$left] == [[1, 2, 3], [4, 5]]
 
@ConorO'Brien :|
@ConorO'Brien ...
@ConorO'Brien it should be an atom instead
that's what sane languages do
 
@ASCII-only an atom? as in?
@ASCII-only you presume attache is sane
 
11:28 PM
@ConorO'Brien as in lisp atom. basically a symbol in other languages
 
@ASCII-only symbols are disgusting in a language like attache, where the focus is on data manipulation rather than stateful computations
 
@ConorO'Brien well it's saner than 99% of PPCG languages. if that's not sane IDK what is looks at starboard nvm
@ConorO'Brien what's so bad about symbols
 
@ASCII-only well it still is pretty sane, but it doesn't allow the notion of "normalcy" or "sanity" bind it
 
@ConorO'Brien even Mathematica basically uses symbols
@ConorO'Brien of course. same with my theoretical language :P
where function calls aren't a thing - only operators
 
@ASCII-only don't get me wrong, symbols are awesome, just not in Attache. it further confounds the notion of data manipulation IMO. symbols serve little purpose outside of naming conventions and "unique variables"
 
11:30 PM
@ConorO'Brien yeah that's the point
 
@ASCII-only even Mathematica has a builtin for detecting the upness of caprine mammals
@ASCII-only ... to confound data manipulation?
 
@ConorO'Brien they're supposed to be used more as a primitive identifier - i.e. they're supposed to be faster than strings
 
@ASCII-only speed isn't a concern in attache lol, it's written in a scripting language
 
@ConorO'Brien PyPy is also written in a scripting language >_>
 
it's also a JIT
 
11:33 PM
Don't 99% of scripting language implementations use JITs anyway nowadays
 
no clue
I'm designing a language with an implementation, not implementing a language with a design
 
ngn
is anybody here on ppcg implementing a language in which speed (as well as brevity) is a concern?
 
I'm sure it's always a concern, just not the primary one
 
@ngn I intend to. Obviously since it's me though it'll probably be abandoned within half a year
 
ngn
@ASCII-only most languages have that fate :)
 
11:39 PM
@ngn isn't that assembly language?
 
ngn
@DestructibleLemon assembly is hard to program in, it's not compatible across cpu architectures, there's overhead for creating a valid binary
 
@ngn so you're talking about a language like c, which has a good optimiser
 
ngn
@DestructibleLemon c is too verbose
 
D is nice
 
@DestructibleLemon Nobody on PPCG has made a fast assembly language. probably. for all we know minxomat could have done it >_>
 
11:44 PM
@ngn so you want c but with shorter syntactic elements
 
ngn
I'm trying to implement something similar to k, so I was curious if there are other super-fast minimalist languages I should look at
 
@ngn APL or J :P
 
@ASCII-only what i'm saying is that every language has to be run by something, and at the lowest level it is always assembly/machine code
 
Anonymous
@ASCII-only umm
 
ngn
@ASCII-only yeah, they are all the APL family, including k which is a cross-breed of APL and Lisp
 
11:45 PM
@ngn actually, how many aren't based off APL :P
 
ngn
@ASCII-only Forth comes to mind, it's minimalist, I'm not really sure how fast it is
 
@ngn IIRC it's faster than C in certain situations
 
ngn
we need more challenges to create evolutionary pressure :)
 
most user-created languages will not exceed C or assembly or similar
 
We don't have fastest-code factorial
Someone should write that
I want to submit Mathematica, extremely f*cking fast: Factorial
 
ngn
11:54 PM
@Pavel the factorial function grows too quickly, the fastest would be just table lookup
 
@ngn You can just ban lookups, since that's optimizing for the test cases
 
@ngn yeah but what language has infinite table lookups
 
Mathematica computes the factorial of 10 million in like 20 seconds
That's pretty good
 
ngn
oh, I was thinking just 64-bit ints...
 
I'll write an arbitrary-precision factorial fastest-code later today
 
11:59 PM
@Pavel completely? or does it approximate?
 
00:00 - 21:0021:00 - 00:00

« first day (2650 days earlier)      last day (2209 days later) »