« first day (108 days earlier)      last day (94 days later) » 

7:32 AM
Got a sec, @JonathanAllan? (I'm busy getting errors too and want to get back to it)
curious to know if my comment about nested macros answered your question
and if you know about \let as well
... @humn from here if/when you'd like ...
I seeuu manshuu, here's what i've met so far:
was hoping for a more fractal-y effect but there are also space limits, haven't even decided on the colors
 
 
1 hour later…
8:49 AM
@humn yes and no... the # of #s thing (excuse the overload :p) means I can create a def in a def, but does not help with recursion much, because I'd need to write that explicitly for the length of "MedusaXYZABC..."
I mean the line can be something like
\def \| #1#2#3|{\def \| ##1##2##3|{##2#2#1##3##1} \| #1#3#1} \|
but it would have to go deeper
 
do you know about \let?
 
no I know nothing
:p
hence the comment about just typing away :)
 
most people don't even know that they know nothing
 
and the reference is a bit self-referential
seems to expect you to know all sorts
 
hehehe
 
8:51 AM
nice
 
\let \A \B reassigns the definition of \B to \A
then you can \def \A to be built on \B ... at least in regular (La)TeX
 
oh I was thinking to keep redefining \|#1#2#3|{...}
 
but i haven't gotten it to work here
interesting|!|
 
try the snippet ^
 
(hang on.. gotta set up a lab bench for this)
 
8:53 AM
should give eM|dusa|
but as I say then we need to recurse, and MathJax does not like it (and we cant def \|#############1....)
 
your snippet works for me!
right, getting the recursion to bottom out is where i got stuck and decided to just post the thing anyway
 
also we'd get down to ||, at which point it would complain too
(I think)
 
yeah, i almost feel bad about allowing for that possibility, but you should've seen the requirements before i simplified them!
 
I saw the first puzzle (the quine) and I just "ugh dunno"
I don't even understand the answers :p
 
hey, this is one way to learn... did you notice that the person who solved the quine so sweetly is the designer of MathJax?!
 
8:57 AM
yeah I saw your comment
 
i almost feel like we're getting free consulting now
maybe we can make a feature request
did you understand my mechanical solution to the quine?
 
I wrote an answer for...
26
Q: Latin square compression

Nathan MerrillA Latin square is a square that has no repeated symbols in the rows or columns:. 13420 21304 32041 04213 40132 And as many Sudoku players know, you don't need all of the numbers to deduce the remaining numbers. Your challenge is to compress a Latin square into as few bytes as possible. You ...

but then saw exactly the same had already been done :(
 
whoa! all those solutions in just 2 days?!
 
basically Lemar code / Factorial base for the permutations
 
not familiar with the term (maybe not concept either)
 
9:00 AM
um..
 
thought i did see you touting primordials recently, though
or was it prinary? (or the same thing?)
 
sorry, Lehmer
 
you can call me Shirley if you'd like
 
here...
In mathematics, the notion of permutation relates to the act of arranging all the members of a set into some sequence or order, or if the set is already ordered, rearranging (reordering) its elements, a process called permuting. These differ from combinations, which are selections of some members of a set where order is disregarded. For example, written as tuples, there are six permutations of the set {1,2,3}, namely: (1,2,3), (1,3,2), (2,1,3), (2,3,1), (3,1,2), and (3,2,1). These are all the possible orderings of this three element set. As another example, an anagram of a word, all of whose letters...
 
think that Latin square puzzle might've hooked some Code Golfers. recognize one of them at least
 
9:02 AM
that section describes it
and this is the same as my code, effectively,
7
A: Latin square compression

DennisPython 3, 10772 bits (1346.5 bytes) def compress(rows): columns = list(zip(*rows)) size = len(rows) symbols = range(size) output = size - 1 weight = 25 for i in symbols: for j in symbols: choices = set(rows[i][j:]) & set(columns[j][i:]) o...

 
they're the one i recognize... half of the answers on Code Golf ask for help from this one
 
although maybe could get it down to 10672 bits by exploiting that the requirement is only up to 25*25, not sure though (decompression needs to figure out size and compression need to avoid overlap)
 
looks like they're not used to explaining their answers
 
overlap = hash collision basically
 
you hashing? or talking about compile-optimization-like hashing
 
9:05 AM
well effectively
you are creating a unique number from the Latin Square
 
didn't you already solve this problem, and then some, with your sudoku program?
 
no
just like Dennis
 
gotta admit i only skimmed the description of the puzzle
 
each number has some possibilities and so you encode which one it is in a sorted order as your next "digit"
 
ah... now i see the magic word... compression time
 
9:07 AM
mine was prob a touch faster
just cos I used bisect
 
why not post it? too much trouble because you'd feel obliged to actually include an explanation?
 
but will only be noticable when the squares get big :p
 
size matters!
 
no point unless I can beat the compression achieved :p
 
i guess
 
9:09 AM
and I equal it AND use the same method
so I may as well copy and post :p
change var names :p
 
ah, you'd know if there's really nothing new to tell
 
it's very similar :p
oh well
so how does \let help do you think?
 
the bisect difference sounds interesting in words...
 
bisect is a module in python
 
there's probably already a module that compresses a latin square
 
9:11 AM
if you have a sorted list you can use bisect to find out where a new number should go quicker
or, equivalently, find where an item is in the list faster than list.index(item)
 
guess the list isn't a tree
 
yeah it's not. You could implement a binary tree too, but no need
if you have a sorted list already
 
except for runtime efficiency with random insert... nothing new to say i guess
 
yeah not doing any
imagine you have a 25*25 Latin square
 
go on
 
9:13 AM
you look at top left number and it could be [0,1,2,3,4,5,6...24]
you see it is 5
so you can bisect the list to find 5 is at index 5
and then pop it out
next number is 24, say
so it's at index 23 now
etc
never inserting
just finding and poping
 
always indexing... i guess finding 24 involved minimal search
 
your list remains sorted
maybe could do something clever cos the list is actually a sorted & of two sets
when you get to second row first column, it cant be a 5
yeah Dennis just uses .index
 
right
 
25 is pretty small
 
does it boil down to permuting rows and columns independently?
 
9:17 AM
um
 
... about \let, though, it allows \A to be defined in terms of \A where each version of \A is actually different
... that's in TeX, but when i gave it a quick go in MathJax i got the same infinite recursion error you did
 
if you had a single permutation of 25 objects (like a row is) then you can represent it as a number in ! base. As you move down the rows the numbers get smaller because you can ignore the numbers already found in the columns
 
triangle-ary
 
yeah I'm not sure how to make it recurse down and stop :p
@BmyGuest - nice maze :)
 
the word maze!
i was going to ask why it's not
 
9:21 AM
ideas for maze problems circulating my brain...
 
.. here's a piece of mathjax cheese for your brain, but then i wanna get back to a metamathjax picture...
the idea, not worked out, for terminating recursion is to wrap another function and delimiter in there...
.. overlap them.. so at one point it looks like
\A .... \delA \B....\delB
and then later
\A ... \B ... \delA ... \delB
something like that, so \B won't get called when it is in \A ... \B .. \delA
 
ok, but there does not seem to be a way to define N levels dependant upon the input
 
right, i can't find anything resembling a counter, for instance
 
we need to have it work with ||, |M| |Me| ...
|blah|
 
still not sure it's possible, but i want us to have a chance before that same person sneezes and solves it
 
9:25 AM
:p
 
.. thanks for giving it a whirl anyway!
... before (again) i go ... didn you not volunteer for moderator for a reason? i didn't because i'm too flakey
(i secretly want someone who might like the kind of music you do,.. says more than will be written here)
 
reason is I don't know if I can commit time to it, or if I'd be any good!
think Deu, Alc, Emr possibly a good 3
 
did A enter the ring?
 
eventually
4hrs ago
 
ah, all 3 have thoroughly solid Puzzling credentials! i was hoping for someone i knew a little bit
(which is about 4 people here, one of whom disappeared about a month ago)
wish we could have way more total, really everyone who wants to would be fine with me
 
9:33 AM
who disappeared?
 
q_a
 
oh!
weird
 
@JonathanAllan Thanks. Had a lot of fun watching the comunity take it apart in very short time.
 
prob be back
 
hoping q_a returns too! ...... the maze is already solved?! no fair, i wrote and deleted a comment saying how i hope it takes many times longer to solve than to construct
... ohhh, i'm heading back to editing ... pleasant permuting!
... uh, duh, just noticed that the latin squares puzzle IS at Code Golf.. no wonder some of the people looked like they belong ther .... bye again
 
 
1 hour later…
11:00 AM
yeah, i'm hoping q_a returns soon too - he was a cantankerous asshole, but he was our cantankerous asshole
5
as for the election, my top choices would be Emrakul, GPR, and Alconja
manshu clearly cares a lot about the site too - i'm a bit worried about the language barrier though
 
11:26 AM
@Deusovi we think alike, your choice of GPR would be my fourth :)
 
Sid
11:36 AM
What language barrier? I suppose everyone here understands English and can write in English
 
@Sid: English isn't Manshu's native language. It doesn't affect too much, but there are some occasional misunderstandings
 
11:57 AM
@Deusovi Yes, I thought about putting it in my reasons-list. :p
 
Sid
It isn't mine either... but I don't feel it would affect much.. @manshu u write just fine..
 
Thanks @Sid :)
 
 
2 hours later…
2:34 PM
hello, may i know if anyone of you has tried pokemon go?
 
f''
I came up with a terrible cryptic clue
force around chip is in glass (4)
 
3:00 PM
seems like someone is on an Editing spree
http://puzzling.stackexchange.com/review/suggested-edits/22046
:D
 
:D
 
3:58 PM
0
Q: Welcome to Cubing

ABcDexterAfter the critical feedback on the previous question, here is the new question... It so happened that one day, your nephew came to you and said that he wanted you to play a board game with you. You were getting bored so you agreed. The only problem was that you had no dice to play with and ...

 
4:13 PM
@f'' huh, only four letters? seems like it should be more due to clue length/type
not many four-letter clues that aren't double defs or deletions
(also, i'm preparing a puzzle for the maze theme >:D )
 
 
3 hours later…
f''
7:15 PM
@Deusovi yeah, only four letters, and it's not a double-def or deletion
everything is kinda dodgy, I did say that it was a terrible clue
 
@f'' is it mica??
 
f''
@WesleySitu you got it!
 
 
4 hours later…
11:50 PM
ah, clever - i got the IC part, but didn't think of MA, and didn't know the word "mica"
 

« first day (108 days earlier)      last day (94 days later) »